It's always puzzled me that, in armies, officers form a separate hierarchical ladder from the NCOs and enlisted soldiers.

Armies could have a single hierarchy, top to bottom, as in the simplified diagram below on the left. Instead, all armies have two distinct ladders, with one strictly above the other, as on the right.  (Reminds me of those wacky non-standard integers.)



The usual answers are obvious but irrelevant: Yes, some people shoot straight to a position high on the ladder. You could do that with either model. Yes, even when those lower down on the ladder have more experience and wisdom, it can make practical sense to have a hierarchy. Yes, the higher someone is, the higher the level of the decisions they make. You could likewise do these on a one-ladder model.

It's said that officers "decide," while non-officers "just carry out  orders"; or that officers choose strategy, and non-officers do tactics. But everyone makes decisions, on their own level. A private makes decisions for himself, a corporal for three soldiers, and a colonel for a thousand, each one in the context of their orders from above. One soldier's strategy is his superior's tactics. And the distinction is not based on command: New army doctors automatically become officers, even if they don't command anyone. Doctors are non-combatants, but fighter pilots are combatants par excellence, don't command anyone, and are all officers.

These answers don't explain why there need to be two ladders. I asked at Quora without a convincing answer. Historically, the distinction was based on social classes, but that doesn't explain why every army follows this arrangement, including those in very different societies.

Similarly: What's a corporate executive? (I'm talking about large companies here; small companies and startups are different.) I understand that there is a management hierarchy, but why the arbitrary distinction between a senior manager and a junior executive? Aren't those just two rungs on the ladder? In corporate-speak, an executive is called a "decision maker." What a strange term! Isn't a manager or even a lowly "individual contributor" also a decision maker -- at the scope that their own managers allow? (I should add that the two-ladder system is not as developed in business as it is in the army  or in medicine. There is no career ladder for non-execs that extends arbitrarily high, though always below the execs.)

Not all professions work that way. Actuaries have ten levels, based on passing a sequence of exams. And though some areas of engineering distinguish an engineer from a technician, software engineering has no such dichotomy: Some software engineers make more money, and some make broader decisions or manage others, but there is no two-way split.

In medicine, on the other hand, there is a clear distinction between doctors and nurses. There are different status levels among doctors and among nurses, but a PhD in nursing stands on the other side of a clear border from a beginning MD. Similarly with lawyers and paralegals. These dichotomies stem from licensing restrictions, which in turn are descended from medieval guild practices. But why does it have to be this way? Why not just rank medical personnel, or legal personnel, in a single continuum from practical nurse through rockstar brain surgeon. (Is that a title?). There would still be the understanding that some people will never climb beyond a certain point, while others can jump straight to a higher rung.

The answer lies in LessWrong's concept of "agentiness": Making "choices so as to maximize the fulfillment of explicit desires, given explicit beliefs." Less abstractly, it is sometimes described as "reliability and responsibility." Agenty types get to be called "Player Characters" or heroes. ("Agenty" and "agentiness" are made-up words, and the standard terminology is "agent" and "agency.")  I think "agenty" was made up to point out that while all humans are agents to some extent, some do it far better than others.)

In the organizational context, officers and executives are meant to be agenty, while enlisted/NCO and non-executives are not. The officers and executives plan towards achieving goals, while everyone else executes defined tasks. The officers and executives make high-variance decisions, with high risks and high returns, while everyone else has the job of just doing their job consistently and not messing up.

Is agentiness a natural kind, a cluster in thingspace, a joint-carving concept? Might agentiness just be a mix of features that occur to varying degrees in various contexts?

We might say that agentiness is a continuum: Everyone has some, but some people have more than others. Lower-downs sometimes have goals, and higher-ups often act like cogs. Moreover, the agentiness of officers and executives is strictly in the context of their superiors' goals: They may be agenty, but not for their individual goals. It would be more accurate to say that in their roles they are meant to be agenty, on behalf of the organization.

Some people are non-agenty in some of their social roles and agenty in others. For example, I know workers who readily admit to being lowly cogs in a machine, but who have tremendous achievements in setting up and leading non-profits outside work hours. Some hard-driving workaholics are milquetoasts at home. Some caring, wise, foresightful parents are limp rags at work.

But agentiness is a real concept, at least so far as the officers and executives go. Their roles are implicitly defined by agentiness. Armies and corporations decide which people have it (or at least are meant to). These organizations agree with LessWrong that agentiness is a natural kind.


New Comment
144 comments, sorted by Click to highlight new comments since: Today at 8:59 AM
Some comments are truncated due to high volume. (⌘F to expand all)Change truncation settings

Historically, the distinction was based on social classes, but that doesn't explain why every army follows this arrangement, including those in very different societies.

You may also notice that the dress uniforms of pretty much every military are similar as well (especially compared to historical differences.) Also compare Emperor Hirohito to King Edward VII. People in most countries have also adopted western standards of business dress and western corporate hierarchies as well.

You could view this as a form of cargo-cult copying. Non-European countries wanted to emulate the strength of western societies, but it can be hard to know what differences actually need to be copied and which are arbitrary or irrelevant.

Yes. But war is a Darwinian process, and if one army figured out a better way to do things, they would win. That's not to say that a one-ladder system would make the difference, however.

But war is a Darwinian process, and if one army figured out a better way to do things, they would win.

Er, all else being equal, maybe. There have been only a handful of total wars fought between states. There is zero reason to think armed forces have been optimized to that extent.

Well... it is generally agreed, I think, that the Wehrmacht was, man for man and gun for gun, a better fighting force than any of the Allied armies that defeated it.
OK ... but they had the same two-ladder system, so that is not an example of an alternative to the two-ladder system.
The point is that most wars are won or lost on non-military grounds: economics, politics, technology, strategy, and even just the size of the army. For the one component of military organization to be naturally-selected, you would need many more wars and generations than have in fact existed.
Yes, but my point is that the organisation of even victorious armies is not necessarily optimal, or better than that of the army that it defeated. There hasn't been enough selection (and arguably, armies are insufficiently accurate replicators anyway) to make Darwinian arguments have the power you ascribe to them.
Chance plays a huge role in natural selection.
"But war is a Darwinian process" No, it's not. If "Darwinian" means "pertaining to or resembling biological evolution", which it generally is understood as, then there are certain characteristics of biological evolution that one can reasonably expect to be present for the term to be proper: a genotype governing the development of organisms and consisting of genes (units of heritability), a phenotype physically expressing a particular genotype, a copying process that is largely faithful in transmitting genes but occasionally mutates them, a method of combining genes from two individuals, and a selection process based on phenotype. What is the "organism" in the case of war? The genotype? The phenotype? The genes? The copying process? Etc. Ask Billy Beane whether the mere existence of competition means perfect efficiency. Warfare involves signaling, brinkmanship, collusion, negotiation, bluffing, path dependence, local equilibriums, information asymmetry, superstition, stochastic processes with samples sizes completely inadequate to estimate underlying parameters, and more. There's no reason to think it's efficient.

In the organizational context, officers and executives are meant to be agenty, while enlisted/NCO and non-executives are not.

Maybe this is the theory, but in practice it doesn't turn out this way at all. It's been my observation that officers who mess up do not have to face consequences as severe as soldiers do. In fact there have been investigations into military operations that have confirmed this.

Similarly, in the corporate hierarchy, senior executives can run their companies into the ground and still get bonuses (as happened in the 2008 financial crisis) whereas low-level employees would be faced with termination of employment.

However, it is true that in general soldiers are not taken to be as responsible for war crimes as commanding officers are. However, this again comes down to scope. A soldier can only ever order the murder of a handful of men; a general can order the slaughter of a million.

I have to agree with you that it only makes sense to have a single hierarchy.

Officers are held accountable to a lower degree by their own army. They are held accountable to a higher degree by other armies. That is completely consistent with agency; they are the corporeal manifestation of the abstract concept of "the army". When someone says "The US Army made the decision to launch an attack", they mean "The officers of the US Army made the decision to launch an attack". If officers are punished by the army, the army is punishing itself, because the officers are, as far as the decision-making is considered, the army. This, not merely scope, is why officers are treated differently as far as war crimes are concerned, and why criminal prosecutions tend to focus on CEOs, not on all the low-level employees that actually put the fraud into practice. Also, organizations tend to give people higher up in the hierarchy more of a vested interest in the organization. This means they have more at stake, which in turn means that when they mess up, they end up with more than a lower-level employee in an absolute sense, but less in a relative sense. For instance, an executive might be promised a bonus of up to $10 million. The executive might mess up and still get $1 million, and that looks like they're being rewarded for screwing up because they're getting, in an absolute sense, more money than an ordinary employee who did their job perfectly, but relatively speaking, the executive just lost 90% of their bonus. Of course, "Don't complain about my $1 million bonus; I lost out on $9 million that I could have gotten" doesn't tend to go over very well.
Militaries have internal conflicts and punish their senior officers. This goes not just for the army but for other branches of the military (not sure why you're putting so much emphasis on armies). The bonuses for the 2008 financial crisis were in many cases higher than the executives had recieved in prior years.
I quite agree. In my post I am simply wondering why the dual hierarchies exist, not whether the systems work as they should.
Doesn't that follow from the agenty/non-agenty distinction? An agenty actor is understood to make choices where there is no clear right answer. It makes sense that mistakes within that scope would be punished much less severely; it's hard to formally punish someone if you can't point to a strictly superior decision they should have made but didn't. Especially considering that even if you can think of a better way to handle that situation, you still have to not only show that they had enough information at the time to know that that decision would have been better ("hindsight is 20/20"), but also that such an insight would have been strictly within the requirements of their duties (rather than it requiring an abnormally high degree of intelligence/foresight/clarity/etc.). Meanwhile, a non-agenty actor is merely expected to carry out a clear set of actions. If a non-agenty actor makes a mistake, it is easy to point to the exact action they should have taken instead. When a cog in the machine doesn't work right, it's simple to point it out and punish it. Therefore it makes a lot of sense that they get harsher punishments, because their job is supposed to be easier. Anyone can imagine a "perfect" non-agenty actor doing their job, as a point of comparison, while imagining a perfect "agenty" actor requires that you be as good at performing that exact role, including all the relevant knowledge and intelligence, as such a perfect actor. Ultimately, it seems like observing that agenty actors suffer less severe punishments ought to support the notion that agentiness is at the least believed to be a cluster in thingspace. Of course, this will result in some unfair situations; "agenty" actors do sometimes get off the hook easy in situations where there was actually a very clear right decision and they chose wrong, while "non-agenty" actors will sometimes be held to an impossible standard when presented with situations where they have to make meaningful choices between unclear
That's not the point though; the point is that agenty actors are understood to be doing higher work as reflected in their position in the hierarchy, and thus are expected to be smarter than their lower counterparts, and be able to make better decisions. That expectation should logically carry higher responsibility and accountability with it, otherwise what distinguishes a soldier from an officer?
You're conflating responsibility/accountability with things that they don't naturally fall out of. And I think you know that last line was clearly B.S. (given that the entire original post was about something which is not identical to accountability - you should have known that the most reasonable answer to that question is "agentiness"). Considering their work higher, or considering them to be smarter, is alleged by the post to not be the entirety of the distinction between the hierarchies; after all, if the only difference was brains or status, then there would be no need for TWO distinct hierarchies. There is a continuous distribution of status and brains throughout both hierarchies (as opposed to a sharp distinction where even the lowest officer is significantly smarter or higher status than the highest soldier), so it seems reasonable to just link them together. One thing which might help to explain the difference is the concept of "agentiness", not linked to certain levels of difficulty of roles, but rather to the type of actions performed by those roles. If true, then the distinguishing feature between an officer and a soldier is that officers have to be prepared to solve new problems which they may not be familiar with, while soldiers are only expected to perform actions they have been trained on. For example, an officer may have the task of "deal with those machine gunners", while a soldier would be told "sweep and clear these houses". The officer has to creatively devise a solution to a new problem, while the soldier merely has to execute a known decision tree. Note that this has nothing to do with the difficulty of the problem. There may be an easy solution to the first problem, while the second may be complex and require very fast decision making on the local scale (in addition to the physical challenge). But given the full scope of the situation, it is easy to look at the officer and say "I think you would have been better off going further around and

In the military, I always thought that it was a method of ensuring obedience. Fear in the battleground is supposed to be one of the most intense fears experienced by humans. And getting people to go into the battlefield and act with even a semblance of competency requires a lot of brain programming. Thus, the fact that an enlisted man can never even aspire to be an officer (with a few exceptions) gives the officer an incredible source of authority; in the mind of the enlisted man, the officer is almost infinitely better than him. This is also the reason why bootcamp is so intense: programming obedience.

This might also explain the doctor/nurse duality: the institution desires that in a high-pressure situation, the nurse never disobeys the doctor. Though I'm not so sure whether this explains it fully in this situation.

Obedience can be demanded in a one-ladder system.
Yes, but there is a question of degree. The claim is that two-ladder systems with no allowed lower-ladder to higher-ladder transition effects a much greater amount of obedience than a comparable one-ladder system. In other words, you'd have to expend a lot more effort in training if you wanted a similar amount of obedience in a one-ladder system.

Your description of the dichotomy is interesting. But it doesn't explain why some professions have it and others don't. Is it merely historical accident for each separate profession? Or did agent-y armies actually out-fight non-agenty ones when those arose?

Also, I agree with one of the answers you got on Quora, which gives another reason for multiple chains of promotion/status. Along any one chain, going up a rung is a promotion most people want. But different chains are like different sub-professions: the skills and experience may not transfer from the top of one chain to the bottom of another, and people may simply not want to change their careers that way. A manager may be "higher status" than an engineer, and tell him what to do and be paid more, but the engineer may not want to become a manager and may not be a good manager if he does.

In an army the distinction is clearly purely historical or status-based (why else are MDs officers and engineers are not?) But for doctors vs. nurses, or lawyers vs. paralegals, I think it's part of the explanation; although a status differential is clearly present, it's a status difference between professions, not between groups of people.

Is it merely historical accident for each separate profession?

Just like the two ladders in militaries is a holdover from a more classist society, the doctor/nurse divide is at least partly a holdover from a past (more) sexist society. Even today about 90% of nurses are women. This might be interesting if someone had access past the paywall:

The answer from Quora stating that it had to do with different skill sets is completely ignorant of modern day military practices. For instance, Master Chiefs in the navy, especially once they get the specific qualification tailored to their department, don't stand watch and are completely managerial in nature. You will never see a Master chief do any sort of labor intensive job unless they are breaking the rules. This is even true in combat units where high ranking enlisted Navy SEALs generally default to desk jobs just like their officer counterparts. This is not always true, but it's most often true. This transfer takes place as you increase in rank and you'll start seeing the beginnings of it as low as E-5 in some branches, higher in others. The only true difference is that officer deployment rotations generally have them doing more than just their main job. For instance, a nuclear officer will also be an officer on non-nuclear ships for certain tours. So while a Master Chief will be doing the same job as a O-1 to O-3, he will always do it within the same department or same type of department - with exception to Washington jobs. While the overall titles are different the concept of the jobs are the same. For example: A Division Master Chief, deals with personnel and divisional duties while a Divisional Officer deals more with personnel and to a lesser extent with divisional duties, but still signs off on schedules. A Department Master Chief deals almost exclusively with personnel and has very little to do with maintenance or Departmental duties - their only real involvement in departmental duties is related to personnel use. A Division Head (usually an O-4 or O-5) deals heavily in Divisional duties and schedules. As far as actual "watches" go, or "shifts," the E-7 to O-3 can all have the same shifts with exception to a few, in some fields. Still, when it comes down to it, you have a situation where senior enlisted are basically doing the same job as a O-1 t
A priori, I'd say any enlisted/NCO would prefer being an officer (if they didn't have to work for it, etc.). There are very few people who actively prefer a subordinate role; everyone wants more money and power, at least in principle, though some want it more than others. Most managers are not executives. The executive/non-exec border is mostly a dividing line between some managers and others. "Individual contributors" are completely out of the running here, just as privates in the army are nowhere near being officers (in contrast to Chief Master Sergeants, about whom we might ask why they are not officers.) There are also a very few "individual contributors" who are execs, such as a tiny handful of top technical people in some tech companies. Yes. Now, why are these treated as different professions rather than as different levels of the same professions? Other "professions" have radically different tasks, prestige, and compensation for different levels.

There are very few people who actively prefer a subordinate role

That may be so, but there are a lot of people who would actively prefer NOT to take on serious responsibilities.

There may be, for that does not characterize most master sergeants, nor more most senior (non-exec) managers, most of whom would most certainly like to be execs. [edit] But I agree with you -- if we follow LessWrong's line of thought about agentiness, then yes, a lot of people prefer to avoid responsibility.
There are lots of things people might prefer if they didn't have to work for them. :) Because they're at the top of their career, so why should they give that up to be at the bottom of a new career? This, I have read, is what some of them would say. Of course, that begs the question of why there are two ladders, but given that there are, it seems a reasonable response.
OK, good points. To the extent that officer status correlates with agentiness, and that some people are not agenty, then some people would not want to be officers. I am not sure about the antecedents in the preceding sentence, however.
4Eliezer Yudkowsky9y
That strikes me as an awfully sheltered, possibly even Jewish thing to say. I've heard this is true in the Israeli army, elsewhere not so much.
While I suspect I know what you mean, I'm left somewhat agog here at the implication that "Jewish" is an extreme form of "sheltered."
Heh. That might be this "Jewish humor" I hear so much about.
Given that in my experience and by what I've heard, self-deprecation is a common element in Jewish humor, yes.
To the extent that the framework in this post is true, it is possible that some people are put off from the requirement to be agenty. As to whether Jews think that way less or more than others ... hmmm.... And as to whether I am sheltered from non-agentiness: I see a lot of non-agentiness, no shortage of that.
Most people would like to have a higher-status role. That doesn't mean it's not a separate profession requiring different skills. That's true. I didn't mean to conflate the two, but to offer engineers vs. their managers as another example of distinct subordinate and superior classes where it makes sense to have two hierarchies. Partly because they are in a large degree different professions requiring different skills, and not translating experience well.
Indeed, I've heard people make the argument that only having a single hierarchy is a common problem in the field of software engineering, since it forces programmers who want to be promoted to do a job they're not necessarily well-suited for, and which they might not like. Having separate "programming" and "management" promotion tracks was considered a clearly superior alternative.
Surely there are not two alternatives being compared here but three. (1) A single hierarchy with programmers at the bottom and managers at the top and fairly straightforward progression all the way up. (2) Two hierarchies, one for programmers, one for managers, with the managerial one viewed as above the programmers' one. (3) Two hierarchies, one for programmers, one for managers, truly parallel, with comparable levels of status and pay available on both. There is an obvious reason for programmers (and others) to prefer #3 to #1 because it allows people to carry on doing what they're good at and enjoy, but #2 (which is what one typically gets in practice, and I would strongly suspect it even when there are notionally parallel hierarchies) is not such a clear improvement.
What does "above" mean? Managers necessarily tell the programmers what to do a lot of the time; that's just part of managing people. For humans, this seems to be inescapably tied to higher social status. Is there a separate sense of "above" you refer to? Higher paychecks for managers are common. But then I might as well ask the more general question of why different job-tracks have different paychecks at similar levels of experience, seniority, and skill percentile. There are various reasons; one is that the top-level executives deciding on paycheck policy are usually promoted from the top ranks of the management track, rather than a technical one.
In principle, that two-hierarchy system, #3 above could allow for true parallelism: The most senior people on the technical ladder would have a boss, like anyone else, but this boss would be someone who is way up the management ladder. Also, in this hypothetical system, the top non-manager technical people would have similar compensation to top managers. In practice, the most senior people on the ladder simply merge into the management ladder, so that at that point there is no non-management technical ladder. So, #2 above is far more common.
Some companies have option #2, a technical ladder for non-managers, where the managerial ladder is strictly superior in terms of informal social status and compensation. A few very advanced companies may offer #3 or something like it. To the extent they do, we see another case of two ladders, but this time with a very obvious reason, much less a mystery than the officer/enlisted or exec/manager distinction.
Officers might not be so good at doing sergeant work, and doctors not so good at nurses' work, but no doctor would get in trouble with the law for doing a nurse's work, but vice versa, the nurse would. Likewise for officers and NCOs. In practice, that might be true, but this whole thread is asking why the official distinction exists. [edited for brevity]
I've seen many doctors who couldn't do a nurse's work. They could maybe fill in for an hour if no-one else was available, but they couldn't really replace one. Soldiers are a special case because junior officers lead their men into battle. So officers, at least young and fit ones, should be able to replace ordinary soldiers. My best answer has two parts. The first is the exaggeration of status differences that happens when two different professions work closely together. Any initial difference in status - maybe one profession has a higher bar to entry (diploma / entrance exam / professional licensing), or maybe it's a historical artifact (noble officers with commissions vs. drafted soldiers), or maybe one is the manager of the other and gets to order them around. Then the "upper class" tries to preserve the "gap" and prevent the formation of just such a single-ladder system as you describe. The upper class closes ranks against the lower class to prevent social mobility, as it were. The second factor is that the two classes are at least partially distinct professions and require different training or qualifications. So any new people coming into the system have already decided which class they belong to.
Sure. But the System doesn't think that way. A nurse can get in trouble with the law for doing what a doctor does. can a doctor get in trouble with the law for doing what a nurse does? Yes, that's it. No, most managers are not execs. I can't see much qualitative difference between the" professions" of senior manager and executive , except for this implicit presumption of agentiness.
Agreed. There is no qualitative difference between senior manager and executive beyond the increased scope of responsibilities and scale with which your decisions affect others, both of which could be attributed to the word "agentiness."
Could well be, but why does this concept of "executive" as a distinct group exist? Why not just have some managers who are more senior than others.
Perhaps there's simply no need for legislation that would prevent a doctor from working as a nurse. No doctor would even consider this option unless the situation is seriously screwed up. A nurse working as a doctor can do much more damage than a doctor working as a nurse. Or put in another way, an incompetent doctor will do much more damage than an incompetent nurse.
I don't see why. Either can kill the patients they work with, or set bad standards for other doctors/nurses.
A doctor has responsibility over many more patients. A doctor's competence is more complex, so he has more ways to screw up. A doctor can make more autonomous decisions, often makes them alone, and can't as easily ask help from others. He also leads a team, so he doesn't have to execute all his mistakes himself.
I disagree with your implication. Most doctors and nurses are responsible for specific treatment aspects of a few tens or hundreds of individuals. A small minority of doctors influence larger numbers (e.g. department heads), but even then many of their instructions are carried out by nurses and junior doctors. Nurses who administer treatments have ways to screw up that are not available to doctors who merely select those treatments. A doctor can order a harmful procedure; a nurse can actually administer the wrong drug by mistake. A doctor can make a mistake in surgery; a nurse can make a mistake in intensive care. I think empirical data is needed to elucidate the amount of mistakes (or "responsibility") owned by doctors and nurses. It's not something that can be clearly deduced from first principles.
You disagree with the implication or the claim? Do you disagree with it completely or do you think it's exaggerated? ETA: It seems it was exaggerated, but right [;,/ns/Chapter&itemId=/content/chapter/health_glance-2011-26-en&containerItemId=/content/serial/19991312&accessItemIds=&mimeType=text/html]. There are about 3 nurses per doctor on average in OECD countries. That's one of the key points I made. A nurse can't make other people carry out mistaken orders. A person carrying out a mistaken order isn't necessarily incompetent, although in easy to spot cases they are. Sure. A doctor has hundreds (or thousands) of diagnoses and medications to choose from. That makes me pretty confident that he has more options for fatal mistakes, and that the probability of doing the right thing is much lower. I think the more difficult question is what constitutes incompetence, and what levels of incompetence are comparable in nurses and doctors. I strongly suspect that our main disagreement comes from this aspect. I think I agree what Joshua said here [] so debating this point much further isn't of much interest anymore.
The claim was that doctors are responsible for more patients. I'm not sure that is correct; your link about there being more nurses says, Anyway, what I meant by the 'implication' (should have elaborated, sorry) was the implicit claim that because doctors have responsibility over more patients, that means they can do more damage if they screw up. But even if a single doctor does more damage than a single nurse (and I'm not sure of that), 5% of doctors screwing up would still do the same amount of damage as 5% of nurses screwing up. A doctor knows thousand of diagnoses but for a patient complaining of knee pain most of them are completely irrelevant and would never be diagnosed by mistake. Also, a nurse can administer any treatment a doctor can prescribe, so nurses also have theoretically thousands of medications to administer by mistake. Finally, most nurse also has a range of mistakes, having to do with physically taking care of the patient, which aren't available to most doctors, who don't perform dangerous procedures every day.
Certainly true but not relevant. Fair enough. We could continue this forever, but I wouldn't find that especially rewarding. I quote myself: Which requires more incompetence: making a wrong diagnosis and prescribing the wrong medication, or giving the wrong medication, when you're given the name of the medication? I'd say the latter, and so they wouldn't be comparable. Peace out.
I don't think that degree of damage is the essential point. Both can do significant damage.
Let's take a hypothetical but rather typically populated 24 patient hospital unit with 1-2 doctors and say 4-8 nurses on shift, plus other staff. Let's assume they're all competent. Replace one doctor with an incompetent doctor, or, replace one one nurse with an incompetent nurse. Do you seriously not see any difference?
I'm pretty sure you misunderstood what I said. Read it again. If not, try to think of examples more broadly. Your example misses the point. Edited to be less snarky.
Thanks, I've edited it.
Yeah. Sorry about the hostile tone.
I don't think MDs are automatically allowed to practice as nurses, if that's what you mean. Also, nurses have specialization just like doctors do, and only those with the needed training and certification are allowed to perform various special procedures. And I'm pretty sure doctors aren't allowed to perform those procedures. Why does it matter? MDs are special in the eyes of the law, but that's just something special to MDs, and indended to curb malpractice and charlatans, not just to preserve a guild's power. Executives aren't a legally special class. For army officers the question doesn't arise. I can't see much segregation between the two, either. Senior managers often become execs. Maybe it's a difference in the corporate cultures of our experiences, and your agentiness model fits the culture you describe.
I'm pretty sure that executives have legal liabilities that non-execs lack, although that might apply in most cases just to the top execs. The question of being legally allowed to do certain things? There are lots of duties in the army that only an officer can legally do but an NCO cannot, but not the other way around. I'm thinking of very large companies, including those I have worked in; and those I have read about. Startups allow agentiness to everyone.
Particular roles (CEO, board member, CFO, ...) have particular legal responsibilities. I don't think there's any law addressing executives or "top executives" as a class. There are roles that require being an officer, sure. But more technically, they always require being an officer of a certain minimum grade. Other positions require being a non-officer of a certain grade or above. And plenty of technical positions require specific training and certification, regardless of officer status. I would describe it as startups requiring agentiness from almost everyone qualified to work in a startup. Otherwise it becomes just a small company and probably fails. The startup where I work is just now transitioning into a post-investment non-startup company with a board of managers. And so, for the first time, we're explicitly looking to hire non-agenty people to fill some junior roles. Agentiness is very much the word I'd use to describe some of our staffing decisions. But we still don't have anything like an exec vs. everyone else distinction. I'm one of two technical architects, a programmer, very agenty, and definitely not an "executive" (and I don't have any legally binding duties beyond an ordinary employee with a contract).
Interesting example of non-agentiness being considered a plus for some jobs.
It's not a plus, it's an acceptable minus, a trade-off vs. a lower paycheck.
I have personally known and known of several people who prefer being enlisted men, NCOs, or warrant officers to commissioned officers simply because commissioned officers inevitably end up in a management role, while others can more or less stay in a role where they actually do things directly for the entirety of their career.

Another data point: the British Police have a single hierarchy. They are all called "officers", but in this case the word does not distinguish them from "enlisted", for which there is no analogue, but from people who do not have police powers (e.g. admin staff). Everyone enters at the bottom, although there are fast-track opportunities.

all police departments that I am aware of in the U.S. are single hierarchy. The dual hierarchy is a military thing. In the U.S. when there has been a draft (involuntary conscription) one is only ever drafted in to the lower hierarchy. The officer hierarchy was always strictly volunteers.
Also, even without a draft the lower and upper hierarchy have different induction methods. Military academy for the upper hierarchy, etc.
Interesting. That's a nice counterexample.

In my experience (being ex-military) the enlisted corp exists to mold subject matter experts, while officers have more of a executive/decision making role. It's hard to have both a subject matter expert and someone who has to make decisions based on the input of many different subject matter experts in one person, so those two roles were split between the enlisted and officer corps. Senior NCOs usually act as advisers to their immediate reporting officer.

We had briefings about this quite a bit in basic training, but that was years ago so I forgot most of it.

I have to disagree. Doctors are subject matter experts, and they are all officers. Your average doughboys/grunts and their NCOs are not subject matter experts, unless you push the definition, and they are enlisted.
I believe the main distinction was primarily historical when nobles and aristocracy commanded peasants. I had always thought that commissions (from the Queen/King or head of state) used to be put on sale by the state, similar to how France at one point used to sell public offices. In today's more modern times, one can become an officer by dint of having a post-secondary education. At least in Canada, you are typically an officer when you enlist provided you have a bachelor degree and pass certain intelligence tests. Everybody else (NCOs or enlisted) typically become technical SMEs due to lack of upward mobility. Doctors are SMEs, but they also have extensive post-secondary education. Average grunts and NCOs don't start out as SMEs, but given enough time (provided they survived) become an expert would have made perfect sense.
Even more recently, I think it was that enlisted men hardly made any decisions at all. Isn't the modern idea of the moderately agenty enlisted man a result of post-WWI squad-based mobile combat?
Yes, technically the exact difference between officers and non-officers is that officers receive a commission by the sovereign. That's a formalism and does not explain why there is a two-ladder system.

If you look for the commonality between various multi-ladder systems (are there ever more than two ladders?), you will notice that it originates historically in the drawing from multiple pools of candidates into the same occupation (like fighting or healing). Eventually the ladder may become a lost purpose, preserved only because it's an integral part of the system (a deep local maximum of efficiency, for example). The pools could differ by land ownership, education and/or training level or even by gender. If the original reason for the double ladder no lo... (read more)

Traditionally the distinction is that employees do the work, the managers manage employees, and the executives manage managers. In the XX century there was also real distinction between different ranks of managers in certain spheres, notably law firms and investment banks. They were organized as partnerships so the boundary was between employees (including managers) who were paid a salary and partners who had an equity stake in the company and received a share of profits besides the salary.
Actually, there are many non-exec managers who manage managers, so that can't be the defining distinction. There are also a tiny handful of execs who don't manage anyone, though that might be just an exceptional case.
The definition of "executive" is fuzzy and there has been some inflation over time. Look at titles, e.g. vice president. Fifty years ago it meant you're a big shot, nowadays it usually means you're a middle manager. Note that there is also a separate, legal concept of "officers of the corporation" which is a different thing.
OK, so the two pools originally were the founders and the employees, then ("capitalists" and "workers" in Marxist terms). Before it all morphed together.
No. The basic definition of executives, from the perspective of the capitalists who own the company, is responsible employees who take care of the business.
Yes, this is one of the strongest candidates for an answer. Still, I think that organizations do change, sometimes keeping historical forms as an archaism. I can easily imagine a counterfactual historicla shift where sergeants' social status rises and lieutenants' social status falls, to the point that de facto the strong boundary between officers and non-officers disappears. So, just explaining the two-ladder system as a historical remnant doesn't quite answer it..
That's not a "still", it's exactly what I meant. After the original ladders take root, it is much harder to reform the whole system than to adapt it to the changing circumstances. I would like to see such ladder inversion examples. While it is true that a new lieutenant has to learn from his or hers sergeant before really talking command, there is still a difference in the social status and education between them.

Separating people into classes allows you to treat the classes differently. All the people who are trained for leadership roles can go to university together while it's not necessary to have foot soldiers in the same university courses.

Yes, university graduation does seem to be on the other side of a definite divide in our society, as compared to high school in our society. And there are certainly social class divides. I wouldn't push the agentiness idea there, however.
Actually, maybe some people do have explicit lessons in agentiness. (And they probably want their children to have the same lessons.) But I would guess the agentiness is just a subset of some larger curriculum. Also, CFAR should make agentiness lessons. Making people rational and agenty seems like a good combination for world optimization.
What would you do specifically if you wanted to teach agentiness?
I imagine something like teaching people a "think about the problem -- gather data -- make a decision -- do it" loop, starting with simple problems and gradually moving more meta. The emphasis would be on doing; on closing all the started loops. At the beginning there would be dozens of simple problems, all of them have to be solved very soon. Not just hear the theory and profess an agreement, but repeat it so many times it becomes your natural reaction. This would create a habit and an expectation that problems can be solved without procrastination. (Something like the "success spiral" in motivation literature.) Then move to more complex things, and perhaps add more steps to the loop, like "reflect upon what happened". The idea would be to train people out of feeling helpless and stopping (i.e. not merely stopping doing, but also not gathering data, not discussing the problem, etc.). I imagine that when a person with this training would see something that can be fixed easily, the person would a) recognize it as belonging to the class of the problems that can be fixed easily, b) make a cost-benefit decision, mostly about the opportunity costs, and in case of a positive decision, c) fix the problem. And the next problem, and the next problem. If you had this person as a friend, your observation from outside would be that many problems just somehow disappear when this person walks around. Then, if you were interested, there is a chance this person would think about a way to teach you this skill efficiently, too. I imagine that given our human limitations, an important part of the skill would be to have a supervisor or a support group, and regularly report to them your problems and solutions; sharing knowledge and encouraging each other. I would expect that something similar already exists, but it's very context-specific, not generalized. For example how to behave as a spy or a soldier, or how to survive in a jungle... with a lot of domain-specific knowledge, and s

There are different status levels among doctors and among nurses, but a PhD in nursing stands on the other side of a clear border from a beginning MD. Similarly with lawyers and paralegals. These dichotomies stem from licensing restrictions, which in turn are descended from medieval guild practices.

Are you sure you understand what these professions do? There's probably less overlap in the skillset of doctors and nurses than you think. A doctor can't do a nurse's job without some training. Might apply to lawyers and paralegals too, but I don't know.

You ... (read more)

An interesting data point is airfoces. There everyone actually flying a plane is on the officer hierarchy and the NCOs/enlisted are simply support staff.

Yes. As mentioned, this contradicts the notion that being an officer is strictly connected to command status.

I don't see how your evidence proves your conclusion. Militaries and many corporations choose to invest agency in some small fraction of their employees. That doesn't make agency a natural kind.

My impression is that this sort of two-layer hierarchy is much more about history than about any philosophy of how organizations should run. You mention the possibility that this is social/historical but I think don't really rebut it adequately.

Right, it doesn't. But my conclusion was not that agency is a natural kind, but rather that "these organizations [implicitly] agree with LessWrong that [agency] is a natural kind." Indeed it is. But why do all the armies in the world, as far as I know, have a two way distinction, including armies with roots in different cultures? Even ragtag rebel armies generally have a two-rank system of soldiers and "commanders." Why has no one experimented with a straight one-ladder hierarchy? Even the wacky Japanese Self-Defence Forces, which pretend to be a completely civilian organizations do it this way [].
When you mentioned different cultures I was imagining cannibals in the jungles of Borneo, or something. The JSDF is a completely ordinary western-style military, patterned after other modern armies. I don't think it counts as evidence that this is a human universal.
From what I have read in the recent book by Jared Diamond, fighting in New Guinea, is of a completely different type -- quite non-hierarchical. Can there be hierarchical armies without the two-ladder hierarchy?
Is the fighting in New Guinea on a large enough scale to require a hierarchy of any kind? A group of just a few hundred men, without strongly segregated roles that need coordination (they all fight together in the same way), is just a warband. It doesn't need a complex command structure to coordinate; it might help, but doesn't have the training and regulations to enforce it. It's easier to be egalitarian than to have to worry about political infighting during actual fighting.
That's part of Diamond's point: These societies are organized on a small scale. Other similar societies scaled up and moved to a chieftain system. But in any case, even small, elite, independent army units have the usual officer/enlisted dichotomy, even if in the elite units this is a less clear distinction. Possibly, if the line of this post is correct and I am not taking it too far, this is because the SEALS, Sayeret, etc are implicitly expected to be agenty in a way that most soldiers are not.
I have a suspicion that during meatgrinder wars (e.g. WW2) the two hierarchies if not merged then became a lot closer. Survivors who showed skill were promoted up including from sergeants to lieutenants. I also wouldn't be surprised to learn that the Soviet military in the 1920s and 30s was close to a one-ladder deal.
Interesting. Yes, there were battlefield commissions []. Note that such promotions were given rarely and usually with a "brevet" designation to keep it from being exactly like the "real" thing. But to the extent that this existed, it might fit into the framework of this post as a recognition of extreme and exceptional agentiness. Battlefield commissions were not, officially, given for skill, hard work, nor for filling the slot -- the usual reasons for a promotion -- but rather for "outstanding leadership on the field of battle." Do you have a reference on that?
I recall reading that Napoleon promoted enlisted men who showed conspicuous bravery to officers (I believe continuing a practice started by the French army during the revolution). The historian I read who discussed this practice said that the lack of education of these officers did prove a hindrance to the French at times, but there were also definite advantages to the practice; it produced officers who were conspicuously brave, and set a good example, and it gave the enlisted men incentive to try harder if promotion was possible. But perhaps the biggest benefit was that it guaranteed Napoleon could replace his losses; in most European countries at the time, only aristocrats could be military officers, and losses in the Napoleonic wars were high enough that some countries ran short of remotely suitable aristocrats to employ as military officers.
Enoch Powell rose from Private to Brigadier during the course of the Second World War. [], though this was very rare.
He was commissioned because he was educated, so I don't think that's a good example.
Nope, and I am too lazy to go search, but it would be consistent with the early Soviet emphasis on egalitarianism and their need to construct an army almost from scratch.
Their recruits with any experience in other, "real" armies naturally try to emulate them. The desperate situation of a new rebel army probably doesn't encourage experimentation with new ways to organize. If a rebel army has no-one with actual military experience, then it probably doesn't survive long enough to become a data point - or it just looks like a huge mob or a set of independent bands.

And the distinction is not based on command: New army doctors automatically become officers, even if they don't command anyone. Doctors are non-combatants, but fighter pilots are combatants par excellence, don't command anyone, and are all officers.

I don't think all these are invariant between armies of different countries.

To the extent that the separation between officers and enlisted men is universal, I think it's probably because most countries militaries are modeled after Western first world militaries, which co-evolved over the history of Europe, ... (read more)

Yes. Why? The usual answer is that officers "decide," "plan" or the like. I don't see a qualitative distinction between deciding, planned etc at different levels, other than this nebulous concept of agentiness.
I've never been an enlisted man or an officer, so I don't have any experience with what either one entails, but at a sufficient level of abstraction, every job comes down to some combination of deciding and acting. The fact that officers "decide" for a larger number of people doesn't mean that one effectively learns to make the sort of decisions an officer has to make by learning from the practices of noncommissioned officers and scaling up.
I recommend Absolutely American by David Lipsky, about the West Point experience. It left me with a lot of questions about what is going on in the officer training program. The graduates seem well-trained, but still, they are ordinary schmoes like you and me, not some sort of stereotypical leadership figure -- emotionally and physically strong, determined, foresightful, brave, etc. And why were some of these people sent off to the finance corps? Can't the army fill the finance corps in another way? I also wondered why there are both West Point and ROTC -- both produce officers, so is one "better" than the other? Or are they thought to produce the same level of quality, but the two programs just offer a choice to the potential officers? But to bring it back to the point of this post, the concise answer is that officers are trained to be a different class of person from the enlisted -- an arbitrary distinction which may correlated with many characteristics but ultimately is a formal classification of soldiers into two types.

You don't want empathy between bosses and grunts. Folks on different ladders have less empathy than those at different rungs on the same ladder.

It is easier for a Level 2 Boss to order his Level 8 minions around than it would be for a Level 12 boss to order around his Level 10 minions. Similarly a General will wield his Soldiers more dispassionately than he will his Not-Yet-Generals. The reverse is also true. It is easier to have faith in THE COMMANDER than in a peer who has many more grades than you.

While there are certainly good reasons to maintain emotional distance between the people giving orders and the people going out and getting shot at, I strongly suspect that social separation does more of the work there than formal grade separation. The latter can be a means of enforcing the former, but there are other possibilities as well. Western-style air forces or naval aviation are probably the best example: all the people doing the shooting are officers, but they're getting orders from a significantly higher echelon of officers (who, in the Navy, might not even be on the same promotion track). Even in armies, though, there are significant jumps in danger and responsibility between company-grade officers and field-grade, or between field-grade and general officers.
As mentioned in the comment on fraternization , the dichotomy may well be a useful anti-empathy device, but only at one of many possible cut-off points.

Perhaps in your corporate ladder discussions it may be useful to mention unions vs. managers to make the point clearer. Focusing on senior managers vs. executives is a weaker description because most people see that as one unified ladder, and not two separate ladders.

Similarly: What's a corporate executive? I understand that there is a management hierarchy, but why the arbitrary distinction between a senior manager and a junior executive? Aren't those just two rungs on the ladder? In corporate-speak, an executive is called a "decision maker."

... (read more)
OK, thanks, I guess it depend on the organization. In the large companies I am familiar with, "Executive" is a definite formal status, and most managers are not executives. But I see that "senior management" can be used [] as a synonym for "executive." Still, most managers are not executives, even where that terminology is used. Also, but some fast-track up-and-comers are executives -- these might be called junior executives. I don't think so. Do you mean an executive is responsible for a function like marketing across the corporation? There are definitely VPs of Marketing and VPs of Development in charge of one part of a company. There are non-execs who are in charge of marketing for some smaller piece of the company. There are execs who are responsible cross-functionally for a profit-and-loss line of business, not one function. Instead, it seems to me that executive is an arbitrary, though formal border between two groups, and, though no one says so out loud, the underlying difference is what I said in the post.

I don't think it's correct that the structure described here fits militaries cross-culturally, except in name. In the US and most Western European military structures, the senior NCOs are the critical link. The degree of authority and autonomy given to senior NCOs in the west is fundamentally different from that in most other military structures. I can't comment on this from personal experience, but every US military service member I've known who has served extensively alongside militaries outside the US, Canada, or Western Europe has commented on this, an... (read more)

Thank you. This strengthens the question: Why this weird, very separate, category of "officer"? Given that the agenty/non-agenty distinction is not real, why the arbitrary boundary? Why not just have a ladder?
Well, one answer may simply be that militaries are class-and-tradition-laden bureaucracies that are hostile to change. Certainly this has been the experience of the US when attempting to build a western-style NCO corps into the militaries of allied developing nations. Another answer might be that it's already possible in principle: somebody has probably already mentioned this, but one can go from a NCO role into an officer role; the traditional term for this is "mustang officer". Another answer might be division of labor: traditionally, western military officers are required to gain a lot of breadth in their careers, both through academic education and through assignment to different types of field commands, logistical positions, and political positions. This poses a (common) problem for officers who have a personality type that wants depth in one type of military occupation; that role is more traditionally a senior NCO role. It may be, however, that this type of division of labor produces the most effective fighting force; I don't know.
[-][anonymous]9y 3

How often are the two tiers of agentiness

"Capable of hiring/firing people" and

"Not capable of hiring/firing people?"

I mean, at some rank in an organization, you sort of have to get the power of adding and/or kicking people from your part of the organization. That strikes me as a big step in power that can require entirely different sets of talents. I'm an adequate programmer, but I have no practice at selecting other people for it.

That being said, I don't know for instance, if a low ranking (but finished with training) officer can kic... (read more)

A manager of a local branch of a chain store or fast food restaurant certainly has the responsibility of hiring/firing workers, but is far from an executive.
That is a good point. It seems like there is clearly also some kind of factor in the size of the organization itself. As an example: If I form Mike's Burgers as a potential new chain and I incorporate it when I have one store, it is entirely possible for me to simultaneously be both a store manager, and a company executive of the corporation, whereas the CEO of a bigger chain might have thousands of stores and so doesn't individually manage any of them. Am I correct that when most people think of CEO's, they are thinking of CEO's who are not also working substantially at another job such as store manager (or programmer, given some software startups?)
My post was about execs in large companies. I have edited it to make that clearer. Certainly things can be different in small companies, particularly startups.

A similar thesis is found in Competent Elites.

Yes, except that in my post, I am saying that the organizations implicitly define the officer/exec roles as as having agentiness, not that officers and execs have agentiness. (Also, Eliezer's Competent Elites post is about people who have something far rarer than "mere" agentiness, rare as that too is.)

I think that in the military, the "no fraternizing with enlisted personnel" rule might be one reason why a hard separation is useful. This kind of rule requires a cutoff and can't easily be replaced with a rule like "no fraternizing with people of a rank three or more below your own." For instance, how would you set up the housing arrangements? Also, promotions would be awkward under this system, since you would always have a group of people you previously could fraternize with but no longer can.

Yes, creating this arbitrary dichotomy keeps lieutenants from fraternizing with enlisted men. But it doesn't keep Chief Master Sergeants from fraternizing with privates or generals from fraternizing with lieutenants. So, taken merely as a way to prevent fraternization across ranks, the dichotomy is of little value.
Well, just because the rule doesn't by itself prevent all possible cases of inappropriate cross-rank fraternization doesn't mean it has no value. There are other norms and practices that discourage generals from hanging out with lieutenants, e.g. generals usually get fancy lodging separate from the lieutenants. I suspect that cutting off lower-ranking officers from fraternizing with enlisted men prevents what would otherwise be one of the more common problematic cases. If the military were even more concerned with this problem, it could have three or more groups instead of two, say, enlisted, officers and super-officers. But there are also tradeoffs to having more groupings, so the military sticks with two (part of this might be historically contingent, maybe three groups would work just as well but everyone is just copying the consensus choice of two).
Good points. But they don't explain this arbitrary dichotomy. Indeed it does -- 23 or so groups [], which are the ranks. That's plausible, though we should be cautious of reverse reasoning: Did the Lieutenant/Sergeant border arise to prevent fraternizing across those levels, or is fraternizing across those levels considered extra-bad becauses it crosses the officer/NCO border? I am not convinced that this is a good explanation for why the dichotomy exists in the first place.
You should read up on the notion of Schelling point [].

What are you saying about executives?
Are you saying that they are structured like armies?

I don't think that's true, or even that people use "junior executive" at all consistently. My understanding is that companies often have people "on the executive track." These people go through normal management ranks, but get promoted more quickly. Are they "junior executives"?

And do they exist because it's necessary to train people to fill executive jobs, and young people are better than experienced but older senior managers; or because those individual people are being promoted via politics and personal relations?
It could be either reason, but why is it at all necessary to choose some in advance to move quickly up to the executive levels. Why? Why not just pick the best person for each job, sometimes taking people from a lower level to shoot up faster than others, but not designate them as a special class? This is what occurs for promotions within the non-exec category or within the exec category. And in armies, why choose some people in advance to be officers? Why not just let everyone climb the same ladder, but occasionally choose some good ones to jump up a few rungs?
The second theory would explain that: maybe they're someone's special pet (and promoting them even faster would attract unwanted attention).
No, I am just saying that "executive" is a distinct class. They are not just higher level managers. There are legal definitions to this, but they follow on the organization setup rather than the other way around. The companies define that, which is exactly my point. The distinction between "manager" and "executive" seems arbitrary but real, and I wonder why they invented it.

I think "agenty" was made up to point out that while all humans are agents to some extent, some do it far better than others.

Personally, "agenty" makes me squirm a bit and I wish we'd just used "agency".

Yes, likewise for me. I used "agenty" in this post because it was specifically about LessWrong's perspective.

A somewhat relevant statement by Teišeba Biainili in the Quora discussion, " The enlisted man swears to follow all orders, the officer does not."

My guess is just that the original reason was that there were societal hierarchies pretty much everywhere in the past, and they wanted some way to have nobles/high-status people join the army and be obviously distinguished from the general population, and to make it impossible to be demoted far down enough so as to be on the same level. Armies without the officer/non-officer distinction just didn't get any buy-in from the ruling class, and so they wouldn't exist.

I think there's also a pretty large difference in training -- becoming an officer isn't just about skills in war, but also involves socialization to the officer culture, through the different War Colleges and whatnot.

That division is in fact not universal and not clear.

First, in some armies promotion of enlisted men and NCOs is or was in fact main source of officers. One example is German army before and during WWII. Link in Russian

In short, no higher education was required and training itself took 6 - 9 months.

Second, USA has warrant officers, USSR had (and some post-Soviet states now have) praporshchiks, and there are many more categories that are between enlisted and officers.

And reason why that two-ladder system with classist origins is not univesally abandoned a... (read more)

But why does it have to be this way? Why not just rank medical personnel, or legal personnel, in a single continuum from practical nurse through rockstar brain surgeon. (Is that a title?). There would still be the understanding that some people will never climb beyond a certain point, while others can jump straight to a higher rung.

The only reason I can think of is that the establishment believes that mastery in nursing does not imply even a beginner's level competency in doctoring? (similar to how mastering ecology doesn't give one even beginners compe... (read more)

There is no single "master nurse" title. Nurses take specializations (just like MDs can become e.g. ophthalmologists), or smaller courses in specific proficiencies (eg to administer specific treatments). Maybe a nurse that took every course in existence would know as much as an MD (though I expect there would be bits missing here and there, particularly theoretical biochemical knowledge), but a regular nurse certainly doesn't.
Titles don't have something directly to do with competency. I would guess that most MD forget most of their theoretical biochemical knowledge 10 years after finishing their studies.
That doesn't mean such knowledge shouldn't be required from MDs, including transitioning nurses. Or if it does, then MDs shouldn't have to learn it in the first place.
If most of them are allowed to work without that knowledge it defacto isn't required for working as a doctor.

Historically, the distinction was based on social classes, but that doesn't explain why every army follows this arrangement, including those in very different societies.

The claim that all societies use this model is inaccurate. The counterexample that springs to mind is the Roman army; I'm fairly certain that there are plenty more.

Interesting. I foiund a couple [] of pages [] which say that there was an officer/non-officer model. But perhaps those pages are just shoehorning the historical Roman army into modern forms. Do you have a reference to the contrary?
Actually, one of the sources you just linked (Wikibooks) states that officers were usually promoted from the ranks: For further sources saying the same thing, see here [], here [], here [], or here []. See also this []:

With all respect, there isn`t any NEED for officers and a two-track military. This is 100% a holdover from the European class system. Just think about the expression, "an officer and a gentleman." What do you think GENTLEMAN means -- someone who is polite? The distinction between "officer" and "man" is entirely historically based--and ridiculous. Like lawyers using the designation "esquire"!

Why have no militaries worldwide adopted a one-ladder system? There are hundreds of militaries, and at least some of them have the ability and desire to break away from old European models. Most importantly, military systems are tested in the Darwininian struggles. If a one-ladder system is better, someone could have adopted it as a way of beating the enemy.
The "Darwinian struggle" argument is fallacious, as the status quo is not necessarily the best way of doing things, especially in a conservative, class-based society like the military. History is full of examples of how the generals (ultimate expression of.the officer class) got it wrong.
OK, so someone should try organizing differently and whup the old-time two-ladder armies. The most likely candidate for this would be a ragtag militia trying to form itself into a national army while under pressure to beat enemies.