"Mistakes Were Made" Is One Of The Worst Apologies You Can Give

The psychology behind how we convince ourselves we didn't do anything wrong and how it destroys crisis recovery

We’ve all seen statements arrive after a public scandal that use the passive voice and contain phrases like “mistakes were made” and “lapse in judgment”, and we typically all have a similar response to them. When I come across one, I imagine us collectively rolling our eyes at it. These read as non-apologies because the sanitized language and distancing of the person from the problem come across as a lack of accountability and refusal to fully take responsibility.

I could say “don’t do this in your response” and end this article here, but telling people (including ourselves, unfortunately) not to do something, at least in my experience, rarely results in us not doing that thing again. We need to understand the why that leads us to do these things if we want to stop ourselves from doing them, and the why of a bad response isn’t as simple as “I want to look better”.

A blame-shifting, distancing non-apology and insufficient response often comes from the mental gymnastics we do to convince ourselves that we’re not actually the bad person in the situation; it’s less of a deliberate attempt to hide something and manipulate the situation to go more in our favor than it is a result of our convincing ourselves that we’re not wrong in the first place.

Albert Bandura’s main argument in Selective Moral Disengagement in the Exercise of Moral Agency helps us understand why we end up saying spectacularly self-defeating things when we’re caught doing something we shouldn’t: we don’t usually engage in harmful conduct without first justifying the morality of our actions to ourselves. We all have internal moral standards and feel bad when we violate them, so when we do, we disengage our conscience to avoid feeling like a bad person. Unfortunately, not feeling like a bad person doesn’t extend to not being perceived as a bad person, especially when it comes with a terrible crisis response.

One reason this type of thinking is difficult to catch in ourselves is that we often don’t know we’re doing it. You might be thinking, “I don’t sit around trying to justify why I’m not a bad person when I do bad things,” and you might not, but that doesn’t mean your ego isn’t subconsciously doing exactly that without your knowledge. It’s actually quite difficult to acknowledge that our brains do shit with our thinking without our knowledge. Probably because it’s not exactly an anxiety-reducing idea—we very much like to feel that we’re in control of everything we think. Even people who are experts in the idea of thoughts we don’t know we have struggle with the reality of this. Just this week I saw a post on (I think) Threads (which I unfortunately didn’t save and thus can’t credit here) in which a creator mentioned being asked to identify their implicit biases during implicit bias training when implicit bias is literally defined as “a negative attitude, of which one is not consciously aware”. “Tell me about the thoughts you don’t know you have” is an interesting instruction.

However, understanding how this subconscious thinking works and how we convince ourselves that what we did wasn’t actually bad (or was even good) helps us bring some of that subconscious thinking into our conscious mind and actively do something with it.

Making Bad Behavior Look Good

Many factors go into how we try to twist our descriptions of our behavior to make bad things appear good. If any of these are in your response (or even just in your thoughts around the situation you’re dealing with), that’s a sign that you’re about to make a mistake in your response.

Moral Justification

Moral justification is how we try to make our damaging conduct personally and socially acceptable, portraying it as serving socially worthy or for moral purposes. We then feel ‘safe’ to act on what we perceive as a moral imperative and inflict harm on others while feeling like we’re good people. If this sounds far-fetched, consider how ordinary people who join the military become soldiers willing to kill. They redefine the morality of killing in a combat scenario, and it becomes protection and defense, even in cases where actions go beyond those that can ever be justified (Nazis); as Voltaire said, “Truly, whoever can make you believe absurdities can make you commit atrocities.”

War is the extreme example. We also see moral justification everywhere in everyday life. When Elizabeth Holmes started Theranos, I doubt she set out thinking “I want to commit massive fraud to get rich.” She saw herself as a visionary revolutionizing healthcare and saving lives who had to cut corners to achieve her world-changing goals, so she likely engaged in some moral justification when none of that was materializing, convincing herself that lying to investors and patients was necessary to overcome obstacles, framing the endpoint as a moral imperative.

In a crisis statement, moral justification sounds like, “While my methods may have been unconventional, my goal was always to protect the interests of <xyz>.” And this type of phrasing fails because audiences generally won’t buy the premise that a noble goal excuses unethical behavior, especially when the person delivering it still hasn’t accepted that what they did was wrong.

Sanitizing Language

Bandura also highlighted how we change the way we speak about events and our involvement in them to distance ourselves from our actions and their impacts, as activities take on entirely different appearances depending on what we call them (a military operation that kills civilians ‘resulted in collateral damage’).

In crisis PR, sanitizing language makes for a horrible crisis response, and we’ve all seen attempts at crisis comms that highlight these bad responses. An affair is a “lapse in judgment.” A lie is a “miscommunication.” A massive data breach is a “security incident.” A systemic failure of leadership is an “oversight.” Firing thousands of people becomes “workforce optimization.” Remember Boris Johnson’s response to his government being caught breaking their own COVID lockdown rules? He didn’t apologize for breaking the law or going to parties when people weren’t allowed to visit their dying relatives; he said he believed he was attending a work event. Even in his eventual admission and the media reporting around it, the word ‘misled’ was used instead of ‘lied’.

Look out for agentless passive voice in apologies and statements. That’s sanitizing language. “Harm was caused”. “Mistakes were made”. “Procedures were not followed”. Accountability is missing from these distancing phrases; they make the speaker come across as a passive observer of events that happened rather than someone who actively played a part in causing the effects. Why this fails in crisis responses is that it (a) tells your audience that you’re not prepared to name and own what you actually did and (b) shows that you haven’t accepted the reality of what your actions caused. This makes the statement read as performative and does nothing to actually distance you from the effects of your actions.

Exonerating Comparisons

Another way we try to make bad things look better is by comparing them with other acts. How behavior is perceived can be influenced by what it’s compared with. This one shows up a lot in politics, where a politician who is criticized for a decision then immediately deflects by talking about things a previous rival administration did as vindication.

In crisis communications errors, this comes out as things like, “Our product had some safety issues; however, Competitor’s product is worse and, overall, we have a great safety record.” This approach fails because it deflects accountability to a person or entity that materially has absolutely nothing to do with your situation instead of owning your role.

Obscuring Personal Agency

Speaking of deflecting accountability, another way we convince ourselves our role in causing harm is less than it was in reality is by disengaging from our own personal agency. If we can convince ourselves we didn’t really cause something, maybe because we were just following orders or we were part of a larger group of people involved in the event, that minimizes our self-condemning reactions and then minimizes our ability to take responsibility.

Displacement of Responsibility

Displacement of responsibility happens when we use the “I was just following orders” defense. Bandura notes that authorities often keep themselves intentionally uninformed by creating implicit agreements and insulating social arrangements that leave those at the upper levels in an organization somewhat irreproachable. Then, when things go wrong, the subordinates are blamed and portrayed as misguided or overzealous.

This shows up from both directions in corporate crises, where mid-level executives caught in unethical practices claim that the culture demanded it or senior leadership implicitly authorized it, while senior leadership claims they had no idea what was happening, resulting in a cycle of responsibility displacement in which everyone involved can point at someone else and there’s no true accountability anywhere. Elizabeth Holmes, for example, tried to displace responsibility onto her former partner, alleging psychological abuse that impaired her judgment.

Diffusion of Responsibility

Diffusion of responsibility makes everyone responsible in a way that none of them end up feeling responsible. In group decision-making settings, otherwise reasonable and considerate people can make awful decisions that cause harm because any harm you cause as part of a collective can almost always be attributed to the behavior of someone else in the group.

The Boeing 737 Max crisis is an example of diffusion of responsibility, as the failures that killed 346 people resulted from a massive, complex corporate structure where responsibility for safety, design, and regulatory compliance was so diffused that catastrophic risks were normalized. Everyone involved was focused on their specific jobs (engineers on plane components; managers on cost and schedule; regulators who deferred to Boeing’s own safety assessments), and when the crisis hit, the initial responses failed to produce a centralized acknowledgment of failure, instead focusing on procedures, pilot error and systemic complexities.

Crisis statement errors from diffusion of responsibility typically use that passive voice I mentioned above, attributing error to something somewhere in the group but not stating who or what and where the actual responsibility lies.

“It’s Not That Bad”

Being able to justify harm to ourselves also requires that we minimize the effects of our actions, because as long as we reduce the impact of our behavior on others, we’re able to ‘shut off’ the feelings that cause us to self-censure. Bandara notes that it’s easier to harm others when that harm and resulting suffering isn’t visible to us. If decision-makers aren’t in the room with the individuals who have been harmed, they find it easier to morally disengage from them. In his paper, Bandara uses the example of keeping cameras and journalists away from battlegrounds during war, because the images that would come out would erode public support for military action. Because they would make the consequences visibly human. When we engage in the “It’s not that bad” kind of moral disengagement, what we’re doing is dehumanizing the victim, and that’s where a lot of crisis responses fail.

These are the crisis response failures that minimize the damage, sometimes not mentioning it at all, other times burying it at the bottom of a long statement that spends most of its time talking about the effects of the situation on the person/organization responsible rather than those affected. If you’re leading your statement talking about how much the situation has affected you, you’re at high risk of writing a minimizing statement that directly tells your audience that you care more about how your actions affected you than how they affected others.

Blaming the Victim

Related to minimizing the effect, we have blaming the victim. If we dehumanize the person or people our actions have affected, we reduce our capacity for empathy for them and make it easier to rationalize harming them. Again with the extreme example, war often involves presenting the ‘enemy’ in as dehumanized a way as possible to make it easier to kill them. In political rhetoric, we see dehumanizing language used about immigrants to make it easier to support politics that cause widespread suffering.

This type of dehumanization is rarely so explicit in corporate and personal crises, but it does show up. Think about statements regarding layoffs where people are described as “headcount” or the “workforce” or reduced to statistics; or where an individual is being ‘cancelled’ for harmful actions and describes critics as ‘keyboard warriors’ or ‘a vocal minority’. Any language that reduces the people harmed by your actions to an abstraction is a form of dehumanization and a barrier to taking accountability.

Closely related to this is blaming adversaries or circumstances, where people responsible for an issue view themselves as victims of it (e.g., when someone has done something wrong and they blame the media for the backlash they’re getting), or victims are blamed for bringing the effects you’ve caused on themselves.

In a crisis response, blame-shifting or misattributing blame is one of the absolute worst routes you can take because it implies that you don’t think you did anything wrong and the real problem is that your audience is too sensitive or too easily offended (we’ve all experienced people trying to make things sound like apologies that aren’t—”I’m sorry if you were offended…”).

If your response is heavy on things like, “The media took my words out of context”, “This is a witch-hunt”, and “Cancel culture is out of control”… (especially if this is what you’re leading with), take a step back before you write a statement that demonstrates a complete lack of accountability and moral compass regarding the situation.

Moral Disengagement Is Often Progressive

As Bandura notes, moral disengagement happens over time. Just like most crises aren’t one isolated event that blows up. Most are a result of a ‘slow burn’ series of choices and procedures that collectively weaken a person or organization’s ability to recognize impending disaster until it’s happened. A person may initially perform mildly harmful acts that they can tolerate with some discomfort; then, as their ‘tolerance’ for doing so becomes strengthened through all of the (often subconscious) avenues described above and they become more able to justify the action and diminish the harm, their level of ruthlessness can increase until they can do the things they would previously have found abhorrent without any negative feelings about them. The inhumane practices become routine. I don’t need to go into great detail with examples here as (I’m assuming) we’ve all been reading the news over the past few years…

So a crisis statement tells us about more than just the specific incident it’s addressing, as it’s also a reflection of the character/culture of the person/organization issuing it. A bad statement that’s full of sanitized language and diffusion of responsibility might indicate that someone needs better comms training, but it most often reflects the cumulative values of the person/organization that have, over time, built up to an extent that allowed the situation being addressed to happen. A CEO giving a performative apology for an ethical breach indicates a lack of concern and accountability regarding the breach on a surface level, but underneath that there’s often a progressive moral disengagement that has been allowed to continue unchecked in a way that made bad practices feel normalized enough that nobody saw the disaster coming.

This is the difference between crisis management and communications. The communications aspect, when done well, helps bring your audience’s perception of the situation closer to the reality of it that you are (again, if done well) taking responsibility for. The management aspect is getting into that underlying build-up of suboptimal practices that allowed the crisis to happen and putting steps in place to prevent it from happening again. Then, back to communications, letting your stakeholders know what you’re doing to make sure it doesn’t happen again. Because you have to acknowledge that “mistakes were made” to do all of this, effective crisis management and communications requires that you don’t do all of the moral disengagement described above. If you distance and dehumanize and minimize, you don’t see the problem. You can’t fix a problem you refuse to see. And this, by the way, is often how I get people who don’t want to take any responsibility to the point of accountability because even if they find it hard to initially accept any role in the matter, almost everyone can agree that they don’t want to be in the situation they’re in ever again. When accountability is the way through and out the other side of an issue and a sort of ‘protection’ against future similar issues, it becomes much easier to take than when it’s perceived as just ‘admitting wrongdoing’.

Taking The Performance Out Of The Response

It’s not unnatural to want to respond immediately when you or your organization is being pressured to make an apology. The problem with making statements or apologies (not every crisis actually needs an apology; sometimes only an acknowledgement is required) under pressure is that the most natural feeling while doing so is defensiveness, and the most common driver for making the statement is a desire to ‘make this go away’, which tends to lead to statements that look like accountability or apologies but are really just defense mechanisms. They’re performative, and they’ve probably got moral disengagement in them. Not sure if you’re writing one of these? Check for:

1. “I’m sorry if” statements

2. “It was never my intention” statements

3. Mentions of your words being taken out of context

4. Mentions of opponents or how others are representing you (‘those seeking to score political points’; ‘the media said’; ‘online mobs/bots’)

5. Statements that reiterate a value that your actions have shown you actually don’t have (e.g., “I’ve always supported xyz [when you’re being criticized for harming xyz]) or make vague promises (e.g., “I will continue to fight for what is right”)

All of these tell you where you’re falling into moral disengagement traps. #1 is blame-shifting to the audience’s reaction; #2 is minimizing impact by focusing on intent; #3 is displacing responsibility; #4 is misattributing blame; #5 is trying to use a comparison to make the current situation look better or shift attention to something good to minimize the actual harm currently being caused.

It can actually be helpful to write but not (I repeat, NOT) release an ‘apology’ or ‘statement’ that’s entirely driven by how you feel in the moment so that you (ideally with help from a crisis strategist who can help you understand how to take that initial response through to understanding and accountability) can look through where your thought processes are blocking you from strategically and honestly addressing the situation in a way that sets you up for rebuilding instead of burning everything to the ground. I’ve had clients do this because I want to see everything as it is in their mind in the moment so I know exactly what we’re working with and where to focus conversations when building the real approach.

You might write that initial statement and think it’s fine because it’s true to what you currently feel, and truth is extremely important, but the problem with this reactive truth-telling is that it’s premature. The truth that you want to get to is the truth that includes accountability and is built on rational strategy rather than that emotional, defensive reaction we all experience when we do something that we, regardless of how much moral distancing we’ve engaged in, wish we hadn’t done. When you write and release a statement from that reactive mindset, you end up with a statement that the public instinctively rejects because even if they don’t know the psychological underpinnings of a morally disengaged response, they pick up on it and recognize (a) the absence of genuine accountability and (b) the fact that you care more about making the issue go away than repairing the harm you’ve caused.

Once you’ve got that initial reactive response and looked for the flaws in it (remember, the biggest enemy is not, in fact, the media or your angry audience), you can begin asking the hard questions. It’s always going to be an uncomfortable process (which partially goes into how so many bad apologies make it to print… we don’t like uncomfortable processes and avoid them).

• What exactly did you do… (without the sanitized language)

• Who was harmed by your actions… (without minimizing the consequences)

• Why did you make the choices you made… (without avoiding acknowledgement of personal agency)

• How does what you did align with your values… (without moral justifications)

This is uncomfortable because you have to confront and break down the psychological defense mechanisms you’ve built to protect yourself when you’ve done something that, when you look at it without those defense mechanisms, makes you feel bad about yourself. That’s the hard part about this. Nobody wants to feel bad about themselves, and everything I’ve mentioned in this post is something we all naturally do to protect ourselves; these mechanisms aren’t in themselves moral failings. They’re just not conducive to actual behavioral change and reputation rebuilding. Almost every natural instinct we have in a crisis is one that will make our response bad (I’d like to make a complaint to whoever designed our brains, honestly). So here’s something that might help you accept the discomfort of working through your truths: there’s nothing inherently wrong about being wrong; it’s an opportunity to learn and improve for the future. Furthermore, doing a bad thing doesn’t make you a bad person. What makes a bad person is doing a bad thing and then continuing to do it regardless because you can’t acknowledge that it was bad.

Finally, acknowledging harm is not the same thing as expressing intent. We get intent and effect so twisted together in a way that makes us express both inadequately in a crisis. We want to make sure we don’t come across as though we intended to hurt people, so we focus on what we actually intended to do and leave people thinking, “OK, but you’re ignoring the effect you had”, and then we get caught up in thinking that if we acknowledge the effect we had, we’re basically saying, “I intended to cause this harm.” But you’re not. You can acknowledge the resulting harm in a way that does not implicitly say that you caused it intentionally. I once reversed my car through a closed garage door. I thought I’d opened it already and wasn’t paying attention. It caused $600 worth of damage. Nothing about that suggests I meant to do that, right? “The garage door didn’t open, and now I’m stuck with a $600 bill I can’t afford. I don’t know how I’m going to pay for this and groceries for the rest of the month. I can’t believe these companies make garage doors that cars can just drive through without a warning. I was just trying to get to work; I didn’t mean to do it. The door should have opened…” All of that was my initial response, completely devoid of accountability.

When you’ve removed your moral disengagement from your thought processes and confronted the actual truth in the matter, you can construct a response that is honest and takes accountability for the right things. That response should contain:

• Acknowledgement: Name exactly what you did wrong with no minimizing language, in active voice, without euphemisms and sanitizing language.

• Accountability: Name the impact of your actions and where you are accountable. With specificity. No excuses or “I’m sorry if…” statements.

• Explanation: Explain why you did what you did; again, no excuses. Put your actions in an honest context (people respect an honest explanation that makes you look bad more than a deflection that tries to make you look better).

• Commitment to Change: Explain what you’ll do to make sure it doesn’t happen again, making sure that this is specific, achievable (don’t set yourself up with promises you can’t keep), and directly tied to the harm caused.

Again, this will probably be uncomfortable to do, because most apologies are; they necessitate vulnerability that most of us never actively want to show, and it breaks down our defenses, requiring moral engagement and the willingness to look at what we did and actually feel the consequences of being disappointed with our own actions. It’s easier to blame the media and use sanitizing language that makes us feel better than it is to say “I did this, I was wrong, and I’m sorry”, but the moral disengagement that goes into that is what makes ordinary people do extraordinary damage without guilt, and none of us want to be that person.

You can also look for the aspects of moral disengagement I’ve discussed here next time you come across a crisis response ‘in the wild’ to help you decide how to take it. Look at what’s not being said, whether they’re using the active voice or passive voice, whether there’s any acknowledgement of harm or commitments to change. Is there evidence that the person has done the work to understand the reality of what they did? If you’re not seeing those things… you might be looking at a psychological defense mechanism.

Source:

Bandura, A. (2002). Selective Moral Disengagement in the Exercise of Moral Agency. Journal of Moral Education, 31(2), 101–119. https://doi.org/10.1080/0305724022014322