A few months ago, I attended a talk at Mozilla by Ted Nyman (of Github) titled “Scaling Happiness”. The video is freely available.
Nyman argued that companies with minimal formal structure are better for workers (specifically, better at maximizing workers’ happiness) than more traditional companies. Whenever someone asks whether something is “better”, I ask “better for whom?” Whose happiness was Nyman talking about? He didn’t say, but when I think about happiness, I ask what’s best for women, for people in GRSMs (gender, romantic and sexual minorities), for disabled people, and for people of color, since not too many people seem to think about what’s best for people in these groups. (For the record, I’m in the second and third of those groups, though I’m usually not perceived that way in one case, and often not perceived that way in the other.) Happiness for the dominant cadre in the software industry — that of people who have white privilege, male privilege, cis privilege, and heterosexual privilege, and who lack visible disabilities — is not the same as happiness for everybody.
I don’t mean to say that happiness is a zero-sum game, that when abled white cis men are happy, that inherently takes away some of the limited pool of happiness from disabled trans women of color. Rather, part of the problem is that people who have privilege perceive happiness as a zero-sum game; part of their happiness comes from seeing themselves as better than others.
I think most people in the tech industry or in open-source or free culture communities know what I’m talking about when I say “structurelessness”. Perhaps you work at a “flat” company that encourages employees to make up wacky job titles to put on their business cards, calls everybody a “team member”, or renders everyone uncertain about who their boss is. Or maybe you’ve only worked at more structured, hierarchal organizations: ones with managers, a complicated organizational chart, ranks, and hierarchy. You probably know the distinction even if you’ve only been on one side of it.
Does structurelessness eliminate competition, abuses of power, and status hierarchies, or does it just drive them underground? To break down the question, let’s look at a few ideas about structurelessness, some of which are from Nyman’s talk and others are just things I keep hearing from people in the free/open-source software and culture world.
- Authenticity: people are happier when they are able to be who they really are at work.
- Informality: people are happier when they’re able to be informal at work, such as by wearing T-shirts with holes in them or saying “fuck” a lot.
- Conduct: formal mechanisms for guiding behavior aren’t that important, since in a healthy organization, people will just be nice to each other.
- Leadership: people whose job it is to manage aren’t necessary when people can just manage each other.
- Accountability: formal goals and performance metrics just get in the way of getting a job done.
As a nod to structurelessness, I’ll take on these points in no particular order.
First, why would a feminist argue in favor of structure when structure so often means hierarchy, and hierarchy is deeply entwined with oppression?
It’s true, I’m not a big fan of hierarchies. Maybe they can be used for good, but I haven’t seen a lot of that in reality. At the same time, though, it’s also a fallacy to think that simply declaring we’re not going to have hierarchies makes hierarchy go away. Jo Freeman wrote about this in the ’70s, in her essay “The Tyranny of Structurelessness”. Based on her experiences in feminist organizing, she found that groups of people (like feminist consciousness-raising groups) that declared they weren’t going to have a formal structure devolved into unofficial hierarchies, which were much harder to challenge and hold accountable.
Do you trust people to see you for who you are? I mean the question in two senses: (1) Do you believe that it’s even possible for you to communicate who you are to others without a great deal of effort, and (2) Do you trust others, as a general rule, enough to assume they will behave cordially towards you once they know who you are?
In his talk, Nyman talked about how in typical companies, many relationships are inauthentic. That is, people don’t act towards each other in the ways that they would in the absence of a rigid, externally imposed set of relationships. At least stereotypically, people don’t behave towards their bosses, or to their subordinates at work, the way they behave with their friends. He argued that people are happier when they can present themselves authentically and have authentic relationships, and that a less structured organization fosters such relationships.
If you are queer, or trans, or have mental illness, or all of the above, you probably know something about the perils of presenting yourself as you really are. Dan-Savage-style coming-out narratives notwithstanding, many of us who are placed socially in these ways find that we cannot be completely authentic in all aspects of our lives. I definitely want to express myself, but I have to balance that against other needs, like being able to make a living in a capitalist society. If I dressed the way I’d prefer to, if I talked more openly about the times when my depression and anxiety prevent me from getting work done, I might find it harder to fit in, to stay attached to a professional group, to stay employed, than I already do. So instead, I wear T-shirts and cargo pants, and I let people think (at times) that I’m merely disorganized or not that committed to what I do.
In my opinion, it takes a lot of privilege to assume either that greater authenticity leads to greater happiness, or that the only reason you would leave who you are at the door when you step or roll into work is the formal, organizational structure of the place where you work.
Moreover, being your authentic self in front of somebody else requires trust, and outsiders have very good reasons not to trust insiders. For me, part of what I mean when I say I lack a certain amount of privilege is that every day at work, I make calculations about who is safe to interact with and who is unsafe. Of course, there are degrees of safety and it’s not a binary choice. For example, every time someone uses “crazy” as a pejorative — suggesting that what I am is also a label to insult an idea with — that decrements their “safety” score inside my head. Almost everyone uses this word in this way — even I still do, given that I’m not free of internalized ableism — which is why I say it’s not a binary property. If my company became totally flat and got rid of all structures, processes, and goals, I wouldn’t be able to have authentic interactions just because of that. I’d still have this calculus of safety I have to apply all the time.
And what about when who you are makes people uncomfortable? If you’re queer, trans, kinky, poly, disabled, you probably either have spent a lot of time trying to blend in, or you have stories about when people become uncomfortable upon realizing some aspect of who you really are, and having to comfort them. (Or both!)
To take a completely different example, do you really want to encourage people to be “who they really are” when who they really are is a harassing creep? Maybe having to be a bit inauthentic at work serves an equalizing function, like a uniform. If you know what the rules are, it’s more likely that you’ll be able to follow them and less likely that you’ll be cast out for breaking a rule you didn’t know existed.
Informality and isolation
Usually, no one tells little kids on the playground who to play with and who not to play with. But even very little kids start forming hierarchies of exclusion when left to their own devices. Vivian Paley’s work, as documented in her book “You Can’t Say ‘You Can’t Play'” showed both that in groups of kindergarteners, leaders emerged who got to decide which kids got to play and which kids got excluded; and that a teacher could change that by imposing the simple rule that “you can’t say ‘you can’t play'”. And increasing the amount of inclusion in the group made the kids in it feel more accepted, on the whole. An advocate of structureless organizations might argue that Ms. Paley should have just let her pupils be their authentic selves and form their own social alliances. But at least according to Paley’s account, imposing the rules made the kids happier — contrary to Nyman’s claims about structurelessness and happiness.
Now, perhaps kids are just different from adults. I also don’t think it’s necessarily human nature to form hierarchies in the absence of formal rules. Fundamentally, I don’t care whether that’s because of nature or nurture. No matter what combination of nature or nurture it is, as human beings we have the latitude to choose what we will value. Personally, I value inclusion, and while I can’t prove logically to somebody else that this is something they should value too, I think there’s plenty of evidence that inclusion and the overall happiness of people in a group correlate. And, as Paley’s kindergarteners show, inclusion does not necessarily naturally arise from structurelessness.
Isolation is closely related to an insidious way in which people who believe themselves to be good can perpetuate oppression: the withholding of mentorship. In another context, that of law schools, Pamela J. Smith wrote about how even when Black women gain admission as law students, informal social barriers to the development of mentoring relationships with faculty members are a form of discrimination that is difficult to challenge (“Failing to Mentor Sapphire: The Actionability of Blocking Black Women from Initiating Mentoring Relationships”, reprinted in Critical Race Feminism, Adrien Katherine Wing ed.)
Informal mentoring between apparent peers is mediated by social power dynamics as well. In her book Leaving The Ivory Tower, Barbara Lovitts wrote about the importance of tacit knowledge in determining whether Ph.D students succeed or fail. Many graduate programs are quite structureless in a day-to-day way; despite having a clear hierarchy (tenured faculty, tenure-track faculty, non-tenured instructors, postdocs, grad students), new graduate students must navigate a system with very little formal structure in order to learn the unwritten rules of the game. The difference between being a popular person and an unpopular one in grad student social groups can be the difference between academic success and failure. Would fewer grad students drop out because of isolation if there was a more formal process for initiating beginning students?
In my personal experience as someone who, earlier in my life, didn’t resemble most of my colleagues, lack of mentorship is a major structural barrier to success both as an academic computer scientist and as a software engineer. And I think lack of structure translates into lack of mechanisms to encourage formal mentoring relationships — something that has a disparate impact on women, people in gender, romantic, and sexual minorities, people of color, disabled people, and everybody else who may not feel comfortable approaching someone of higher social status to ask for support.
Likewise, people with disabilities that affect how they process tacit social cues — such as people who are on the autism spectrum — may have a much easier time contributing harmoniously when the rules are made clear than when they must access all resources by guessing at a system of unwritten rules. Since ability to write software isn’t contingent on being neurotypical, barriers to entry for neurodiverse people mean excluding a portion of the talent pool for no particular reason.
When a marginalized person joins an organization, in the absence of structure, isolation and lack of mentorship can combine to render them powerless and unable to ask for — or perhaps even express — what it is that they need. In such a situation, it’s easy for that person to then be labeled “unproductive” by the very community that has, without even knowing it, made it impossible for that person to learn and grow. Formal structures are one way to level the playing field and make sure that everyone has the same opportunities, regardless of whether senior folks in the organization find them initially easy to relate to or identity with.
Codes of conduct and diversity expectations
I don’t know how a structureless organization would maintain or enforce a code of conduct. Maybe in such an organization, everyone just likes each other so much that it’s not necessary to have one. But codes of conduct aren’t needed because people aren’t nice or because they don’t like each other; they’re needed because different people have different expectations about what kind of behavior is appropriate in which contexts. It doesn’t seem to me like getting rid of formal structure solves that problem.
Codes of conduct are just one way to help a group become or remain diverse, by ensuring a safe environment for everyone and providing mechanisms to address breaches of that safety. Without formal structures, how does a company make and keep itself diverse? While the practice of affirmative action is often inaccurately derided as “quotas”, a few tech companies do go as far as to institute numerical quotas for hiring women. I would suspect that such a practice, and even more flexible affirmative action concepts, would conflict with informality. But how does a structureless organization avoid devolving simply into hiring friends?
In general, how do you make sure that an organization without structure doesn’t default to recreating the same power hierarchies that exist in its underlying society? I asked this question during the question and answer period at Nyman’s talk, but it got a little lost in translation. Nyman’s answer amounted to “we won’t hire racist or sexist people”. But that’s not good enough. Everyone raised in a white supremacist society has unconscious racism, and everyone raised in a patriarchy has unconscious sexism. It’s obviously inadequate to dismiss the possibility of recreating systematic oppression “because most of us are good ethical people”. Nyman himself admitted that Github is getting less diverse.
Unless it literally consists of a collection of people, each working alone — in which case you’d wonder what makes it an organization — in an organization without people formally titled “manager”, people will have to step up to manage each other at least sometimes and to some extent. How do you take initiative and assert power — in the absence of a structure that makes that power legitimate — when you’re already culturally oppressed and disempowered? If nobody is a manager, who will be most successful in, say, asking that their team institute a “run regression tests before committing code” policy: a tall, white, able-bodied, cis man; a short, Latina, disabled, cis woman; or a fat, Black, genderqueer person? When is it possible for people to really treat each other as equals, and when do they infer hierarchies when not given a formal hierarchy to look to?
What about when you’ve been punished in the past for trying to regulate others’ behavior instead of “knowing your place”? If you’re perceived as female, knowing that girls who assert power get called “bossy” and women who assert power get called worse, but also knowing that your leadership skills will eventually be called into question if you don’t assert power, structurelessness starts looking like a double bind.
Without goals and performance metrics, how do people get held accountable? I don’t just mean accountability for delivering on the promises one makes as part of doing one’s job. How about, for example, not finding a subtle way to fire somebody for discriminatory reasons and make it look like it was performance-related?
In his talk, Nyman acknowledged that more “formal” processes are necessary for handling harassment: he acknowledged, “you can’t just go to anyone” if you’ve been harassed. But what else, falling short of “harassment” as such, might require a formal process?
I’ve been pretty negative about structureless organizations. But there might be positives. Are they more open than more traditional companies to people with less formal education, or whose biographies are otherwise non-traditional? (I don’t know.) Do they make it harder for entrenched managers to retain power by virtue of seniority? (Again, I don’t know.)
To be fair, there isn’t just one set of processes that could arise when an organization sets aside formal structure. The majority could end up ruling most of the time. Or an organization could make decisions based on consensus. Or it could be cloyingly called a “do-ocracy”, in which decisions get made by whoever has enough time and energy to implement the consequences of the decision. I still think there’s the risk of majority rule, though, and the problem with that is that decisions about basic rights, respect and dignity can’t and shouldn’t be made by a majority. Where do basic rights, respect, and dignity come into this discussion? The number of occupations that are at least potentially a route into the middle class, at least theoretically available to anyone who has acquired a certain skill set that is possible for anyone dedicated to acquire, is steadily decreasing. If you’re in a social class such that you need money to live, learning how to program isn’t a bad way to go. But that will only continue to be true if tech company jobs are open to any qualified candidate, without the hidden price tag of humiliation based on one’s race, gender, disabilities, or sexual orientation.
Majority rule is, then, a problem because majorities often opt to keep minorities in their place for the benefit of the majority. And yes, a group made up of entirely people who see themselves as good and ethical can and will deny basic rights, respect and dignity to people based on gender, sexuality, ability, race, class, and other axes of oppression. The world might be different someday, but we can’t get there by pretending we are there.
Your thoughts, readers?
Thanks to Geek Feminism bloggers Sumana, Mary and Jessamyn for their comments.
I agree pretty strongly with almost everything you say, and I think I will leave my disagreements for another time.
Hierarchy isn’t something imposed on human beings, it’s something that’s built-in to being human. We are wired for dominance behavior, and it’s probably better to have that out in the open, where we know what the rules are, as you say.
I’ve had at least one co-worker say to me, about a fairly flat, open company “I’m glad there’s so little politics here.” When in fact, at that very moment, I was experiencing some fairly toxic politics. Flat organizations can make it worse because they can often be a green light to all sorts of dominance behavior by all individuals.
I think the qualities that produce satisfaction in workplace participants are competence, autonomy, and connection. I think the desire for autonomy is expressed multiple times in Nyman’s talk, but also the desire for competence. However, I think what’s also being expressed is a desire to not have to negotiate with co-workers, and this is, well, impossible.
“We are wired for dominance behavior”
I am skeptical of claims like this because they justify existing power structures so well, and science (particularly social science) has such a history of reasoning backwards from claims that justify existing power structures to evidence selected to support those claims.
I’m all for skepticism. In your piece, you’re making the case that dominance behavior is universal, just not stating it so baldly.
I am skeptical of most of the claims of Evo Psych, but I don’t believe in the blank slate, either. You’ve just told me you don’t trust research, so citing some will be useless here, so let me put it in my own words:
We’re all creatures of dominance and submission, however, we also get to be mindful about it, and to remap those drives toward pro-social ends. For example, you can strive to defeat your own procrastination, the weight-cycling industry, or sex discrimination, or whatever. These are all exercises of the drive to dominance.
I don’t accept this particular ideology, and request that we drop this thread, as I’m feeling rather ‘splained to.
I seriously, honestly, do not know what you mean by “I’m feeling rather ‘splained to”, and I’d like to.
It seems I’ve committed some breach of discourse, but I’m clueless as to what that might be. It’s like one of those unwritten rules you wrote about.
No, sorry, it’s not my job to educate you. Learning to be respectful and not condescending is everyone’s responsibility.
No, see, I don’t read Tim’s statement as claiming non-belief in research. I read your statement as super condescending, only very vaguely on-topic, and also incorrect.
Dudes who have institutional power making these kinds of “let me explain how you’re wrong about things” derails = ‘splaining. Do you require further education, Jay?
This is absolutely brilliant. I feel uneasy about “flat” organizations, too – and I’ve had very difficult experiences with flat organizations outside of tech, too.
Totally agree with Doctor Jay’s comment that connection is an under-referenced component in workplace satisfaction. I would add that for most of us, anyway, a sense of meaning is important – ie that our work feels like it contributes to something meaningful.
I’ll be referencing this piece often, Tim. Thank you for doing a brilliant job.
I’m also glad you treated this at length, Tim. You’ve wonderfully articulated so many points I’ve felt many times, not just working in tech but also in other job situations that purported to value everyone equally. Adding to my anguish was–until now–no idea how to express what I grew to think of as betrayal.
I found this far too apt for me today as I’m about find out if I get to keep my first professional job as a librarian in a school which claims to have no hierarchy. I was hired to rebuild the library but those in charge who control things in a very top down manner have never given me respect.
These issues of mentorship especially ring true to me, because most of the structure is within departments, I don’t have one. I have a colleague on the lower campus but the school knows it should have working libraries but doesn’t value what we do. And since I’m new and was hired to rebuild a program that they haven’t looked at with that much priority, I’ve been on my own. I got told that I didn’t do enough, I didn’t assert myself enough as I felt unsure and isolated. I’m lucky that I do work as a librarian because I’ve made a lot of informal connections with kids and small ones with faculty where I’ve helped them out. They just haven’t translated enough for those in charge.
There’s something very perverse about an organization that says we have no hierarchy, but oh the choice of if you get to keep your job comes down to three people and not the community you’ve been serving. I would lay even money on it that if they end up firing me, the community will be surprised and not happy.
Thank you for this post, I needed it to remind myself that they don’t get to decide my value. My patrons do.
This is spot on the mark. I’ve experienced all the things you describe: being more inauthentic in a structureless group (because I had to impress everybody all the time, not just the manager I was having 1:1s with); more than usual difficulty being an effective leader as a woman; isolation; being perceived as less productive because I couldn’t ask for the structure I needed.
Furthermore, I noticed something that’s a dark secret of flat organizations, which is that they often have a tyrant at the top anyway, one who gets more control of everybody because there are so few managers empowered to protect their people. If that person exerts their control only occasionally, through blow-ups or passive-aggression, people may not even notice the degree of control they have.
Thank you for this post, Tim! I especially liked this insight:
I specifically want to give people a citation regarding the importance of mentorship in software companies. Andrew Begel and Beth Simon’s “Novice Professionals: Recent Graduates in a First Software Engineering Job” says that it’s implicit information that new CS graduates lack, not coding skills, and that formal onramp processes help massively — not to mention the cognitive diversity implications. There’s a preview online of the book it was collected into (Making Software, edited by Andy Oram and Greg Wilson) that has some stuff you might find useful: Reflecting on Pedagogy, Implications for Chance, and Misconceptions That Hinder Learning.
Many people also do better with clear expectations regarding days off and vacations. Netflix famously tells us in its culture slidedeck that individual works choose their work hours and days; the Valve employee handbook & Netflix culture slides circulate as though they were the blueprints for utopia. The other night I was chatting with a few friends, and one of them — on his own! — specifically brought up “flexible” time off as a problem. He said that he wished he could give more talks at conferences, but because there’s no policy about how much time he can take off, each time he’s invited to give a talk, he asks himself how mad he wants to make his boss. And so he doesn’t give talks. And not taking enough time off leads to burnout and suboptimal work.
Also, I can’t quite articulate it, but there’s some relationship here with Shanley Kane’s company culture critique and with how libertarians think of governments. Programs that address systemic problems are dismissed as waste. This frustrates me.
I think this is right on target. I work in a big and fairly structure-less organisation (= 2 levels: top management + everyone else, at least on paper, and everyone is supposed to have a say in decision-making). The internal politics can be quite vicious. I happen to be in the part which is considered “second class”, and is therefore largely excluded from decisions (though it has the advantage of also being less affected by the worst of politics). This is also the part that has the highest concentration of women, people with disabilities, people with kids, etc. I am a woman, with an invisible disability. There is no overt discrimination or prejudice. However, there have been any number of decisions made that are best for the “department”, but not best for people with physical impairments, or parents/carers, etc. I fought some of those, but it is an uphill battle.
At the same time, I ended up a project manager/decision maker for a team of people. While we would like to have consensus every time, in practice it isn’t always reachable, or else takes too long. I happen to hold strong opinions on many topics. I never wanted to be a manager, but one day I realised that I am running group meetings, taking many decisions and otherwise managing the project, and other people in the group are just letting me get on with it. I try my best to not steamroll people and to make sure everyone is heard, but I know I sometimes get my way just because I am willing to push. I can easily see how people who are less confident, or less privileged, would be at a disadvantage in this setup.
At some point this became “official”, in the sense that I started calling myself a project manager. But I think this has helped make it better, in some respects. Once I accepted that I have power, I also became more conscious of how I use it and how it affects the group
Pressed “enter” too early. I am all too aware that even though me becoming a manager was not intentional, it was probably a combination of privilege and particular personality features that made it happen, rather than “merit” or “experience”. In fact, I would say that there is at least one person in the group who is more experienced than I am, but just less outgoing, and more than once I wished I was better in listening to them. So the point is, such things can happen even in teams that are supposed to be equals, and it does not necessarily lead to best outcomes.
I had a similar realization a while back, and I think this is really important. It’s not how much or how little structure you have, necessarily, but how well your organization recognizes the power dynamics in play.
If giving people titles accomplishes that, do it. If documenting how decisions are made does, the same. Transparency and accountability comes from building a system where all participants can understand how things are done, and why.
I can attest that this is 100% the case in the animal rights movement (which is 80% women, and dominated by men). There is a growing movement that opposes the professionalized organizations like PETA and Mercy for Animals…the shift to the more “grassroots” “democratic” form, however, simply leads to more ingroup/outgroup maintenance and miniature dictatorships…and often mass trolling and harassment to protect those boundaries.
Structurelessness is likely not possible in social movements…I finally had enough of it and perform my activism solo through blogging, publishing in academic journals, and teaching….which is probably more than I would accomplish in a group setting anyway.
I’ve worked in both high structure and small-structure environments by now. My current company isn’t a completely flat one, we do have overt structure; just not a lot of it. Very roughly:
2. Department Heads
3. Everyone else
We’re still pretty small (~20 people) so we don’t have room for much more structure. But, we don’t have a lot and even within our departments there are unsanctioned divisions forming. It takes some social perception to notice they’re there, and even more social skill to work effectively within that kind of environment.
It’s tricky for me since I work in a title where I spend a lot on capital expenses, which requires justifying purchases. That requires buy-in, and noticing that it isn’t my department-head who makes those decisions but the CEO, and my Department Head needs to be deeply involved in the process anyway.
One fracture line I’ve noticed is work-styles. People who work best hanging out at each other’s desks (or in G+ Hangout) are their own group, distinct from those who work best through IM and group-chat. When this fracture line crosses a functional group, you get factions.
In an intensely social environment like this personal presentation matters quite a bit. You need to look like a team player. If that means repressing things you’d rather not repress, then so be it. Presenting as something as distinctly othering as trans is a bold step, and not one taken lightly. Having more explicit hierarchy means the social is taken out of the internal politics, and personal presentation (authenticity?) becomes less perilous.
Lovely article and critique.
I guess it should be clear, if only in hindsight, that “structurelessness” can’t avoid having a structure of some kind, and that structure has implications as you rightly critique above.
But surely there must be a way to structure without hierarchy. I work for a worker-owned cooperative that is doing well in some areas, but not very well regarding diversity, probably due to many of the points you mention.
It depends in part on what you mean by hierarchy. Some tasks require a person to be in charge of doing things, and it’s often useful to appoint committee chairs, even if they rotate. It’s important to work to keep the person who is “in charge” from misusing that power (which is actually very hard, even when everyone involved wants that to be the case), and to provide opportunities for everyone to take on the level of responsibility and commitment they’re comfortable with.
This is something I spend a lot of time practicing, and I still don’t think I’m very good at it yet. It’s just so different than what our surrounding cultures create and expect.