It seems clear to me that most people (even the most brilliant engineers apparently) still need something as a reference, they can't just be let on their own.

But that reference doesn't need to be a single manager. We could start to think of the team as the manager. The team can perfectly give feedback, handle friction, distribute boring work and so on. We already have plenty of stupid daily standups, team meetings, weekly all-hnds meetings and so on, so it wouldn't really be hard to have a team daily standup where problems can be brought up to everyone and solved cooperatively.

You can lead without managing (much) and you can manage without leading at all. The latter always seem to suck all of the joy out of my work, to the point where I dread coming in.

The problem is they should have fired all of the managers but kept all the leadership. It takes some soft skills to know the difference.

I guess one way to look at is management as an activity - at some scale you have enough of it it makes sense to have someone(s) specialize in it, rather than distribute amongst the group. But once it is a specialized role, it's harder for the person in that role to do other things. So if there are too many of them, or their time is not directed the right way, it's a net negative.

It’s advertising for one set of priorities, which introduces friction for others. Friction is amoral. Leadership can be immoral (in which case, removing leadership can be a good thing).

But I think I’m the specific case I illustrated (management without leadership) few of the qualities you allude to will be much in evidence. Managing for the sake of managing is, well, bureaucracy, isn’t it?

The Morning Star example (which I read of just now) seems to have the leaderless thing as a cultural backdrop, and new employees receive training in it (by doing). OTOH, you can't just do away with managers in a structured organization and expect it to work, the culture is not there.

You are confusing skill with power.

Hierarchies exist to wield power and exert control over employees in order to extract financial value from them.

There's plenty of people who own businesses, buildings or land and have terrible "soft skills".

Totally agree. That's probably another big mistake of the google experience.You can't just change completely the organizazion and expect everything to work fine the next day. Change should be done gradually, ideally I would gradually assign responsibilities to the team, while leaving a manager there to supervise and slowly fade out.

Exactly the kind of managerial activity devs often want to get away from.

Do you have a link to more information about this? I only heard about this recently.

Neither regular management is. There's plenty of stories of failed management. At least if it's a whole team doing the management, toxic behavior by one individual can be controlled. If the toxic individual gets to be manager instead, then there's nothing to stop him.

The VSM theory poses that, for any complex system to become sustainable, it needs to adapt to changes in the environment that would tear it apart. Enduring organizations are structured as a loose fractal hierarchy, where a core "policy-making" subsystem decides what is the defining essence of the organization. Other types of subsystems would distribute the signals and process, them to get work done towards what constitute 'progress' and 'change' in the system.

This core sends regulatory signals to the other subsystems to react and adapt against environment external signals, to keep the whole system alive and working efficiently. The strength and variety of the external and internal signals between subsystems could be measurable, defining a model to predict what complexity must possess the organization to deal with its environment and what signal strengths may break it. Thus the whole organization may change greatly and evolve as reaction to changes in the environment, but the core subsystem will decide what parts of the organization must remain constant to keep its essence intact.

According to this model, a leaderless organization would not need to have a single person in a leadership role, but would still need to have this core to define its essential characteristics; otherwise it couldn't work as a coherent organization working towards well-defined goals, but a loose movement pulling in all directions at once.

Edit: elucidate the main points.

[1] https://en.wikipedia.org/wiki/Viable_system_theory

[2] https://en.wikipedia.org/wiki/Viable_system_model

Leaders and managers are different things. You can have a team where the leader and the manager are different people. You can have a team with more than one leader, where they each lead in a particular competency, or on a particular project, or on a particular day because everybody else is in a bad mood. Properly functional teams tend to self select their own leaders, based on how they view their competency (both technically and socially). Every team that works well will have one or more people that the other team members trust to make decisions, provide feedback, guidance and advice.

A leader doesn’t need to have a job title. The behaviour of the team members determine who the leader is, it’s not a role management can assign to somebody. If you’re lucky, the leader of a team and it’s manager will be the same person, because the leaders will usually have more clout than the managers. But that’s often not the case.

https://theanarchistlibrary.org/library/cathy-levine-the-tyr...

Raw, seething resentment that goes nowhere.

Leaders are a short circuit around all that by just telling people what to do, no arguments. If your organisation is small enough that everyone can be brilliant and busy, great, good luck, might work. But even then technical people often appreciate some leadership to reduce the mental load and let them focus on how rather than what and why.

I doubt there are leaderless organisations. There are organisations that purposefully hide their leadership and those that do not. And yes, big management layers are bad. The managers have the same problems as everyone else - humans are limited. If there are lots of managers they can't all be above average and then the basic human problems they were meant to solve come back.

It's true that some people are smarter than others in certain tasks, so it's reasonable to give them more power in making decisions. But it's not reasonable to give them ALL the power. Because no human is smarter than a collective mind of others.

This is not a criticism of you personally, but rather of the internal consistency of your argument.

That's what I suspect as well. If you don't specify who the leaders are, some will emerge anyway. Perhaps there's some advantages there, but one obvious disadvantage is that the org structure is now undocumented.

I think this is missing the bigger picture. I wrote this a few days ago:

“I think what we’re seeing has more to do with unaccountable invisible architectures (law/governance systems, IP and money) that obscure production processes, than that it’s anything to do with human nature or social media, or us being ‘hairless monkeys’.”

Arthur Brock describes it better:

“The story I want to be able to tell my (as of yet non-existent) grandchildren in 25 years.

Once upon a time, our planet was dying. People had organized themselves into patterns with funny names like “limited liability corporations,” “federal governments,” “non-profit organizations,” and “institutions of higher education.” We were unwittingly eating the world and all of its resources to perpetuate the survival of and expand the growth of these social organisms."

[...]

“Since information sources and sharing protocols weren’t integrated, we couldn’t clearly see flow patterns on large scales. The fragmentation made it easy for people to only pay attention to what they wanted to hear, or remain ignorant and unseeing of things they didn’t search out. In fact, most information was received in the form of dramatic entertainment called “news” that was filtered and scripted by the organizations delivering it according to criteria which supported their own influence and growth.

Since currency was only thought of as money, and wealth as accumulation, when people organized themselves into collective patterns, those social organisms were preoccupied with their own growth and acquisition of artificially scarce money. They were structured as cancers — using their resources only for their own perpetuation and growth — killing the planet that is their host.

Do you remember hearing stories about “the Internet?” — the early network for computer communication that started to emerge in the late 1900s… It was a place of great experimentation and chaos. We hadn’t yet discovered how to make all protocols interact with each other, or have shared models of meaning and world-views to operate from. But information started flowing quickly around the world, accelerating change in surprising and exciting ways.“

[...]

"In the early 2000s, there was a breakthrough. A small group of people translated the organizational patterns by which living systems operate into the computing and communications. Soon they invented the field of current-see design to consciously design the DNA of social organisms in accord with the sacred patterns which nourish life. People began to organize themselves into a variety of patterns that look more like ways you see us working together today.

Fractal computing receptors and self-describing protocols enabled information to flow in new distributed patterns of coherence. We built an easy compository for sharing meaning and models for sense-making. Our collaboration tools started evolving much more rapidly with the ability to install new shared protocols in an instant. There was no longer centralized control of information in every group and organization. There was an explosion of new life and new patterns of collective intelligence

In many ways this explosion of social organisms and communication protocols was even more chaotic than the Internet it began replacing. Really amazing patterns started emerging which quickly outperformed the corporations of the past because they were better at having people feel fulfilled, trusted, known, heard and involved… better at sharing resources that were previously hoarded… better at creating outrageous productivity and creative results.

Once we learned how to organize ourselves in these ways, we were able to reverse the damage to the planet and our communities very quickly. Problems that had been accumulating for hundreds of years, we reversed in only a handful years because we could coordinate on a planetary scale in ways that nourished the bodies, spirits, and hearts of everyone involved. We learned the joy of dancing together for mutual benefit, built the tools for us to see the state of our collective dance, and grew to feel the vibrancy of those rhythms together.“

Source: https://medium.com/metacurrency-project/the-metacurrency-myt...

Management concerns the organization of people, tasks, information, incentives and all such things. Bad management is a common plague upon many an engineer, and probably for good reason.

Bad leadership is more commonly felt through its absence. Managers could and maybe should be leaders, but often are not. Leading, of course, can also be done by someone positioned elsewhere in the formal hierarchy. Nowadays, the formal hierarchy (aka the org chart) of a firm may not necessarily reflect the informal structure of a well-functioning team.

This then leads to the question: What is a leader?

Surprisingly, the whole concept is rather elusive. If we do not mean management, what precisely does a leader do? The answer depends a lot on whom you ask. For example, economists have traditionally very little use for a leader. Coordination, information and all things amenable to traditional econ models, are things that a manager does. "Leading by example" is one of the few concepts where a leader has any use in economics of organization.

Other literatures, by contrast, have an overwhelming amount of roles and theories about leadership, to the degree that one might question if the entire concept is well-defined at all.

If you are interested in a shameless plug, we have analyzed the entire corpus of Harvard Business Review - a publication that concerns itself with leadership - to find a semantic definition of leadership "from the horse's mouth", so to speak. Results were fairly surprising - we see changes in roles and contexts of a leader over time: increasing reach and importance of the concept, but also sharper definitions of contexts of its roles. All these things mean that leadership is elusive. It means many things in different context (presentation on youtube under v=z2uceIRQA6g - again sorry for the plug).

In my current view, the most important role for a leader, next to determining the strategic direction, is sense-making (or more precisely, sense-giving). Whenever the expectation of the team differs from reality, a leader's job is to provide the frameworks and consensus for a united narrative, one that allows the team to react and move forward.

It's a bit of a guidebook on how to build this sort of organisation.

There is a sort of standard ("right-wing", "evolutionary") explanation that argues that social hierarchies are of biological nature, similar to hierarchies in some species, they are related to sexual selection and simply somebody has to be an "alpha". Ultimately, I rejected this explanation because I think the evidence for it is contested. Some humans are actively willing to submit, it's not just the threat of violence that causes them to do so, and the deference to authority is not always sexual in humans. It seems that not all hierarchies are simply a struggle for power, although arguably in large-scale human groups, it often becomes the case.

Currently, my favorite explanation of authority as a natural human value comes from the idea that the authorities represent "institutional memory" of the particular human group. So for example, the first authority you meet, your parents, exists to give you knowledge and experience of their life, so you wouldn't learn from zero as a child. Similar goes for the social group as a whole, the authorities in the group (typically the elders) are the people who lean conservative, and make sure that lessons of the past are remembered by the group.

Given this explanation for the role of social hierarchies and authority, I am not sure what replaces this "institutional memory" in "leaderless" organizations. If nothing replaces it, it must become inefficient eventually by repeating past mistakes. Or the leaderlessness might be short-lived and the authority which holds the institutional memory will appear.

However, I agree that in modern large companies, the management is often turned over much more frequently than some employees. This, assuming the actual role of hierarchy in human groups is to gatekeep institutional memory, would explain why some conclude that management is useless. But it is not clear whether going further and entirely removing it is going to be better.

    > But it is not clear whether going further
    > and entirely removing it is going to be better.

The orchestras show that there is an alternative to middle-management. We don't really have a word for this, but we could call it Elders.

A key difference between Elders and Middle Managers is the role of skills. Elders must be at least competent at domain-relevant skills to be relevant, and their respect grows with skill mastery. Whereas, in a MM culture, having skills excludes you from influence. You are either a career manager (entirely without domain skills), or you convert. In the process of conversion, you will have social pressure applied not to get your hands dirty.

1) Tree structure minimalizes communication complexity. It minimalizes both latency and total cost of communication/decision process, which has usefulness for both business and military. Efficient decision process allows quick and decisive power control, which often has superlinear advantages. The other side of coin is that longer and more thorough decision process could lead to better decisions, which may have long-term advantages.

2) Social hierarchies are not cause of struggle for power, but they ritualize and tame it. With that, they limit costs associated with struggle for power in a similar way how ritualized animal fights limits wounds, which would happen in real fights. In social hierarchy each actor acknowledges its level of power and allows surrender without destruction. Ideally, the power struggle within rules of social hierarchy also makes it more meritocratic (e.g., people with better business skills get on higher level in bussines hierchies).

But your institutional memory makes also sense.

I personally struggle with it, as i have strong emotional aversion to both submit or claim power over others, but it is not hard to see advantages (and also disadvantages) of social hierarchies on rational level.

BTW, looking on your nick, aren't you by chance JS1 from abclinuxu?

Ad 1, when there is no central "brain", and decisions are more distributed, a single hierarchy often hinders communication. (There are for example studies that show that the biggest communication barrier in SW teams is not a physical distance, but a managerial one.) It seems when people work in a distributed way they tend to communicate in many different ad hoc groups, as needed.

Ad 2, there is indeed some evidence that more centralized societies are less violent. But it's not clear whether this is because of the hierarchy, or because of the concentration of power. On the other hand, it seems that the societies with multiple different hierarchies like democracies are in fact often less violent than more authoritarian societies. It seems to me if this hypothesis was true, we would see much more formalized ways how the hierarchies are changed, but I think we don't, rebellions are often a very messy business. Perhaps it can be accepted as an explanation why there are trade wars instead of wars, but as the answer to the original question, it IMHO still leaves something to be desired.

In any case, I think there are many good explanations for hierarchies.

> aren't you by chance JS1 from abclinuxu

Excellent guess! ;-)

If nature is unjust, change nature![0] So, um. Let's keep doing it until we can do that.

0: https://laboriacuboniks.net/manifesto/xenofeminism-a-politic...

In childhood, parents play both roles: they punish you for disobeying rules, and they care about you and teach you all kinds of stuff. They exceed you in both strength and skills. But later, these two things separate: there are people you fear but don't admire, and people you admire but don't fear; people whose attention you want to avoid, and people whose attention you dream about.

In the workplace, there are people whom you admire for their skills, and then there are people who are simply given power over you without doing anything admirable. And there are also a few people who are given the power and who use it in an admirable way. I guess the idea is that we should get rid of people who are given power but do not have natural respect. The people who have both the power and respect... dunno, maybe their power should become informal, based on the consent of the governed (i.e. you follow their advice, because their advice consistently turns out to be good). Thus the bearers of "institutional memory" would stay, and the useless ones would go. Some teams would have one authority person, other teams would have multiple authorities with different responsibilities based on their skills.

First, when there is no overt structure, it is replaced by politics.

Second, you can see the result: the difficulty (one might even say inability) to produce new product internally; instead new games come from acquisition: the incoming team might manage a sequel but after that the team dissolves and can't recruit new members for a second sequel.

I do think the dominant system would benefit from a rethink of the "leader/follower" mentality: everyone is needed (else why are they there) and they do different parts. Cleaning the office is important else the whole thing stops working; deciding go/no go on a new product is also important else the cleaners lose their job.

I am not sure how to fix this. The fact that we conceive of hives of bees as having a "queen" (who is in fact the weakest member of the entire colony with precisely zero decision making ability) shows how deeply entrenched this model is.

I seem to remember a quote about Gabe Newell, and how he actually was very much in charge, despite there being no managers.

I have no real feedback on whether or not this structure is effective. I was a manager for a long time, and hated it. I would have much rather been an engineer; but that is probably more a reflection on me, than the practice of Management.

Well, the thing is that commercial companies - which are owned by a party other than their employees [1] - "need" agents of owners' control. Whether they "lead" or not is not the main point. The point is that they answer directly to the owners, and through this they have power and control. Without them, the workers would "do what they want" - which may be a good thing or a bad thing, but it won't be what the owners decide.

So, in that sense, "we" "need" "leaders", yes.

[1] - If workers also own company stock that doesn't changed the structural nature of the ownership relations.

Also, is there an example of a factually leaderless organization, past or present? (Anonymous is not a transparent organization and is not an informative example.)

First Example:

1. In many world states, condominiums / apartment buildings have common building affairs run by a residents' association. While these typically have executive functions assigned to only a small number of people, it's very often that membership in that committee changes frequently, with a a decent fraction of the residents getting to serve on it occasionally, and this executive is never (well, in my experience) perceived as "leading" the residents. These are stable organizations which last for decades doing concrete practical work - with a budget, engagement of contractors and occasional employees, interaction with municipal or state authorities etc.

Second Example:

The CNT, Confedercion Nacional del Trabajo. This is a confederation of Anarchist Labor Union, whose hey-day was during the 1920s and 1930s, but unfortunately got decimated by Francist Fascism. It survived underground and today is active, though not remotely as large as those days. Its structure is a bit involved, but the main ideas relevant to your question is that regional or national bodies cannot make decisions for the local bodies, nor the national for the regional, and their capacities are secretarial and administrative (which of course still carries a lot of power); and even that rotates - regional secretariats assume the role of national secretariats in some kind of round-robin fashion. This in itself does not mean that there's no leadership, but the structure, together with the anti-Authoritarian ethos of Anarchism is a political philosophy and labor-unionism as what's actually practiced contribute to an anti-centralist spirit which (hopefully still today) permeates the organization. It's a "do it yourself together" kind of an attitude rather than "wise people trusted by many have laid out the path for us". Caveat: I only visited with one of the regional confederations once in 2006 and read about them so I can't "guarantee leaderlessness".

See also: https://en.wikipedia.org/wiki/Confederaci%C3%B3n_Nacional_de...

I personally favor ad-hoc & on-demand, transient, hiearchical structures, optionally with robust leadership, as the norm for future organizations. Information systems help in making such orders practically viable. I should add that I don't share your disdain for "wise people". It is true that the trust element, to say nothing of actual wisdom provided, are potential bug sources. But some problems do require the top-down approach, and the period preceding an emergent order (consensus) via bottom-up ("do it yourself") approach is non-optimal for certain contingencies.

In general, I think it pays to be flexible and practical when it comes to organizational structure. You mentioned the "ethos" of CNT, which is imho the fundamental driver of organizational health. No social architecture can save a society from its own nature or character defects.

An org without a leader is rudderless. Can't think of a real, successful example either.

You would need to remove the incentives of selfish actions and backstabbing to see any of the efficiencies with a formal hierarchy. Otherwise each branch becomes a metaphorical tumor.

Second, the article shifts back and forth between leadership and management. I just posted to my blog about the difference, with a picture of an apple and an orange. While they overlap, they aren't the same. By no means am I suggesting one is better, only that big enough jobs require both.

Managers use skills like analysis, planning, record-keeping, and organization on readily measurable things, facts, figures, observable deliverables, and timelines. A successful manager creates compliance. Well-managed teams get the job done.

Leaders use skills like listening, vision, support, charisma, and credibility on intangibles, emotions, beliefs, stories, images, and symbols. A successful leader inspires. Well-led teams love their work.

Managers work in the system, improving it to meet the values and goals driving it, making it more efficient.

Leaders work on systems, adjusting them, leading them to align with values and goals they miss.

Managers use incentives—carrots and sticks—based on the outcome desired.

Leaders help others do what they already want to but haven't figured out how based on attracting and enlisting people whose goals and values align with the team and role.

Managers react to the values and goals set by leaders.

Leaders are proactive, figuring out what values and goals to prioritize.

A manager's skills transfer between companies and industries. A manager can succeed without caring about what the team does. Managers get people acting. An effective manager gives people the satisfaction of a job well done.

A leader must do what he or she leads others to. A leader believes in the cause. He or she can rarely succeed at leading what he or she doesn't care about, but the passion a leader feels for a cause makes the work rewarding and inspiring. An effective leader makes that inspiration infectious.

Well-managed, poorly-led teams and people may feel obliged to work hard at something they don't care about. They may work for material rewards but not value them. They measure their success by how much they do. Well-led, poorly-managed teams and people may feel inspired and passionate but not know what to do. They may feel frustrated with a present inferior to a future vision they aren't moving toward. They may lack a meaningful measure of success.

Well-led, well-managed teams and people embrace the challenges and use the problems of today to motivate and propel them to a future they love by creating effective strategies and plans and acting effectively. They measure their success by how much they reach their potential.

Here's my blog post https://joshuaspodek.com/management-versus-leadership, but it's basically the above.

Only if we don't study our history, and have a very low bar. Ignoring, for a second, the question of leadership-vs-leaderlessness (which is important of course), let's remember that:

1. While Lincoln led the US to a civil war which abolished formal slavery, his own opinion on black people was quite horrid:

> I am not, nor ever have been, in favor of bringing about in any way the social and political equality of the white and black races … I am not nor ever have been in favor of making voters or jurors of Negroes, nor of qualifying them to hold office, nor to intermarry with white people; and I will say in addition to this that there is a physical difference between the white and black races which I believe will forever forbid the two races from living together on terms of social and political equality.

and indeed, the US took very little action to rectify the historical injustice.

2. Churchill was a staunch imperialist in terms of his public career, and is hated by the peoples of the middle east for his involvement in the continued occupation and the continued support and sometimes setup of oppressive puppet governments in British areas of control. At home, he was a tool of the rich in terms of state spending on welfare and other issues. In fact, even when it comes to Nazi Germany, Churchill supported non-intervention in Spain, where a significant British intervention against the Fascists would have had significant effect on later events.

3. Mandela is perhaps the most worthy of respect, but it must be remembered that the ANC essentially gave up on ending material, economic white supremacy in South Africa to end formal, legal apartheid. Whites, to this day, control almost all of the land and wealth. And the worst thing about it is that it is now a black government which enforces local white and foreign-multinational property rights against the landless and the poor.

All that to say that the success of the South African liberation struggle and the success of this summer's BLM protests point to the need for and efficacy of organizations that can effectively execute with minimal leadership as the situation demands.