← Back to context

Comment by fc417fc802

14 hours ago

> There’s a toxic idea built into this statement: It implies that the real root cause is external to the people and therefore the solution must be a systemic change.

Not necessarily, although certainly people sometimes fall into that trap. When dealing with a system you need to fix the system. Ejecting a single problematic person doesn't fix the underlying problem - how did that person get in the door in the first place? If they weren't problematic when they arrived, does that mean there were corrosive elements in the environment that led to the change?

When a person who is a cog within a larger machine fails that is more or less by definition also an instance of the system failing.

Of course individual intent is also important. If Joe dropped the production database intentionally then in addition to asking "how the hell did someone like him end up in this role in the first place" you will also want to eject him from the organization (or at least from that role). But focusing on individual intent is going to cloud the process and the systemic fix is much more important than any individual one.

There's also a (meta) systemic angle to the above. Not everyone involved in carrying out the process will be equally mature, objective, and deliberate (by which I mean that unfortunately any organization is likely to contain at least a few fairly toxic people). If people jump to conclusions or go on a witch hunt that can constitute a serious systemic dysfunction in and of itself. Rigidly adhering to a blameless procedure is a way to guard against that while still working towards the necessary systemic changes.

Often institutions develop fundamental problems because individuals gradually adjust their behaviors away from the official norms. If it goes uncorrected, the new behavior becomes the unofficial norm.

One strategy for correcting the institution is to start holding individuals accountable. The military does this often. They'll "make an example" of someone violating the norms and step up enforcement to steer the institutional norms back.

Sure it can feel unfair, and "everyone else is doing it" is a common refrain, but holding individuals accountable is one way to fix the institution.

I agree with most of what you said but i'd like to raise 2 points

1) the immediate action _is more important immediately_ than the systemic change. We should focus on maximizing our "fixing" and letting a toxic element continue to poison you while you waste time wondering how you got there is counterproductive. It is important to focus on the systemic change, but once you have removed the person that will destroy the organization/kill us all.

2) I forgot. Sorry

  • I suppose that depends on context. I think it's important to be pragmatic regarding urgency. Of course the most urgent thing is to stop the bleeding; removing the bullet can probably wait until things have calmed down a bit.

    If Joe dropped the production database and you're uncertain about his intentions then perhaps it would be a good idea to do the bare minimum by reducing his access privileges for the time being. No more than that though.

    Whereas if you're reasonably certain that there was no intentional foul play involved then focusing on the individual from the outset isn't likely to improve the eventual outcome (rather it seems to me quite likely to be detrimental).

> how did that person get in the door in the first place?

is answered by:

> any organization is likely to contain at least a few fairly toxic people

  • Exactly. The above comment is an example of the kind of toxic blameless culture I was talking about: Deflecting every problem with a person into a problem with the organization.

    It’s a good thing to take a look at where the process went wrong, but that’s literally just a postmortem. Going fully into blameless postmortems adds the precondition that you can’t blame people, you are obligated to transform the obvious into a problem with some process or policy.

    Anyone who has hired at scale will eventually encounter an employee who seems lovely in interviews but turns out to be toxic and problematic in the job. The most toxic person I ever worked with, who culminated in dozens of peers quitting the company before he was caught red handed sabotaging company work, was actually one of the nicest and most compassionate people during interviews and when you initially met him. He, of course, was a big proponent of blameless postmortems and his toxicity thrived under blameless culture for longer than it should have.

  • Of course. I actually think that "we did everything we reasonably could have" or "doing more would be financially disadvantageous for us" are acceptable conclusions for an RCA. But it's important that such a conclusion is arrived at only after rigorously following the process and making a genuine high effort attempt to identify ways in which the system could be improved. You wouldn't be performing an RCA if the incident didn't have fairly serious consequences, right?

    It could also well be that Joe did the same thing at his last employer, someone in hiring happened to catch wind of it, a disorganized or understaffed process resulted in the ball somehow getting dropped, and here you are.

> Ejecting a single problematic person doesn't fix the underlying problem - how did that person get in the door in the first place? If they weren't problematic when they arrived, does that mean there were corrosive elements in the environment that led to the change?

This is exactly the toxicity I’ve experienced with blameless postmortem culture:

Hiring is never perfect. It’s impossible to identify every problematic person at the interview stage.

Some times, it really is the person’s own fault. Doing mental gymnastics to assume the system caused the person to become toxic is just a coping mechanism to avoid acknowledging that some people really are problematic and it’s nobody’s fault but their own.

  • On the contrary. It's all too easy to dismiss as being the fault of a fatally flawed individual. In fact that's likely to be the bias of those involved - our system is good, our management is competent. Behead the sacrificial lamb and be done with it. Phrases such as "hirinng is never perfect" can themselves at times be an extremely tempting coping mechanism to avoid acknowledging inconvenient truths.

    I'm not saying you shouldn't eventually arrive at the conclusion you're suggesting. I'm saying that it's extremely important not to start there and not to use the possibility of arriving there as an excuse to shirk asking difficult questions about the inner workings and performance of the broader organization.

    > Doing mental gymnastics to assume the system caused the person to become toxic

    No, don't assume. Ask if it did. "No that does not appear to be the case" can sometimes be a perfectly reasonable conclusion to arrive at but it should never be an excuse to avoid confronting uncomfortable realities.