Moving from Blame to Accountability (2017)
thesystemsthinker.com
thesystemsthinker.com
Accountability is synonymous with responsibility. Identifying who is responsible for a failure is synonymous with blaming. We prefer to use "hold accountable" in formal and polite settings, and "blame" in informal settings, but it has the same meaning.
Using the correct level of politeness and formality is a hygiene factor.
"There has been a failure in accountability, and we're looking into preventing in the future" can be equivalent to (informally and impolitely) "someone here isn't cutting it and we need to figure who that is so we can start getting rid of him".
What the article actually means is that the consequences for failures in accountability should be measured such that it encourages learning, and disclosing problems, and so on. But this needs to happen through demonstrated actions, not language. It's actually preferable to be blunt, honest, and fair: "someone here fucked up; if you can figure who did what, fix it, and coach everybody up on how to prevent it, you will be more respected. if you hide your mistakes you will be less respected, and eventually of little use here."
If you e.g. push your employees over the limits and tell them to ignore all kinds of reasonable safeguards and regulations. And then, one day, after a ton of near misses Joe slips and falls into the meatgrinder, it would be very easy to blame Joe for skipping the safeguards.
But the true reason for the accident was the environment and culture Joe had to work in and it might have also hit a random different person.
The most evil variation of this is "we'd rather have the increased productivity than spend the resources to actually avoid X so our guidelines say they're to avoid X but really they're to blame a line worker when X happens so we can keep saying we don't want X". See the rash of train derailments and toxic spills we're lately now that railroads have imposed "precision railroading/precision crew scheduling".
Also: A organization can have foggier responsibility structures than "the boss says $X which is why $X is done".
E.g. you could have a boss who says safety is truly important and even does a ton to help with that, while lower management is so obsessed with other metrics that they sabotage that on a conscious or subconscious level while emulating compliance to their higher ups.
This is very common.
The "evil boss"-scenario is the simple one where it should be very clear that Joe is not to blame. This is precisely why it is important to also look at less clear ones where it would be easy to blame the accident on individual human error — instead of finding the actual root causes.
A formal blaming process means that the question of who to blame is decided objectively and people have more of an incentive actually act correctly.
And as others mention, blaming the process itself is reasonable.
And you're right that the article itself only presents accountability as essentially greater politeness.
I am not sure the creation of a formal process (created by those who have the power to create formal processes) is inherently more fair to those without power than an informal process.
Especially if formality is confused for objectivity or correctness. The latter two are likely impossible for human beings to acheive, so a starting point for a better process is going to be recognizing that the it is going to be affected by the blindspots, flaws, and other foibles of the people who created it.
A formal process isn't inherently more fair. It's only potentially more fair. You need to look in detail at the process to see if it's more fair.
The rest of your argument is a bit like arguing mathematics won't help you build a bridge - "The math could wrong and the user could be biased and even the best can't reflect reality exactly".
When something went wrong, and it was my fault, I was expected to take accountability. This was rewarded by not having folks try to smear or attack me. It was a simple statement of fact. We found the problem, and took steps to mitigate (which could include removing me from the project).
No rancor, no name-calling, no status games.
Needless to say, this did not translate to the American management, who liked to do a sack dance.
I've on occasion also educated up the chain on why attacking someone is an anti-pattern. If you can get the higher-ups onboard, then attacks tend to be viewed as a problem in and of themselves. Or, and this is certainly possible, I'm incredibly naive and have just been very lucky.
Pro tip: If you have a manager who acts like this, or even worse, an org, then it is best to find a better work environment. Having said that, these scenarios tend to be in the minority.
One of my best jobs had the department head say to his staff when things go wrong: "I don't want to hear X screwed up. I want to hear why we allowed X to screw up". Why could a junior employee make a change that affects the whole org? What systems are not in place to prevent this? Etc.
There is a tradeoff between frequency and severity of screwups and rate of progress. This tradeoff should be made deliberately; just going from "screwups always bad" will not reach the best balance for the health of the organization.
A better approach is to automate the process in a way that the manual mistake is no longer possible.
He's writing from the perspective of investigations after airliner crashes. His core point is that you shouldn't be looking for who to blame in a punitive sense (assuming nothing malicious has happened), but rather how policies and procedures allowed the error to happen. The human is just the last domino in the chain. Accountability ultimately lives with management that establishes policies and procedures. But management often does't like that idea, because yelling loudly and sacking someone feels like you're being proactive, even though in the net it makes another error more likely.
I have no power to fire other groups, but what I can do is list all the ways the other group will fail and how we have to handle their failures and what that means for the customer. Those groups will never improve, but because I look like a soothsayer, I will get rewarded and promoted.
Is this healthy for the org? Of course not. But I am not the org. If anyone had my insight a few levels up, they would fire incompetent teams and let teams with proven track records take over. I’ve never seen this happen, so I’ll hold my breath.
This seems rather extreme. To me "accountable" means "accountable for your _reaction_" in this context. I don't think (or at least hope that) anybody advocating this stance is seeking to "eliminate nature" so much as accept it as it is and deal with it on a realistic basis.
As an illustrative example I ride motorcycles on the street. It is a fact of nature that people are going to try and merge into me or cut me off and threaten my life. Blaming them and being angry about it leads to poor outcomes in the form of aggressive and dangerous riding. Instead I focus on mitigating risk by wearing appropriate safety gear and anticipating bad drivers' actions so as to avoid it. I do not seek to control others so much as to maximize the value of my interaction with them by taking ownership of it.
So who's fault is it that you got wet? The weather? Was there nothing that could have been done?
Maybe it's your responsibility to plan for bad weather?
The basic flaw in this thought exercise of yours is, no one is actually '"accountable" for everything' with the power or desire to enact the consequences you listed. That said, there is nothing wrong with modeling that I am accountable for everything and accepting chaos theory as a unavoidable attribute of the world. This means that me being accountable and things going wrong does not necessitate any of the things you listed "independent behaviors by others must be extinguished" as this is just a sociopathic way of thinking on the path to destruction.
Also GP's point refers to "Extreme Ownership" by Jocko Willink, and from that title it looks fair to effectively imagine the extreme.
GP mentions how focusing on accountability changes the approach to failures, and I'd totally see people taking the wrong lessons from a project where they didn't stop a colleague from doing something that happened to be a bad decision.
You speak for yourself, not for me.
Those different than me often contribute in ways I did not see. Life is better when I embrace improvement.
Far too often are we outsourcing in order to avoid the difficult and often times very wholistic concern of risk and outcome. Blame serves no shareholder value but to abdicate leadership from their post.
Yes I write highlevel risks about the information I'm given could be wrong (to cover my backside) but you have to be thorough. Its makes you a better technical person or a better engineer. where by you dissect and rebuild information given to check it's right and to spot inconsistanses.
It's like solving a puzzle, but like a good puzzle it can be sometimes tedious, slow and time consuming. but when you find issues that could cost hundreds of thousands of pounds/ dollars through mistakes it really makes up for it.
Leadership
Ownership is certainly a primary part of that, but there is more to this than what most people from the corporate world can envision unless if you have been at a director level (or higher) in a company that is more than a start up.
Ownership directly implies liability. If you fail wrongly this can mean losing your job, permanently losing your career, losing personal money, or even going to prison depending upon the context. Many professions have that, but software does not. In fact software directly stresses the opposite of this: https://news.ycombinator.com/item?id=35889282
Leadership means to own risks and liability, but it also means owning the welfare of the people that work for you. Another primary part of leadership, equivalent to ownership, is communication. Good communication comprises the ability to provide specific instructions to subordinates so that they never must ask for clarity, as though you are using empathy to read their minds. Simultaneously it also means the ability to persuade peers and superiors to change position on an opinion beneficial to your team/mission. If you are good at this it will appear almost invisible to your staff and exceedingly self-sacrificing to your superiors.
The opposite of blame seeking is in my opinion embracing failure and understanding (as an org) that innovations and progress is made by the willingness to fail.
Seeing failure as the result of hard work, a necessary evil, causes blame seeking to vanish imho. Together with accountability this sets the organization op for incremental improvement. Never blame a person. Blame the system or process. The one we made up together.
The real difference is made by the kind of people who interpret mistakes and respond to them. They need to have passion for the job, be perfectionists who search to improve their output and who take negative feedback as valuable information. That's partly nature, partly nurture, but by the time the people get a job it's way too late to fix anything, let alone by using one word instead of another.
But it's also very hard for individuals to not feel blamed when given feedback meant for improving the process.
A fairly common example I like that you've probably heard before is variations on the story of a junior engineer accidentally dropping a production database. Any feedback to that junior engineer about what they need to improve on to avoid doing that in the future is going to feel like blame no matter how it's given since the underlying causes of the problem are at least as much organizational ones.
This etymology also works in other languages - for instance, in German: “Ver-Antwort-ung”.
> Yes, if I don't have the information or knowledge required to answer your question, I will respond by saying "I don't know" or a similar phrase that indicates my lack of knowledge on the subject. However, I will do my best to provide you with the most accurate and helpful response possible based on the information and capabilities available to me.
And sure enough on obvious things it can't know it will say that. "What color is my shirt" for example.
The problem is it will bullshit as confidently as the loudest pub bore, and while it's sometimes right, it's often wrong
Workplace dysfunction (including the blame- and accountability issues from the article) is either coming from issues your boss knows about and can't fix, or considers acceptable, and either way you better work within the given constraints. Any solution about a griefing neighboring department will likely come through some tacit reorganizations when management reshuffles.
Not saying there is anything wrong in particular with the article, or that working through the article can't help if you truly want to reform a team; however such a task will require a manager with sufficient leadership skills and intuition about the true nature of an organization. Such a manager would better not put up powerpoint slides about the difference between blame and accountability at the end of the next daily standup. It's the intangible sprinkling of management magic that makes each team member feel understood and challenged to do better, and the diplomat's skill of negotiating a truce with the enemy within the company that no systems-thinking-article can teach.