Ansuz

« Fixing cut-off margins and bad... | Home | MakeでLaTeX言語のビルドしましょう »

Scarcity, abundance, and lost careers

How should institutions make hiring and promotion decisions, in theory? How do institutions make such decisions, in actual practice? What happens, and what should happen, when someone's career is interrupted? Is it possible to restore an interrupted career, and should that be done? What happens to institutions when society overproduces, or underproduces, elite individuals? This article looks at ways to understand these questions, starting from an historical episode.

The Dreyfus Affair

In 1894 a young officer of the French Army named Alfred Dreyfus was accused of selling secret information to the German military. He was convicted by a court-martial, sentenced to life exile, and imprisoned in horrific conditions on Devil's Island, off the coast of French Guyana. The evidence against him was flimsy to begin with, and almost immediately was discovered to be fabricated; but senior staff in the Army, apparently hoping to avoid the embarrassment of admitting they had made a mistake, attempted to cover up the fact that they were punishing an innocent man.

Dreyfus was Jewish, and that made him appealing as a scapegoat. Many in the Army and in the public were willing to consider him automatically a likely traitor just because of his heritage. These events took place long before the Second World War and the existence of Naziism. Anti-semitism as a political position was certainly abhorrent to very many people in France at the end of the 19th Century, but not to everybody; at that time it was still within what we might now call the Overton window, of political positions that could safely be taken in public discourse. There was even a "National Anti-Semitic League of France" founded in 1889, which was able to openly operate and call itself that.

In the years following Dreyfus's conviction, his family's efforts to clear his name gathered public attention and became inextricably linked with the left/right political conflicts that existed in the French Third Republic. Whether someone thought Dreyfus was guilty or innocent was not just a question about the facts of one court case. It became a statement of aligning with a side, and of implicitly taking a position on many other issues as well. There was constant newspaper coverage; there were leaked documents and rumours; eventually there were multiple court cases, and tampering with evidence, witnesses, and judges; there were duels fought over personal conflicts connected with the situation; there were suicides of important figures and suspicious-looking accidental deaths; violent street demonstrations of both sides occurred all over France and sometimes internationally; scientific and pseudoscientific questions related to forensic examination of documents became significant (see Appendix A); and so on. The 12-year period of political controversy and social upheaval set in motion by the original accusation became known as the Dreyfus Affair.

In July 1895, an officer named Georges Picquart (then a major, promoted to lieutenant-colonel very soon after this appointment) became head of the "Statistics Section," which was the deliberately misleading name of the Army intelligence department. Picquart held anti-semitic political views; initially believed Dreyfus's guilt; and had been a witness at his first court-martial. However, in this new position, Picquart became aware that the espionage for which Dreyfus was imprisoned was still ongoing, indicating that at the very least, Dreyfus could not have been the only traitor involved. Picquart investigated further and found evidence that convinced him Dreyfus had been innocent from the start, and led him to the real spy, a Major Ferdinand Walsin Esterhazy.

Picquart refused to drop the matter after being repeatedly warned away by other officers, who were shielding Esterhazy partly in order to conceal their own culpability. The senior staff conspired to get rid of Picquart, first by sending him to remote locations including Tunisia (a dangerous posting where it seems they were hoping he would be killed), and then imprisoning him on accusations of various offences, including forging documents and leaking information to the newspapers. The process of his formal court-martial was interrupted by a second court-martial of Dreyfus, and by political developments on the civilian side. A bill in the Chamber of Deputies (the lower house of France's bicameral legislature as it existed at the time, roughly equivalent to the House of Commons or House of Representatives) aborted all the then-current Dreyfus Affair court cases in an effort to quell the ongoing public uproar. Picquart was released from prison in 1899, but had been forced out of the Army.

Alfred Dreyfus was brought back from Devil's Island in 1899 to face his second court-martial, which revisited the charges in the first one, examining the new evidence of his innocence, but still convicted him a second time. Then he was pardoned by executive action of the government. That meant they were deciding not to punish him any more, without in any official way admitting that he really was innocent. The wording of the letter of pardon signed by the President of the Third Republic made clear that it was a political and not judicial act. Dreyfus was free to re-enter civilian life, but he had no career in the Army anymore, and what he perceived as his family's honour was not restored. The entire situation continued to be a controversial subject of public discourse; the government's attempt to end that was largely unsuccessful.

In 1906, after a further seven years of public controversy, political and legal maneuvering, and the election of a new government more favourable to the pro-Dreyfus side, the civilian Supreme Court quashed Dreyfus's conviction, ruling that all the evidence supporting the initial charges against him had become legally inadmissible by reason of having been fake. The decision was "without reference" to another court, meaning that he could not be subjected to a third court-martial. Dreyfus was disappointed not to have his name cleared by a military court, and the decision without reference put that out of reach; but unlike the pardon, this was an official admission of his innocence. It was considered to restore his honour as an officer and a gentleman, and it meant he could serve in the Army again.

This description is highly abbreviated. There is a lot more that can be said about the Dreyfus Affair and I have focused only on the points that are most relevant to what I want to say about institutions, promotion, and elite production. See the bibliography below for work covering other aspects of the historical events.

The what-if game

Although there's a lot that can be said about the Dreyfus Affair, the point that interests me most is one detail: how the French government and the Army, as institutions, tried to make amends to Alfred Dreyfus for his mistreatment at their hands.

They played the what-if game. Dreyfus had trained for, and was at an early stage of, what ought to have been a career as a military officer that would last the rest of his working life, when the false accusation threw his life off the rails. After five years of torturous imprisonment and seven years of spinning his wheels as a civilian, he'd lost twelve years out of that career. So they attempted to give those years back. They reconstructed the career he might have had if not accused, and they promoted him to the level within the Army that he would have been at if he had served twelve years of a normal career, working his way up through the ranks, instead of what actually did happen.

Or, so they (specifically, the Chamber of Deputies) said at the time. I discuss below whether the promotion he received actually was consistent with the what-if scenario. For the moment, let's suppose the institutions really did in good faith give Dreyfus, upon his exoneration, all the promotions and other status he would have achieved in twelve years of an hypothetical uninterrupted career.

Was that the right thing to do?

Well, we can say he earned it. He underwent hardship for it. As a matter of justice I'd say they owed him at least what they offered and probably a lot more. It is remarkable that he still wanted to serve in the Army after what the Army had done to him; but given that the career of an officer in the French Army really was what Alfred Dreyfus desired, and that career was unfairly taken away from him, from the point of view of his rights as an individual, he ought to be given it back.

Let's do another what-if. Suppose instead of promoting Dreyfus, they just allowed him back into the Army at the level he had been at when he left. That was the level of someone who had recently completed a program in a military graduate school before joining the General Staff for a two-year probationary period of on the job training, which was almost completed at the time he was accused. If he picked up at that point, exactly where he left off after twelve years, it would make him probably the oldest new trainee in the General Staff's history. Not exactly an auspicious place to launch a great subsequent career. Even if he never faced any age discrimination, he would be unlikely to make an impressive lifetime achievement record before he retired simply because he wouldn't have time. They would be giving him employment, but closing the doors that had been open before him twelve years earlier. So the scenario of picking up just where he left off, would not really restore what was lost.

But being a higher-level officer is a job with duties and responsibilities. Was he really fit to do that job? Normally, someone serving at the level where the Chamber of Deputies said it would put Alfred Dreyfus, would have had twelve years of experience that he did not have. The officer who went through the normal course would have been commanding troops, fighting battles, making executive decisions, or otherwise fulfilling the duties of an Army officer all those twelve years, acquiring the skills that would make him fit for promotion. Dreyfus instead had twelve years experience of nearly dying from tropical diseases, teetering on the edge of sanity in solitary confinement, and getting caught up in a public scandal. It might be said that he had a twelve-year gap on his resume, which would not cease to be true just because he got a new job title.

Suppose you are an enlisted soldier about to go into a battle where you may well die. Do you want to be commanded in that battle by an officer who has spent his whole career up to this point actually doing the job of an officer? Or do you want to be commanded by a man who is the same age and official rank, and who morally deserves to be considered as having the same number of years of experience; but who did not actually do the job of an officer for most of his years?

Furthermore, suppose you are an officer even higher on the chain of command than Dreyfus. You are choosing who will lead a unit on a difficult mission. You need an experienced and skillful officer. Will you choose the man who has twelve years of experience, or the man who morally deserves to be considered as having twelve years of experience? Bringing Dreyfus back into the Army as the oldest General Staff trainee ever, would not have been right; but bringing him back as the least experienced higher-level officer ever, might not be a lot better.

With the possibility that those above him might treat his deemed experience as less than entirely real, even just from Dreyfus's own point of view the promotion might not be doing him a great favour. His future opportunities might not be all he could hope. Would he be treated as an equal by other officers who had advanced to the same rank, but in the usual way? Everybody would know his promotion was honorary and different from the others. The Army as an institution might well perceive him as less qualified than most officers at the same official rank, and it could impair his ability to advance any further after this one special promotion. And this might not be only a matter of perception. It could be objectively true that his ability to do the job of a higher-level officer would be less than that of someone who had followed a normal career path to high rank.

Should a high-ranking officer be deserving, or skilled? Is it more important for the soldiers of the Third Republic to die waving bayonets for honour, Marianne, and liberté égalité fraternité; or to win battles and achieve strategic objectives? Should their officers be honourable, or effective? Is that a false dilemma after all?

These questions should not be answered too quickly.

Consider Picquart. He held anti-semitic political views, but ended up taking the Jewish side in the Affair, losing years of his own career and risking his life, because that was the honourable and morally imperative thing to do - a mitzvah even if he would not have used that word. Maybe his dedication to truth and duty was the right stuff for a senior officer, and actually more important in determining his suitability for promotion than years of experience would be. Soldiers deeply committed to honour, Marianne, and liberté égalité fraternité would rightly prefer to be led by such a man. The abstract principles at stake here really do matter. Without them, the Army might win a battle only to have it mean nothing.

The way I've told this story, it seems the argument is very strong that promoting Dreyfus and Picquart was a bad thing, an example of irrational behaviour on the part of the institutions involved. But I deeply distrust that quick assessment - as I've said, the questions should not be answered too quickly. I feel like I want to say that it was actually a good thing for the Army to promote these men. Working through why such a conclusion would seem appealing to me leads to two distinct lines of thought: two different theories of what promotion might really be for and how an institution should handle it. The two theories have opposite conclusions. Either of them might be right. Importantly, neither of them is universally right for all institutions at all times. One might have been right with respect to the French Army in 1906, and the other with respect to, for instance, a North American university a century later. It all just depends on the prevailing conditions of the moment.

Scarcity

Suppose a fly upon the wall (une mouche sur le mur) at the French Army headquarters overhears a briefing on the general subject of promotions. It might run something like this, translated.

Serving France as a high-ranking officer is a difficult job for men of the highest skill and experience. Good candidates for promotion are scarce. Most lower-ranking officers are not good enough to be promoted higher. Promoting one who is not capable of doing the job has disastrous consequences. We must find the best candidates, and only the best, objectively evaluating them. We must train each candidate to the highest standard he can reach, and be sure not to lose any. When we go looking for someone to fill a position, there are so few applicants who meet even the minimum objective requirements, that we may be forced to leave the position unfilled. Of course, we cannot promote anybody for reasons other than his actually having the necessary skill and experience, because the Army needs the very best men. Vive la France!

An institution operating in conditions of scarcity has more high-ranking positions requiring skill and experience than there are qualified promotion candidates who can properly fill those positions. An institution operating under a theory of scarcity is one that perceives the conditions of scarcity and responds accordingly.

By scarcity I mean scarcity of qualified candidates, as a proportion. There are likely plenty of lower-ranking candidates the institution could at least consider promoting, but almost all the candidates are bad. It is difficult to find the few who are really able to do the more difficult work of higher-ranking positions.

The theory of scarcity has assumptions built into it about the shape of the problem it is describing. The core of scarcity theory is the institutional belief that there are not enough suitable candidates for promotion. In order for that to be a meaningful statement, and for the institution to be able to operate at all, we also must believe other propositions about it being meaningful to describe a candidate as suitable for promotion, or not; and what it actually does mean for a candidate to be suitable for promotion.

Scarcity theory says skill is a quantity. Some candidates have more or less skill than others. A candidate who can do a certain difficult job, at all, necessarily has at least a certain amount of skill; and a candidate with more skill can do that job better, or can do a more difficult and therefore higher-ranking job. Difficulty of the work increases with rank of the job as a matter of definition, not only an observed fact that might happen to be true.

Skill is an objectively definable quantity both for candidates and for the work they do. It may possibly be hard to measure skill, and it may be multidimensional and therefore somewhat harder to compare between individuals, but in fact it usually is easy to measure skill, and whether we can measure it or not, a single correct answer exists to the question of how skilled a candidate is along a given dimension, or how much skill a job requires. Given two candidates, it is meaningful to say "this one is more skilled, that one is less skilled"; and fine gradations of skill exist, so that we will seldom to never be confronted with two candidates and no way to prefer one over the other. Given two positions, it is meaningful to say "this one requires more skill than that one." It is a matter of objective fact which one is which.

With a scarcity of candidates who can safely be promoted to the higher-ranked positions, the institution cannot afford to lose any. The institution in scarcity will offer incentives for everybody to advance as far as they can. The high-ranking positions in an institution operating under scarcity are likely to be well-paid and will certainly be high-status, that is, positions of honour. The institution depends on keeping everybody desirous of those positions.

It is important not to promote anybody who does not meet the necessary objective standards, even if that means positions go unfilled. Such a person could cause damage directly through incompetence in doing their difficult job, but their promotion also indirectly damages the system and the ability of the institution to get the best work from the best candidates, by harming the honour and prestige of the higher ranks. One unqualified individual at a high rank creates the suspicion that standards in general are too low and anyone else at the same rank may be similarly incompetent.

Scarcity means that most individuals at a given rank will be ones who only just barely met the criteria for that rank. Individuals with significantly more skill would be promoted further instead of stopping at their current rank; and individuals with more than the minimum skill for their rank, but not enough to reach the next rank, will be a small proportion because of the overall bottom-heavy nature of the assumed power-law distribution. In any interval of skill levels, most of the population who fall within the interval will be near the bottom. The point that most individuals under scarcity theory end up in ranks for which they are just barely qualified is basically the well-known Peter Principle, although derived from different assumptions.

Beyond being objectively definable, scarcity theory holds that skill can be increased by experience and training. There might be some variation in genetics or similar that might cause individuals to start at different points, or even possibly limit their eventual advancement, but skill primarily comes from experience and training - not from identity. The institution doesn't and shouldn't care who you are. What really matters is your experience and training. Scarcity theory says that good candidates for promotion are made, not born.

Objectively definable skill means it should be possible to write down a precise criterion for a candidate to be eligible for each rank, and an institution that follows the theory of scarcity is likely to indeed write down such criteria. The possibility of training individuals to a higher skill level means training itself can be a way of measuring skill: the objective criteria for promotion can often be along the lines of "You must have completed this course to qualify for this rank," maybe with some kind of a threshold on the final exam score or similar. Credentials and certificates are useful and often used by institutions under scarcity, as a way of abbreviating job requirements into criteria that are easy to define concisely. The idea that skill increases with experience also leads to easily measured objective criteria based on length of experience, as "You must have held this rank for this many years to move up to the next rank."

My paraphrased criteria are stated in terms of necessary conditions for promotion, not sufficient conditions. Just because you completed the required course does not, on the surface meaning of the words, entail that you really get the rank. If there are two applicants for one position, both having passed the training course, the institution will choose the best one - and it is assumed that there is really such a thing as "the best one." But such cases do not occur much in scarcity, because there are not enough candidates passing the course in the first place. All the ones who do meet the objective standards actually will get the rank if they want it. In scarcity practice, the necessary conditions become sufficient conditions.

Scarcity of qualified candidates implies enough high-rank positions exist for all of the few who are qualified, so one of the important incentives the institution can offer to lower-ranking individuals is a credible promise of career advancement conditional on meeting the published objective criteria. If you get the training, accumulate the experience, and take the tests and get high enough scores on them, then you are really allowed to expect the honour of promotion.

An important value embedded in scarcity theory and underlying the promise of career advancement is the value of procedural fairness. For an institution in scarcity there are always rules; the same rules apply to everybody; and everybody agrees that it is important and good for the institution to work that way.

Abundance

Another fly upon another wall, in some other hypothetical French Army headquarters, might overhear a very different briefing.

Almost all the men serving France as junior officers today could also do the work of senior officers if they had the opportunity. Good candidates for promotion are abundant. We don't promote them all, only because we don't need that many senior officers. That does not mean senior officers have a particularly easy job; only our corps of junior officers is just that good. Furthermore, every rank and even every position is unique. To the extent that some officers are better than others, we cannot predict who will be the best senior officers from their performance at lower ranks. And the difference between a good officer and a better officer is inherently a subjective judgment anyway, that we cannot measure in practice and maybe not even in principle. The ability to be a good senior officer is also not something we can teach. Because almost all candidates are good and our ability to evaluate them precisely is limited, it is inappropriate to waste effort purporting to pick out the very best ones beyond basic competence, which is easy to recognize. Our main problem when trying to fill a position is how to narrow down an unmanageably long list of candidates who are all good enough. It is necessary to apply arbitrary criteria to the list just to reduce its length; and given that necessity, it is appropriate to use promotion decisions for social goals, such as liberté égalité fraternité and justice for society. Vive la France!

When almost all candidates are good enough, and good candidates far outnumber the high-ranking positions into which they can be promoted, that is a factual situation of abundance. As with scarcity, it is possible for an institution to follow a theory of abundance: it perceives the facts of abundance and implements a response appropriate to those facts.

Although abundance is defined as a high proportion of candidates being good enough to serve in high-ranking positions, this theory also holds that the sheer number of candidates is large in comparison to the demand for them. There are so many candidates the institution cannot look carefully at all of them when making promotion or hiring decisions. Fortunately, it does not need to.

The abundance theory comes with assumptions that are to a large extent the opposites of the assumptions in scarcity theory. Abundance theory needs to solve a totally different problem and uses different concepts to describe that problem. Having almost all candidates good enough means that fine distinctions to determine which ones are better, no longer are significant. The concept of skill as a measurable quantity fades away; instead, it will be observed that the remaining differences among individuals, although real and important, are subjective, unpredictable, and do not seem to improve with training. Abundance theory holds that the greatest candidates for high office are born, not made, and also are hard to recognize early.

Job duties still need to be performed, and some people can perform them and others can't. But distinguishing bare competence at the yes or no level is not the problem which needs solving in abundance. The institution in abundance apprehends basic competence as easy to recognize, as well as being plentiful. The institution does not put much effort into defining objective criteria for evaluating basic competence, because that doesn't need much effort. Recognizing truly great individuals as distinct from the merely good enough might still seem like it could be valuable, but abundance theory understands the differences between good and great individuals as fundamentally outside the institution's capability to judge or predict - and therefore counterproductive to attempt.

There may even be a denial of the potential existence of great individuals at all, stemming from a focus by the institution on society instead of on individuals. "Great individuals" is seen as simply the wrong way to think about society and the institution's place within society. Such denial carried to an extreme is diagnostic of the "dissipation" state I will describe later, but even in ordinary abundance it is likely to appear in a less extreme form. Institutions in a state of abundance, especially those like universities whose mission directly includes the formulation of philosophical theories, are likely to adopt philosophical frameworks that deny objectivity as a general principle and in particular its application to evaluation of individuals for promotion or hiring.

Then having determined the non-existence of any important objectively real differences between already good enough candidates, the institution has no reason to measure something that doesn't exist. And even if one might disagree with that determination, the fact that almost all promotion candidates really are good means it is harmless for the institution to deny the existence of job-relevant measurable distinctions. The denial does not hurt the institution at all; they have plenty of good candidates anyway. Institutions in abundance do not need objective criteria beyond basic job duty competence.

Rejecting the comparison between good and better candidates also means a much reduced emphasis on training and experience. Training and the resulting credentials may be necessary to look at as a purely technical matter, but are not seen as the main key to solving the real problems in hiring and promotion. Experience is not highly relevant to promotion because the differences that may still exist among candidates are things they bring with them as parts of their identities, not acquired over time.

The institution in abundance perceives the real problems of hiring and promotion as how to narrow down an excessively long list of candidates, all of whom are good, and more broadly how to use that narrowing-down process to create the best outcomes for the larger society. It cannot promote everybody. It has to reject some. How will it choose whom to reject? Objective skill related to job duties is not a problem that urgently needs to be solved; measuring it may be impracticable anyway; and may be impossible, by philosophical definition. So the institution will turn to solving other problems. Seeing its own mission in the context of the broader society, it will attempt to make its hiring and promotion decisions part of making the world a better place.

Abundance as such does not specify which problems in society need to be solved, only that it is the institution's job to help solve them, whatever they are. In France in the first decade of the 20th Century, political division was high on the list of problems, so the Army would make an effort to demonstrate that its promotion decisions were fair to both sides of the ongoing republican versus nationalist conflict. The Affaire des Fiches of 1900 to 1904, overlapping the later years of the Dreyfus Affair, showed the cost to the Army of choosing the wrong issue. The Army had been maintaining a secret card file on officers, using information collected through Masonic lodges, to keep track of which officers were thought to have republican as opposed to traditional political or religious views, giving promotion preference to the republicans. After the resulting scandal the Army was forced to go to lengths to demonstrate its even-handedness. But even the pre-scandal fiches system fit the typical abundance-theory pattern of using hiring and promotion to solve problems in the larger society. The Army institutionally believed that nationalism was a problem in France that needed to be solved, and - given almost all promotion candidates were good anyway - it chose which ones to promote in such a way as to help solve that social problem. It just made a choice that proved to be unpopular of which social problem it should be helping to solve.

The selection of which bigger issues in particular to address does not come from the abundance theory; only the imperative to use hiring and promotion to address some bigger issues. But the issues selected will often happen be ones that can be cast as some sort of equity, because equity as a general principle is directly supported by abundance theory. The fact of almost all candidates being good enough pushes the institution toward the idea of almost all persons being good enough, and therefore toward adopting more general equity principles.

I wrote that scarcity theory embeds a value of fairness. So does abundance theory. But fairness in abundance is a different kind of fairness from that in scarcity. An institution operating on a theory of abundance will prioritize fairness in the sense of equity, fairness in the results for individuals and for society as a whole. It may be seen as fairness with respect to consequentialist ethics (results-based) or on an ethical concept of virtue (inherent goodness of people), as opposed to the fairness with respect to deontological ethics (rule-based) that is given priority in a theory of scarcity. Objective, written, and general rules of fairness in an institution following abundance are tools for achieving fair results for good people, and if the consistent application of a general rule would conflict with fair results in the case of a good person, then that rule ought not be followed in that case.

More about Dreyfus and Picquart

Dreyfus and Picquart were each promoted when they were brought back into the Army, and I am interested in the specific details of how far they were promoted, and why. I'd like to know which of my "scarcity" and "abundance" theories better describes how the institution of the Army handled the situation, and whether or not either theory describes it well at all. Here are some relevant factual points, some of which make the issues I have highlighted less significant in the historical event.

• The decision to promote Dreyfus and Picquart at all was officially made by a civilian political institution, the Chamber of Deputies, which had different priorities from those of the Army.
• The stated intention was to restore their careers: to put these two men where they would have been if the Affair had not happened.
• The call of exactly how far to promote them was made by the Minister of War, Eugène Etienne, who was a career politician and never served in the military himself, but may have had a perspective more focused on the Army's institutional needs than the Chamber of Deputies had as an institution in itself.
• The promotion Dreyfus actually received was only a small one, from captain to major effective on the date of the decision. This may have been the smallest promotion the Army could give and still say it was promoting him at all.
• Although the "what if" question is hard to answer precisely and convincingly, it is clear that the promotion given to Dreyfus was not really big enough to achieve the claimed restorative goal.
• Dreyfus's projected career path within the Army was in the General Staff, an important but basically bureaucratic role; not a combat role. The argument against his promotion because of missing experience might be claimed less strong for this reason.
• The missing-experience argument might also be less strong because the promotion Dreyfus received was so small; his new job duties would not really require much additional skill or experience.
• Dreyfus consistently refused to accept money as compensation, which had he accepted it would have made matters simpler for the Deputies and the Army. His refusal of money removes some money-related questions (like the amount of back pay due for his deemed service in the "what if" scenario) from needing to be answered.
• Dreyfus was admitted to the Legion of Honour when he rejoined the Army, but this was nearly automatic for officers at his new rank, and it was not seen as a special distinction beyond the rank promotion. Picquart was already a member before the Affair.
• Picquart's promotion was bigger than Dreyfus's, from lieutenant-colonel to brigadier-general back-dated by three years. This two-rank promotion, for a shorter career interruption, was seen as more in line with really achieving the restorative goal. It creates a comparison between the two cases that makes the treatment of Dreyfus look even worse; and it may more strongly raise the missing-experience issues. But much less detail on the specifics of Picquart's case is available to me today.

To analyse the extent of the promotions, it is important to have a clear idea of how ranks worked in the French Army at the time. Here is a summary of the English-language rank titles that are relevant to the Dreyfus and Picquart cases, from highest down to lowest:

• Brigadier-General
• Colonel
• Lieutenant-Colonel
• Major
• Captain

There were often multiple different French titles for a given rank level, depending on which branch of the Army the officer served in, such as artillery or infantry. Ranks as such, the headings on that list, represented big steps in prestige: any major was significantly above any captain. But for officers at the same rank, seniority mattered, and was measured by time since achieving the rank. An officer who had been a captain for several years had more prestige, better pension eligibility, possibly better pay already, and would expect to be considered for promotion to major sooner, than one only recently promoted to the rank of captain.

All those mentioned above are commissioned officer ranks. They entailed tertiary education, and the men in these ranks (and they were all men) were officers and gentlemen, considered to be in a distinct and high social class even in civilian life. Young captains like Dreyfus at the time of his marriage in 1890 were not allowed by official regulation to marry women who had less than a certain minimum amount of family wealth as measured by the value of their dowries. In fact, Dreyfus's wife's family was far richer than the minimum; but this bar was set high enough to be a real obstacle to marriage for some of his peers. The point was to maintain a public perception that military officers were a cut above ordinary people. The democratic ideal of eliminating class from society, given at least lip service in the founding of the Third Republic, was very popular in some quarters, but also very unpopular in others, of French society at the end of the 19th Century. Levelling of society was unpopular within the military.

It does not seem that just promoting Dreyfus from captain to major after twelve years really represented how far he would have advanced in a normal career. At the time of his arrest he had been considered an up-and-coming young officer with good prospects. Someone like that could reasonably have expected promotion to major in less than twelve years, quite possibly to lieutenant-colonel within twelve. The questions of how good his prospects at the start of the Affair really were, and how good a career ought to have been hypothesized in the what-if game, are not easy to answer.

Dreyfus had mostly good reviews from superior officers, and some bad ones, in his file at the time of his arrest. Latter-day writers are quick to dismiss all negative comments from his contemporary evaluators as simple anti-semitism, but that may not be an accurate explanation. His ranking (referring to educational achievement, not authority in the Army) was 9th out of 81 graduating officers when he left the École supériore de guerre, as mentioned below, a pretty good showing; but focusing on that particular ranking smacks of cherry picking, because similar rankings at earlier stages of his education had several times put him below median.

It might be argued that his being part of the General Staff - basically an office paper-shuffling job - weakened the missing experience argument against Dreyfus's promotion. I wrote above about the concerns of a soldier who might not want to be commanded in battle by an inexperienced officer. In fact, General Staff officers would be unlikely to directly command soldiers in battle at all. But using that to excuse a questionable promotion amounts to saying General Staff officers have less important jobs than other officers, and don't need real skills to do their jobs; probably not a true statement, and certainly insulting to them. Furthermore, Dreyfus's training was designed for combat command roles, his rank made him eligible for combat command roles, and he did end up in such a role in the First World War once it began. If majors in general had jobs requiring experience, then Dreyfus as a major needed experience too.

Here is a quote from Alfred Dreyfus himself, in a diary entry at the time of his retirement in 1907. The translation is from Whyte's chronological history of the Dreyfus Affair, and the internal quote at the start (not clearly delineated in Whyte) is from the draft of the Chamber of Deputies bill authorizing the promotion.

The government is powerless to repair the immense harm, both material and moral, which the victim has suffered from such a deplorable miscarriage of justice. It wishes at least to replace Captain Dreyfus in the position he would have reached if he had pursued the normal course of his career.

Yet, on 13 July 1906, more than one hundred artillery officers, less senior than me as captains, were already majors. In appointing me a major, the law recognized that with my ranking when I left the École supériore de guerre, my personal dossier, the same would also have applied to me. Consequently it would have been equitable to give me my legitimate position in seniority among the officers of my weapon-section.

On the date of 13 July 1906, M. de Laguiche, an artillery officer of notably less seniority than me as a captain, was already a lieutenant-colonel. But I had not sought an exceptional promotion.

Among the hundred or more majors of lesser seniority than me as captains, the first promoted to commander [sic] was on 12 October 1901. It would thus simply have been fair to give me, at the time the law was voted, the same seniority as major.

The word "commander" in this translated quote does not make sense. That word in English refers to a naval rank that might be equivalent to lieutenant-colonel in the army; or generally to any officer in command of a given unit, regardless of rank. As far as I can tell there was no specific rank called "commander," or a French term that would be well-translated to English as "commander," in the French Army at the time. I think the only interpretation that really makes sense is that in this sentence Dreyfus wrote commandant, which was one of several French terms for the rank usually translated as "major" but which could also be used untranslated in English; that is, the same rank to which Dreyfus was promoted on reintegration.

I think writing "commander" instead of "commandant" may have been an error, perhaps not in the translation but in the editing of Whyte's book. I have found several other apparent editing errors in that book. So far I have not been able to obtain the original French text for this passage, but I am still trying, and when I have that, I may update this note. It makes an important difference because if somehow "commander" in the translation actually did mean "commander" in English, then the suggestion that Dreyfus should have "the same seniority as major" as someone at a rank equivalent to lieutenant-colonel, would be confusing and would imply some other theory of how to do the calculation.

Assuming the reading "commandant," Dreyfus is saying that he wants to be promoted to the same relative position in seniority order among his peers at the time of his rejoining the Army, that he was in when he left. That is a reasonable, though certainly not the only reasonable, way of interpreting the stated intention of the Chamber of Deputies to put him where he would be "if." He is defining his seniority in terms of who is below him: the comparison group is captains with less seniority than Dreyfus at the time he was kicked out of the Army. None of those men should end up above him in seniority now. As a compromise he limits it to those captains who only made it to the rank of major, excluding the redoubtable de Laguiche and any others who may have been promoted to lieutenant-colonel. Insisting that the Army really should promote nobody from below Dreyfus to be above him during the lost years of his hypothetical reconstructed career, not even de Laguiche, would mean Dreyfus also demanding the rank of lieutenant-colonel (two ranks of promotion instead of one) upon his return.

I have a lot of sympathy with treating the focus in the "what if" scenario as "don't promote more junior officers past him." Having somebody one considers lesser promoted past one, hurts. It has happened to me. There's a university where, if they ever wanted to hire me (won't happen, but), well, they would damn well have to put me higher on the organizational chart than Professor so-and-so who was promoted past me years ago without deserving it. But that approach to justice is very much focused on the individual, on giving someone what he wants and has morally earned, and it may really be more a matter of retribution than restoration. It may not be the right answer to how to deal with the Dreyfus case from the point of view of the institution of the French Army. The idea that it is vexatious to promote someone more junior past Dreyfus, considering him relative to others, does not really address at all the question of his absolute fitness for duty: whether years of rotting in prison, and more years in civilian life without performing Army duties, really gave him the skill to serve in the role of a higher-ranked officer.

A different way to do a similar calculation, still relative to others but much less favourable to Dreyfus, would be to consider other officers who were above him when he left, as the comparison group, and then promote him as far as possible without passing any of them. Very likely, at least one of his peers defined that way was never promoted at all - and then Dreyfus would not be promoted, either. I cannot strongly argue that that calculation makes any less sense than the calculation Dreyfus asks for; it is just the same thing with a different arbitrary choice of the comparison group.

If I were trying to design a general rule for procedural fairness in this kind of case, given we agreed on this kind of relative calculation in the first place, I think the rule that might really be fairest would be to combine them. Pick a representative comparison group including officers who were both above and below Dreyfus when he left, and still in the Army themselves when he returned. Then promote him to the same order statistic of rank and seniority within the comparison group as when he left. That is, he should end up with the same number of men in the comparison group junior and senior to him as before, even though probably not the same individuals because some would have been promoted faster or slower than each other. If anyone passed him, this calculation would force it to be equal numbers passing him in both directions, which seems to keep him at the same relative level if the comparison group is really representative of his peers.

The order-statistic procedure would probably give him a little less seniority than what he asked for, but still significantly more than he was actually given. It still has a theoretical bias in his favour because of not counting officers who might have left the Army. There's an interesting question in how far the "what if this Affair had not happened?" scenario should factor in unlikely possibilities that could occur in 12 years of an officer's career: like his dying unexpectedly, committing some other crime and being punished for that instead of the false espionage charge, doing unusually well and being promoted beyond the rank of major, and so on. The answer to that question would bear on who should be included in the comparison group.

I agree with his position, as far as this point goes and if we agree that promoting Dreyfus at all was the right way to give him justice, that on one calculation basis or another, he should have received more deemed seniority with his promotion than just the day of the Chamber's decision - especially because they gave Picquart extra seniority on what ought to have been the same moral theory. Procedural fairness: the same rules should apply to everybody.

Picquart received a promotion of two full ranks, plus seniority. He had been a lieutenant-colonel when he left the Army in February 1898, and was reinstated as a brigadier-general in July 1906, skipping past the rank of colonel, with his seniority as brigadier-general counted from 10 July 1903. By doing that the Chamber opened the door: they made it clear that they recognized the seniority date as relevant to the restorative goal, and as a thing they could adjust in order to achieve that goal. They also created an inevitable comparison between the two recipients. Multiple commentators said that the decision ended up being fair for Picquart but not fair for Dreyfus.

I have not been able to find a detailed explanation of the basis on which Etienne decided (or claimed to have decided) exactly how much rank and seniority to award Picquart. Although Picquart was kicked out of the Army by forced retirement in February 1898, the Army had been trying to get rid of him with postings increasingly distant from Paris as far back as November 1896; so the interruption of "the normal course of his career" might properly have been judged longer than just the time he spent officially retired, let alone his pre-trial imprisonment connected with the abandoned court-martial. In October 1906, Picquart was appointed Minister of War in the cabinet of Georges Clemenceau, succeeding Etienne.

Dreyfus did seek the additional promotion to lieutenant-colonel, applying to the new Minister of War Picquart for it in November 1906, a few months after his reintegration to the Army. The application was denied.

Burns writes that Dreyfus and Picquart were not personally friendly, despite their linked destinies. The Clemenceau government had benefited from its connection to the Dreyfus Affair, but Dreyfus's conservative political opinions and upper-class family background were not aligned with the new government's left-wing stance. They liked him better as a symbol than as a man. The Dreyfus family and Picquart had fallen out in 1900 over differences in strategy after the second court-martial, linked to personality conflicts between members of the family's legal team; and Picquart retained his general anti-semitism throughout the Affair.

I think Dreyfus may have believed that even though it was a bigger request, it would be more possible for the Army to give him the promotion to lieutenant-colonel in November 1906 than to change the deemed seniority of his promotion to major after the Chamber of Deputies had fixed that. He may also have thought that the Clemenceau cabinet, and Picquart as its minister, would see the what-if calculation differently from the previous cabinet. On a scarcity theory, if higher ranks do in fact require skills developed through training and experience, his spending a few months at the level of major might have allowed him to demonstrate that his skills really were good enough for the further promotion.

But the Dreyfus quote above dates from his retirement in 1907, and it may be his saying in 1907 that he would be satisfied with promotion to major with seniority represented a compromise from an earlier hope for restorative promotion all the way to lieutenant-colonel. It also seems significant that the promotion to lieutenant-colonel he requested through Picquart, if granted, would have been a recognition of him by the Army, and by a Minister of War with real military credentials, therefore more honourable; whereas his promotion to major was something imposed on the Army by the civilian politicians.

Money questions and specific performance

The Dreyfus family was wealthy to begin with, and Alfred had married up: his wife's relatives, the Hadamards, were even wealthier. He had access to some luxuries during his student days that most of his classmates lacked, making him subject to envy and resentment on the part of the other students. Later, his wealth fed into anti-semitic stereotypes and hatred. But it also meant that he did not really need his Army pay. That no doubt factored into his decision to refuse any attempt at compensating him with money for the mistreatment he had suffered. He always maintained that his goal was to restore his and his family's honour, and that it could not be done by writing a cheque.

If money had been more significant, then there would be calculations to make in relation to his hypothetical reconstructed career. For instance, we might go back and calculate the back pay he should have earned for each year of service, with his hypothetical rank at each point. That in turn would make the deemed dates of any intermediate promotions all the more significant; and then to really do it right we should also keep track of interest on the money, accounting for the fact that he was receiving it late. Because he categorically refused to accept money, all these questions are off the table and were never resolved.

In the English legal tradition there is a distinction between the jurisdiction of a court of common law, which will award only money to persons who have been wronged; and a court of equity, which can potentially award (among other possibilities) a remedy of specific performance. That is, the equity court can require somebody to do something, beyond merely paying up, in order to right a wrong. The term "specific performance" is now most often used narrowly to describe enforcement of contracts in particular, but I am using it more generally here to describe remedies for other harms as well as contract breach, which does reflect this term's history. Most courts in most places that inherit from English law today combine the "common law" and "equity" jurisdictions in a single court; there are just a few places which retain separate courts of equity. But even in courts that descend from the English tradition with a partial or complete equity jurisdiction, there is a strong preference for awarding just plain money if at all possible.

The Dreyfus case is a good example for why common law courts prefer to award money damages. There was really no specific performance that could have made him whole. Failing to promote him would have been unfair, and would have failed in the stated goal of putting him back where he ought to be; but promoting him also failed in that goal. They could give him the official job title of a more experienced officer, but they could not cause him to really be the more experienced officer he deserved to be at that point in his career, and without that experience, the title would be worth much less than it should. If only they could have just paid him off and had him go away! Although money also would not recreate his career, it would have fewer externalities, and the question of how much money ought to be paid, despite its obvious complexity would still be much simpler than the question of how to compensate for the aspects of specific performance that were impossible to fulfill.

But France did not (and still does not) follow an English common law tradition. The legal code system used in French courts usually prefers specific performance; and this was not really a court case anyway. The Chamber of Deputies was under no obligation to follow judicial customs in compensating Dreyfus. It is understandable that he would see mere money in any amount as an unacceptable substitute for the career he had lost, or even just those parts of a bright military career he could still have. It is also to be expected that the public would demand to see a greater gesture than money, and that therefore the Chamber in its efforts to placate the public would attempt to grant more than money.

When he retired in 1907, Dreyfus's official rank was reduced to that of captain. This technical demotion appears to have been customary for an officer retiring from the Army, and not seen as any dishonour. He was given a pension according to his accumulated years of service. See Appendix B for a discussion of exactly how many years of service that was. The retirement was not complete or permanent. Dreyfus became part of the "reserve officer corps." When the First World War began in 1914, the Army called up the reserves, bringing him back to the rank of major. He served in the garrison near Paris for most of the war but fought in the Second Battle of the Aisne in 1917. He was promoted to lieutenant-colonel in September 1918, shortly before the armistice in November of that year.

More about scarcity and abundance

Here is a summary of the attributes of scarcity and abundance, including a few new ones not yet mentioned in the discussion above.

I am hesitant to even mention gender here, because of the danger in the 21st Century climate that as soon as I mention gender at all, readers will perceive this entire article as being about gender and nothing else. This is not an article about gender. Gender was not relevant to the Dreyfus Affair, notwithstanding at least one 21st Century author's attempt to re-cast the entire Affair as about gender starting from the revelation that anti-semitism is actually just a form of misogyny.

This article is not even about the Dreyfus Affair, despite the thousands of words I have written on that example. This article is about institutions.

But given the important role gender did play in another example I want to use - the example of North American universities in the second half of the 20th Century - I would be remiss not to mention gender at all. The values of objectivity in scarcity and subjectivity in abundance are values many writers have often associated with men and women respectively. It is factually true that in North American universities, the time when the institutions operated under scarcity theory was the time when they served a male-majority population and promoted only men, and the time when the institutions operated under abundance theory was the time when they served a female-majority population and often promoted women. I don't think the correlation is only coincidental. For institutions in which gender is relevant, which are not all institutions, male power seems to be part of the pattern of scarcity and female power seems to be part of the pattern of abundance. This correlation feeds into my association of abundance and scarcity with the yin and yang lines of the I Ching, and is further developed in the cycle of facts and theories discussed below.

The media used by an institution also seem related to the institution's theoretical orientation toward scarcity or abundance. Institutions in scarcity with a heavy focus on rules and objectivity gravitate toward using forms of communication with definable meanings, that is, media that use words and are recorded. Because institutional functions occur over time and are not tied to individual human beings - the subjective human element in decision-making is deliberately avoided - there is a preference for asynchronous media. That is, media like written memoranda and (in the present day) email.

Institutions in abundance, conversely, will prefer media that maximize exposure of the subjective human element: those that are nonverbal or allow for nonverbal side channels, such as telephone and video, and certainly real-time media that force human beings to be directly involved with each other. Media that cannot be recorded, or merely are not recorded, are seen as allowing human beings to express themselves without the constraints of future retroactive rule application - which is seen as a positive benefit of these media in abundance and a terrible flaw in scarcity.

I would like to be able to say that the media difference between scarcity and abundance institutional promotion theories is the difference between "hot" and "cool" media described by Marshall McLuhan, but I just cannot make sense of his definitions. McLuhan writes that hot media are immersive media, those which do not require thought from the participants; whereas cool media are more distancing media, with lower sensory fidelity, requiring conscious thought of the participants. By those definitions as I just stated them, the institution in scarcity's preference is for cool media, especially written words: those media that force individuals to use their conscious instead of unconscious mental faculties. The institution in abundance, especially with its emphasis on forcing individuals' emotional states to be legible to each other through real-time interaction and side channels, has a strong preference for hot media like the telephone.

But that seems to be my abuse of McLuhan's terms, because he actually says written words are hot and the telephone is cool, just the opposite of what I write above. A possible explanation is that in his 1964 book he was referring to very specific forms of media exactly as they existed in Toronto in 1964: for instance, his hot radio meant the then-new high-fidelity stereo FM broadcasting, which was mind-blowingly immersive to persons hearing it for the first time, and his cool television meant over-the-air analogue black-and-white, which demanded much from the viewer's imagination. Another clue to "hot" and "cool" is that the difference between two-way and one-way media may be significant. The closely related real time versus asynchronous distinction seems most relevant to abundance and scarcity. That distinction is easier to make precisely, less subject to change as technology changes, and frees me from having to argue with a highly respected dead man.

An institution operating on a scarcity theory will allow, and even prefer, individuals to communicate across time, and to record it. An institution operating on an abundance theory will push them to use simultaneous and ephemeral two-way communication media. As I discuss in Appendix A, the Dreyfus Affair was interesting for its focus on written documents - non-real-time and associated with scarcity in my analysis - but they were handwritten, and much was made of the "graphological" side channels in those documents. How someone wrote a given document was seen as important beyond the words he wrote, and who wrote a document was central to the Affair. Identity emphasis is terribly characteristic of institutions in abundance; so is secret decision-making, another frequent theme in the entire episode. The use of media in the Dreyfus Affair points in both directions.

Now, stepping back and looking at the entire system as I have developed it up to this point, how can the Dreyfus Affair be understood in terms of these two opposing theories? Was the French Army during 1894 to 1906 really in a state of scarcity, or abundance? Do either of them explain well how the institutions behaved? Throughout my comments I have been highlighting the abundance theory in relation to Dreyfus and especially his promotion. Promoting someone as a way to give him back the value of twelve years of experience he never had makes sense in abundance, where the broader social goal is important and the projected negative consequences of such a promotion are not. I think that is the answer: the Chamber of Deputies promoted Dreyfus because they were following a theory of abundance as I have described it, and under that theory, it was the right decision. At the very least, they were playing to an audience that believed abundance theory; the public perceived a high-ranking Army officer as just a more honoured low-ranking officer. If the abundance theory accurately reflected the facts, then they made the right call.

But there are many points in the Affair more consistent with scarcity as I have described it. The Army, as opposed to the Chamber, seems to have operated in many respects more consistently with a theory of scarcity. That may explain why they did not promote Dreyfus much; the Chamber's stated intention was not achieved nor even attempted in good faith. Most of the institutional attributes I link to scarcity theory are attributes of the French Army at the end of the 19th Century. The conflict between scarcity and abundance may parallel the conflict that existed between the Army and the civilian political institutions of the time. Maybe the political institutions were using abundance while the military institution was using scarcity.

It seems clear that Alfred Dreyfus himself was steeped in scarcity theory. He was an officer's officer, with personal values closely parallel to those of the Army institution. He may have been willing sometimes to make appeals to abundance concepts in order to get from the institutions the honour he thought was owed to him for his merit, but he surely was keenly aware of what would be missing from that remedy. Remember that it is institutions in scarcity who make promises to individuals. The very idea of being owed honour for one's personal merit is fundamentally a scarcity concept, expressed in scarcity language.

The conflict between scarcity and abundance theories leads to my next few points. What happens when scarcity and abundance are mixed in the same institution?

Inflation

Suppose there is an institution in the factual situation of abundance. It has plenty of candidates for hiring and promotion, and almost all of them are good. But this institution is operating on a theory of scarcity. It is using processes and policies designed for a factual situation in which most candidates are bad. I call such a combination "inflation." Inflation is misunderstood abundance. It may be linked to what some writers have called "elite overproduction."

The institution following an abundance theory correctly, in a factual situation of abundance, will knowingly turn away many good candidates. It will treat that as a necessity, perhaps unfortunate, but an opportunity to fulfill larger policy goals. The institution in inflation, using a scarcity theory in a factual situation of abundance, will also turn away many good candidates. But the institution in inflation will turn away good candidates mistakenly believing that they really are bad - at least in comparison to others, with an unquestioned assumption that comparing good candidates to each other is possible.

Institutions operating on a theory of scarcity use written objective criteria for promotion and hiring. Confronted with the fact of a huge surplus of candidates meeting the criteria, the scarcity theory says that that means the standards are too low and should be raised. An institution in inflation will raise its standards unreasonably high, while keeping them written and objective.

Often, the inflation effect shows up with respect to educational credentials in particular. It goes all the way up. A job that uses only skills taught in elementary school, will require a high school diploma. A job whose duties are taught in high school will be advertised as requiring a Bachelor's degree. A job that could be done by any PhD holder, will only be open to those with postdoctoral fellowships too, and many single-author papers in top-ranked venues.

Note that although educational credentials are a good example of something objectively measurable that can be subject to inflation, credentials are not the only example. The same thing can happen along other dimensions, even something so prosaic as body weight for a job where that is relevant. Too many good applicants, and the requirements will retreat beyond reasonable limits.

Because scarcity theory deeply embeds the idea that most candidates are bad, the institution that inflates its objective criteria to narrow down a huge pool of candidates will not perceive itself as introducing an arbitrary filter just to narrow down the pool. The institution is more likely to perceive the issue as something wrong with the validity of credentials, an inflation of the candidates' scores rather than an inflation of its own demands. High schools aren't what they used to be. They're graduating too many kids who can't read. We need to require a Bachelor's just to be sure candidates will have everything it says in the high school curriculum. Or even, yeah, well, in that country they hand out PhD parchments like lollipops, so we need to see publications too, and they'd better not be in Hindawi open-access journals.

Another way an institution in inflation may respond to abundance facts is by creating new positions. Because it is following the theory of scarcity, it assumes that most candidates are bad; but hard-to-ignore clues suggesting the factual existence of many good candidates may be interpreted as only meaning that the entire pool of candidates, and the institution's world, are growing. Then the institution tries to grow with the world. Because the real issue is structural - an underestimated proportion of good candidates - the attempt at growth will not be enough. The institution grows with the total pool of candidates, but the also-growing proportion of good candidates means the number of good candidates grows faster than the total pool. Institutions that produce their own candidates (such as universities) also produce more of those as they grow, creating a feedback loop that neutralizes much of the benefit. From an individual candidate's point of view, opportunities for prestige and recognition will seem to be vanishing despite an increasing raw number of positions.

Institutions that believe in scarcity perceive a need to motivate candidates to seek promotion; and the institutions do this partly by making promises. If you reach this objective standard, says the institution, then you are really allowed to expect this promotion. They also promise money and status to the ones who do get promoted. In true scarcity, these promises are easy to make and easy to keep. But in a state of inflation, the promise of promotion for every qualified individual cannot possibly be kept. There are not enough positions for them. And the institution that tries too hard to keep that promise, creating new positions it does not need, will run out of money and status to give the holders of those positions, and will end up breaking the other promise. Okay, you got your high-ranking job title, but your responsibilities and what your salary can really buy are much less than you expected. You are a victim of inflation.

When the factual situation is one of abundance, it is absolutely necessary for the institution to introduce arbitrary criteria, just to narrow down the list of candidates to manageable size. That is an opportunity to use the hiring and promotion criteria for broader policy goals, but only if the institution understands it is in a state of abundance. Inflating objective written criteria far beyond the levels where they are actually relevant, is the inflationary institution's reflexive attempt to thin the herd without apprehending that it is introducing irrelevant arbitrary criteria out of necessity. The equity goals that might be achieved in correctly-recognized abundance, will not be pursued in inflation.

Institutions following scarcity fundamentally believe that relevant objective criteria are appropriate, necessary, and even possible. This belief is embedded so deeply in scarcity theory that it is unconsciously applied everywhere, and never examined directly. Really, in a factual situation of abundance, objective criteria are not very important and may not be possible or desirable. The real differences that make one candidate better than another beyond basic competence are not things the institution can measure; it should not try. Furthermore, in abundance theory the differences that make one candidate better than another beyond basic competence turn out to be inherent qualities that cannot be improved with training. So effort spent on training in particular, trying to change good enough candidates to make them better than good enough, is wasted nearly by definition; that just isn't a correct way to think about the differences among candidates.

Versions of "basic competence" inflated to a level where a majority of good candidates will fail to meet them, become counterproductive. Searching for more and tougher objective criteria to use when there are too many good candidates making it through - misperceived as too many bad candidates making it through - leads institutions to start applying criteria that don't work and even correlate negatively with basic competence, but at least they produce numbers, damn it!; because scarcity theory demands numbers. Consider student bubble-sheet evaluations of teaching performance, which are the most inflationary method of job performance evaluation in human history.

Applying inflated training-based filters costs the candidates, the institution, and society at large far more than other filters that could work just as well. How much does it cost the world to put entire generations through four-year college degrees that they don't need, burning years of lives and massive quantities of money, just so an HR department will have what it can say is an objective basis for throwing out a percentage of resumes for a job that really only requires high school? Rolling a die would work just as well, would be fairer at both the societal and individual levels, and would be a lot cheaper. That should be done instead! - but only if the facts really are facts of abundance.

The institution in inflation ends up discarding candidates who really are at least as good as the ones who made it past the filter, and certainly good enough. It has a terrible human cost for candidates. They put great personal investment of time, money (often borrowed), and foregone opportunities just to meet the published objective criteria for career advancement - and then the promise is broken. When they reach the level of training and objectively-measured skill they were aiming for, it is no longer enough. The new standard is inflated out of reach. Or, maybe the official published standard is still the same, but they just don't get promoted, or don't get interviews, and there is no obvious reason why not.

These people believe in scarcity theory themselves, so they believe it is meaningful to say that some candidates are objectively better than others. If they don't get hired or promoted themselves, it must be because they are among the bad ones after all. And it's too late to do something else with their lives and do really well at it.

The institution in inflation is constantly shedding burned-out candidates who genuinely qualified for promotion but did not meet the unreasonably high arbitrary standard to actually get it, many of those for every single candidate who does get promoted. They go out looking for positions elsewhere, but with unrealistic expectations and many lost years of inapplicable experience; an entire social class of Dreyfuses. It's not clear what can possibly be done with, or for, persons who received expensive and time-consuming training to take up promised jobs that don't exist. Parts of their lives have been wasted.

But the broader society is still likely to call the victims of institutional inflation "privileged" or "elite" and expect gratitude and service from them. Words from Psalm 137:3: "they that wasted us required of us mirth"; the subsequent verses are graphic in describing the psalmist's righteous anger. Persons placed in this situation are likely to have a destabilizing effect on society as they try to take back what they perceive as having been taken from them. That is an externality: a cost of the institution's misguided decisions, stemming from the incorrect application of scarcity theory to abundance facts, that ends up borne by the larger society.

Dissipation

Suppose an institution is in the factual situation of scarcity. It has too few good promotion candidates to fill its high-ranking positions, because most candidates are bad. But this institution is operating on a theory of abundance. Its processes and policies assume a situation in which almost all candidates are good. I call this combination "dissipation." It is misunderstood scarcity, and the natural opposite of inflation.

Because the institution in dissipation is following an abundance theory, it perceives hiring and promotion as opportunities to fulfill policy goals that extend beyond itself into the general society. Moreover, because abundance theory states that almost all candidates are good, and residual differences between good enough candidates are institutionally unknowable, the institution in dissipation concludes that it doesn't really matter who gets promoted. Whereas the institution in true abundance knowingly turns away some good candidates along with the bad ones out of mere necessity, the institution in dissipation both turns away and promotes both good and bad candidates indiscriminately, mistakenly believing that to do so makes no real difference either way. Note that the very word "discriminate" is a bad word in abundance theory; the institution in dissipation concludes that that word's opposite must be good.

My description of abundance theory was peppered with the phrase "basic competence." Abundance theory in its best applied form limits the use of objective criteria to verifying basic competence - which is assumed to be easy to recognize, and is not highly salient because almost all candidates do have basic competence when the factual situation is one of abundance. But if the factual situation is actually one of scarcity, then basic competence is not anywhere near guaranteed in the candidate pool. The objective standards that would test for it become much more salient in scarcity than abundance theory expects they should be. The number of candidates failing the "basic competence" standard is much too high to be reconciled with abundance theory's core assumption that most candidates are good. And just as inflation responds by changing the objective standards, dissipation also changes the objective standards, lowering them or more likely eliminating them entirely.

The institution in dissipation believes it is turning away too many good candidates; in fact, it is not turning away enough bad ones. When an educational institution removes standardized tests completely from its admissions regime, having observed or believing it has observed that they were keeping out too many good students, that is a clue dissipation may be in effect - especially if the decision is announced using the language of equity, the characteristic language of institutions operating under an abundance theory. But an institution that only lowers the standardized test scores demanded for admission, from previously stratospheric levels without eliminating them entirely, may simply be recovering from an episode of inflation during which objective standards were raised beyond usefulness. These cases can be distinguished by the theoretical justifications put forward for the changes.

Abundance facts necessitate the implementation of arbitrary "filter" criteria on promotion, just to narrow down the infeasibly long lists of good candidates. Abundance theory recognizes that necessity, but is at risk of concluding all criteria are therefore no more than arbitrary filters, all equally so. In a factual situation of scarcity, there are so few good candidates that it is a serious problem to lose any; so the arbitrary filters implemented by the institution with an abundance theory will end up turning away exactly those candidates it most needs. Faced with facts that necessitate fewer or more open filters, the institution that believes all requirements are equally arbitrary will be as likely to drop a requirement that was really basic competence, as to drop a requirement that was really purely arbitrary.

Moreover, it is worse: the institution will preferentially eliminate the requirements that are most important. In a scarcity situation, candidates need relevant training and experience to do their best. Those will be the first requirements dropped, because they are expensive for both the institution and the candidates.

In a factual state of scarcity it is not enough that candidates just start out as good, virtuous people. Identity is insufficient. Training and experience matter. The institution with an abundance theory will not encourage candidates to gain training and experience, as an institution with a scarcity theory might. Instead, it will focus in all its public communications on the social goals connected to identity that it thinks it has the luxury of pursuing. If against a personal belief in abundance theory candidates believe in a possibility of upgrading their own suitability for advancement at all, they will act accordingly with what they are told. They will establish the identities the institution says it most values. Mission-relevant training and experience will simply no longer exist in the candidate pool once the institution has made clear that even basic competence is unnecessary and higher skill levels will be unrewarded. Even mild scarcity facts will quickly change into extreme scarcity facts once this dissipative feedback loop is in effect to disincentivize experience and training.

A powerful and important institution operating on an abundance theory will even exert a kind of gravity that distorts other institutions. If employers considered "elite" start to drop basic competence requirements, then universities which were at great cost training students to meet those requirements, will take note and start serving, instead, whatever they perceive as the employers' real requirements, perhaps involving identity and tribal signalling. Once dissipation sets in on the part of one powerful institution, it can easily spread to others.

As with the frustrated scarcity theory of the state of inflation, the failing abundance theory in the state of dissipation responds to conflicting facts by becoming more extreme, not by finding a compromise between itself and the facts. External social goals, which are seen as a fortunate addition to the institution's own mission when true abundance applies, become more prominent in dissipation. They replace the institution's own mission.

The institution in dissipation will have no mission of its own; it is captured by external interests. If someone could write down a clear description of a dissipative institution's mission as it perceives itself, this would be just a list of generic talking points, the consensus beliefs of all good people in the time and place. Those talking points could equally well be the mission of any other institution. It would be impossible to read the mission statement and know which institution it came from, because the institution's identity has been dissipated.

Witchcraft as an institution is discussed in my 2019 article on bringing in the idols. There I highlight how (at least for some participants) witchcraft's mission was entirely defined by external social goals that could not be distinguished from those of other institutions. It lost any unique identity, which is what happens with an institution in dissipation.

The institution may change its name, logo (or flag, coat of arms, etc.), and other external branding during dissipation. If so, the changes will not be cast as reflecting a change in the institution's mission, but rather as aligning the institution's image with what was really the institution's true mission all along; the long overdue correction of ancient errors. These announcements will be made using generic talking points with no direct relevance to anything of the institution's own ("We are shocked by the recent revelations about historical mistreatment of Group X and it is incumbent upon us to do our part..."), and the language of equity fairness.

Persons who lack basic competence and are promoted into high-ranking positions will have at best a neutral, and often a net negative, effect on the institution's ability to accomplish its mission. Abundance theory confronted with the effects of promoting too many bad candidates, but disposed to insist that almost all candidates are good actually, will accommodate the observed facts by concluding that the institution is just simply promoting too many candidates in general. Where inflation leads the institution to grow, dissipation leads it to shrink and flatten. Necessary high-ranking positions that the institution really needs, will be eliminated. And losing the ability to accomplish its original mission will reinforce the theoretical conclusion that that wasn't our real mission anyway. Under complete dissipation there is nothing left except the external social goals, in the empty shell of an institution incapable of doing anything to achieve even those.

The best real-life examples of dissipation current in 2021 seem to be the World Health Organization and the US Center for Disease Control. At the height of a century-level contagious-disease disaster, exactly the event that places the greatest possible demand on the actual missions of these institutions, instead of doing their jobs they were policing inclusive language and "teen vaping." They were captured by off-mission social initiatives.

The effect of institutional dissipation on individuals is less obvious than the effect of inflation. Inflation creates a dangerous underclass of individuals prepared for elite status who have no positions to go into; but dissipation causes individuals with necessary competence to simply never have existed in the first place. In factual skill scarcity conditions, skilled individuals are not born but could be made; and with dissipative institutions, skilled individuals are not made. In that respect dissipation entails less cruelty. But it is still damaging to society.

The victims of dissipated institutions may be visible by their absence and the absence of the work they should have been doing. Whenever someone asks, "Why don't we build things like that anymore?", it's possible that the answer is because the institution responsible for building such things was captured by other interests. When someone asks "Why is that other country out-competing ours?", it's possible that the answer is because our institutions devoted themselves to changing our society away from being one that could compete in that field. I wrote that large institutions in dissipation exert a gravitational effect; in fact, dissipation is a metaphoric black hole. It is evident by what it removes from the world.

My paraphrase questions highlighting the missing victims of dissipation are written to imply a change over time. Dissipation is rightly not perceived as just a static condition in itself. It had to come from somewhere, and it has to go somewhere. Actually, other states I have described also entail change. The entire system is always in the process of becoming something else.

The cycle of scarcity and abundance

Each of the four conditions I have described - scarcity, inflation, abundance, and dissipation - is impermanent. Inflation and dissipation are inherently unstable; they cannot remain in effect for long because they are of such a nature as to destroy the institution. The institution will either move away from these conditions or indeed be destroyed. Conditions of scarcity and abundance, being internally consistent and not inherently destructive, seem more stable. But each is vulnerable to change in the external realities of the world that embeds the institution. If the factual situation of scarcity changes into abundance then scarcity theory drives the institution into a condition of inflation. If the factual situation of abundance changes into scarcity then the institution adhering to abundance theory is driven into a condition of dissipation. And institutions with power to affect their world are likely to actually create these changes in the factual circumstances.

Because each will eventually change, and there is a definite direction in which it will change, the four conditions of scarcity, inflation, abundance, and dissipation can be understood as four phases of a cyclic pattern illustrated by this diagram (click for PDF enlargement).

The institution in scarcity is likely to be successful in achieving its mission. Despite the shortage of qualified individuals for hiring and promotion, its theory of scarcity correctly reflecting the facts causes it to offer incentives to individuals, causing them to seek and get the training and experience they need to serve effectively within the institution. But that effort will solve the problem it is intended to solve: the shortage of qualified individuals will become less severe, eventually disappear, and then invert entirely. The effort to train more qualified individuals is especially likely to overshoot the mark of merely producing enough, because it takes years to train them as well as more years for information about the changing facts to filter through to youngsters just deciding on the plans of their lives.

When the factual situation of scarcity is no longer in force, being replaced by abundance, the institution is unlikely to immediately perceive that a change has occurred, and unlikely to immediately change its own theory of operation. Thus it finds itself in a state of inflation, defined as abundance facts with scarcity theory. In this state the institution may still be achieving its mission, but only inefficiently, because of the wasted effort both of the institution and of individual candidates to achieve unrealistic objective standards. It is producing significant externalities through a stream of burned-out individuals dumped into the larger community with needs, expectations, and promises owed to them, that can never be fulfilled. The institution is growing during inflation, but that is also unsustainable, because society's need for the institution's mission is not growing. Assuming the institution does not destroy itself at this stage, the best outcome is that it will correctly respond to the changed facts by adopting an abundance theory.

I wrote "women become men" on the diagram as one of the attributes of the inflation phase, particularly with North American universities during the 1960s through 1980s in mind. The arc of change from scarcity theory to abundance theory in those institutions was accompanied by the rise of feminism, and that was more than a coincidence of timing. Scarcity theory, in its desperation to attract qualified promotion candidates it perceived as rare, was opening up positions to women that had formerly been reserved for men. But it still imposed on them the same expectations that had been applied to men, and due to inflation, the objective standards were in fact steadily increasing. Women newly had the chance to gain status within academic institutions; they were being encouraged to seek status, because scarcity theory gives that encouragement to everybody; but women had to become like men, losing some of the attributes formerly associated with women as a group, in order to do it. At the time, they were still trying to advance in the "man's world" dictated by scarcity theory.

Obviously, this aspect of the inflation stage did not happen in every historically inflationary institution. For example, it probably was not a big force in whichever institutions of the French Third Republic may have been inflationary; but then again, feminism did exist in that society, too. The later role of women in North American universities was quite different, as they became a majority of both students and staff, as the 21st Century began, and as these institutions moved into the abundance phase. In that later phase it was the institutions that changed, rather than women being forced to change - especially because the institutions adopted abundance theory itself, with its connection to concepts and values that might be perceived as feminine.

In the abundance phase, the institution operates under a theory that correctly reflects the facts of a candidate population in which almost all individuals are good. Here, as in scarcity, the institution is likely to be successful in achieving its core mission, if only because that has become easy; and it has the luxury of also attempting to help the broader society in other ways. Objective criteria can be lowered to just a requirement for basic competence, and are no longer highly salient. The danger in abundance is that the institution can be captured by external interests. The extra social goals added onto the institution's own mission may replace the mission; and the arbitrary promotion and hiring criteria needed to keep the institution functional under abundance facts may replace objective criteria even at the level of basic competence. With the incentive for individuals to achieve basic competence removed, the proportion of good candidates decreases until abundance facts no longer apply, and so the abundance theory will no longer reflect the facts.

Once the factual situation has changed back to one of scarcity, as with the change in the other direction, the institution will not respond immediately. It first confronts the new scarcity facts with its existing theory of abundance, and this combination is the state of dissipation. I have fewer good historical examples of dissipation than of the other three phases. Much of what I have written about what dissipation actually looks like is speculative, driven by theoretical considerations.

It seems almost inevitable that just as the institution in inflation will eventually figure out that it needs a new theory, and turn to abundance, the institution in dissipation will be forced to update its theory to reflect the new facts, and will return to a scarcity theory. I have indicated this state transition on the diagram by an arrow labelled "Back to basics" institutional reform (?). But I don't know that that is really what happens, or that if it can happen, it is necessarily what happens.

The best example I know of an institution that seemingly did pass all the way through dissipation at least once, cycling back into a scarcity theory characterized by insistence on relevant skill competence, was the Imperial Chinese civil service, with its famous examination system. It underwent reforms during the 19th Century that did indeed look like a "back to basics" effort; but in the early 20th Century it was abolished anyway. That institution had such a long history that, if these stages really form a cycle as I surmise, it could reasonably have gone multiple times around the cycle. I don't know whether it did.

I have said that these four stages are a cycle but it cannot escape my or a reader's notice that this is a purported cycle with a clear starting point and with what may well be a clear ending point. Real cycles don't have those. I hope that the transition from dissipation to scarcity actually exists, both as a matter of general optimism and because being a cycle is a distinguishing feature of this framework I have developed. Other ways to consider the changes in institutions regard the stages of change as just sequential stages without asserting a loop back to the beginning. So if that cyclic behaviour exists and is successfully predicted by my ideas, it supports their validity. But asserting that this process is a cycle may be an illusion created by excessive eagerness to believe my theoretical claim about what causes the stages. The factual validity of the cyclic sequence may not be proven.

No going back

The inflation stage seems especially interesting at the present time, because of this stage's externality of creating burned-out former promotion candidates. As of 2021 very many institutions seem to be in, or to have recently passed through, inflation. It may be that the unprecedently available communications media of the 20th Century encouraged institutions' cycles to synchronize with each other, so that all institutions will tend to be in inflation at the same time instead of different institutions being in different stages.

There has been a lot written about "elite overproduction," which seems to be exactly a visible symptom of inflation. Scarcity theory holds that the institution needs more qualified individuals, implicitly that qualified individuals can be made, and it leads the institution to set up incentives for making a larger supply of qualified individuals. If that effort continues after it has succeeded in changing the makeup of the candidate population, the result is a waste of unneeded qualified individuals.

The waste of human beings is bad enough for society, but even worse for the individuals involved. They were promised careers of honour and distinction, using the concepts of the scarcity theory into which they have now been indoctrinated, if they could meet the objective standards; they kept their parts of the social contract; and then the promise was broken. What if anything can be done to give them back what was lost? What if anything can they themselves do to take back what was lost? The goals of honour and distinction are not only unavailable but have ceased to exist at all, when the factual conditions are those of abundance.

I started this article talking about Alfred Dreyfus and Georges Picquart, whose careers were disrupted in another way than by institutional inflation, but who still faced the question of how to pick up the pieces of a disrupted career. Their cases illustrate the problems with efforts to make up for lost, maybe promised and certainly expected, careers.

There was no amount of promotion that could have been right for Dreyfus. They could put him at a level reflecting his real skill after twelve years outside a normal career, and then the lost time would preclude his ability to achieve great things later. They could put him at a rank accurately reflecting the hypothetical career years he might have had, but they couldn't really give him the experience he should have had in those years, and without the real experience, the rank would mean very little. Just giving him a pile of money would have been easy, but neither he nor the public would have accepted that, and questions would still be unresolved. In the end, they opted for an unsatisfactory compromise that was not really any of those.

Picquart's case is instructive. He got a better promotion deal, but more importantly, he took a different path, becoming Minister of War, instead of just trying to resume his aborted career on the same trajectory as before. I don't know if that was what he would have chosen, if he could have chosen between it and an hypothetical world in which the Affair had never happened at all. From the perspective of more than a century later, it looks like the final result for him was good. And it is worth remembering that this opportunity only became available to him because of the Dreyfus Affair: without the political fallout from that, the government that appointed Picquart to Cabinet likely could never have come into power at all.

The point here is that there is no going back. Once someone has left their path for any significant length of time, there is no possibility of getting back onto the same path. The best anyone can do is find a new path, which could be as good or better but might not be. An institution that, by creating inflation, has wasted individuals' lives in this way, bears a responsibility to right the injustice. Cleaning up the mess it made itself should be the first social responsibility goal for an institution entering the abundance phase and recognizing that it has power to achieve social responsibility goals. But righting the wrong cannot consist of just simply giving people back their lost careers, belatedly keeping the original promises, because to do so is impossible. There needs to be a way of giving back something else, of sufficient value.

The idea of giving back, or of taking back, lost years of one's life is a major theme in my novel Shining Path. The path in the title is, as well as the name of a specific institution, exactly the path of life that cannot be rejoined after one leaves it. I wrote that more than a decade ago. "Elite overproduction" of the kind we apprehend in the year 2021 was less visible at the time. My own current understanding of the scarcity and abundance cycle is much more recent than that book. And I had other points to make in it, about other topics. But it does contain many examples, including both real historical situations and invented pseudo-history, of people losing their paths and attempting with varying success to find new ones.

In the early 21st Century, in real life, we seem to have entire populations who believe they were promised career paths that were then taken away. The existence of populations under an incentive to take back what they consider already theirs, seems to explain otherwise-surprising electoral results in the USA and the UK. I hope it will not have the same effects depicted fictionally in the final chapters of my novel, and it probably will not; that was deliberately a cartoon plot. But the Boxer Uprising, also discussed in my book, was a real episode of violence arguably associated with scarcity-theory promises broken to a large population. It may be more than just coincident timing that the Boxer Uprising took place during the Dreyfus Affair.

Shining Path also highlights the sex and gender aspects of broken career promises, which I have only briefly mentioned here. Real historical examples all seem to have involved populations of men, and often of young men in particular - like the Boxers. It is to young men institutions in inflation characteristically make and break these promises. My fictional plot involves similarly disaffected populations of young women and especially of female non-humans; but I have no historical precedent for that. Exactly what role gender effects may play in the real-life phenomenon of broken institutional promises remains unknown and unclear, at least to me.

Going back to an interrupted career is also difficult because of the hypothetical nature of the path one is trying to rejoin. It becomes necessary to treat the victims of inflation as if other facts applied than their real life histories; and that is no easier for being necessary. Sarah Parry discusses the prominence of "as if" thought in the modern world in her posting on The World As If, and I've also written about it in terms of a possible cognitive deficit - what happens when some people are unable to process hypotheticals? - in The Imagination Gap. The "what if" exercise may simply be doomed; and in that case, what is equitable for institutions and society to do instead?

Links to other models

I have been hesitant to mention gender and I am even more hesitant to mention the I Ching for the same reason: as soon as I mention it, it may become overly salient to some readers, and cause them to ignore my main points. I will comment only briefly. This is an article about institutions. This is not an article about Taoist mysticism. Nonetheless, I have included the circle of eight trigrams (bagua in Fu Xi's "Earlier Heaven" sequence) in the centre of my chart because it seems to line up well with the stages in the cycle of institutional development I have described, counting one trigram for each proper stage and also one for each of the transitions between them. Unlike other third-party systems I have tried to put in correspondence, this one captures the closed-cycle structure of mine, with all four of the major stages, and the same basic symmetry. Apart from the trigrams, the core abundance and scarcity concepts seem to map well to yin and yang, and the concept of "moving lines" well describes how abundance and scarcity tend to replace one another.

• 離 ☲ fire - glory - scarcity state, individuals advance through merit
• 兌 ☱ lake - satisfaction - supply of qualified individuals increases through education, facts change
• 乾 ☰ sky - expansion - inflation state, institution grows to accommodate qualified individuals it still thinks are rare
• 巽 ☴ wind - flexibility - institution recognizes the new facts of abundance and changes its theory
• 坎 ☵ water - rapid rivers - abundance state, institution adds external goals, individual distinctions minimized versus group identities
• 艮 ☶ mountain - stillness - external interests capture the institution, qualified individuals disappear, facts change
• 坤 ☷ earth - that which yields - dissipation state, loss of identity, institution shrinks
• 震 ☳ thunder - revolution - "back to basics" reform (?), institution recognizes the new facts of scarcity and changes its theory

The stages described by David Chapman in his familiar "Geeks, MOPs, and Sociopaths in subculture evolution" article seem to parallel at least the scarcity, inflation, and abundance stages of my model, especially in the way that the institution tends to grow larger but away from its initial mission as it moves to a focus on social goals. The subculture model is not cyclic - at the end, the subculture dies or goes mainstream instead of returning to the beginning - and it is a model of subcultures, which are not the same thing as institutions. It also is a model explicitly described as no longer valid in the 21st Century; Chapman writes that the days of subcultures are over, having ended around the year 2000.

In some of his other writings, Chapman discusses the stages of adult ethical development as formulated by Robert Kegan. Although Kegan's stages strictly refer to individuals, they have implications for the progress of societies as different development stages become more prevalent in the population. I do not think it makes much sense to assign a distinct Kegan stage to each of the four states in my cyclic model of institutions, but the fundamental distinction I draw between "abundance" and "scarcity" theories does seem to correspond well to the line between stages 3 and 4 in particular. The communal ethics of stage 3 sound very much like abundance theory; and the systematic ethics of stage 4, like scarcity theory.

Chapman cautions against the temptation to link stage 3/abundance with the political left and stage 4/scarcity with the political right. I would repeat that caution, and with it I'm inclined to reject an attempt to map my four institutional stages onto a Twitter-style four-quadrant "political compass" grid, even if the existence of two binary variables gives my model the same shape as one of those grids. The Kegan stages are formulated as a progression, and there are others before 3 and after 4, but another Chapman article on the bridge to metarationality discusses in detail how the 3/4 boundary in particular is permeable in both directions. It seems reasonable to me that institutions may cross back and forth over it multiple times, creating what I've described as a cycle.

When thinking about things in terms of the Kegan stages I like to also compare with the Leary circuits - a similarly-shaped, but different in content, model of human mental development well described by Robert Anton Wilson in Prometheus Rising. Leary circuits 2 and 3, which together cover the mental operating modes of present-day domesticated human beings nearly all of the time, both seem to map onto the scarcity theory in my model, with its focus on hierarchy (circuit 2) and objective rationality (circuit 3). The other two of the most well-attested Leary mental circuits, numbers 1 and 4, might be assigned to the abundance theory by default - and that could point to the reason for the arrangement to be cyclic, if abundance theory tends to lead institutions from the fourth circuit back into the first - but it does not seem to be an excellent fit.

Some closing thoughts

Starting from the Dreyfus Affair and the question of whether giving not conventionally earned promotions to Dreyfus and Picquart was really the right way to compensate them for the harm they had suffered, I've developed a model of two ways an institution may apprehend promotion and hiring: as finding the few best candidates in a factual situation of scarcity, or as thinning a list of candidates almost all of whom are good, to achieve broad social goals in a factual situation of good candidate abundance.

Scarcity theory with its heavy focus on objective promotion criteria best characterizes the theory that officially prevailed in the French Army at the time of the Dreyfus Affair. It also described the operation of North American universities in the early to mid-20th Century, with the exact timing of the peak varying across departments. Abundance, with its focus on the social context embedding the institutions, peaked in North American universities in the first few years of the 21st Century, timing again varying somewhat. As of this writing in 2021, popular news media outlets are good examples of institutions following the abundance theory.

If an institution's theory of abundance or scarcity does not accurately correspond to its factual situation, a dysfunctional state of inflation or dissipation comes into force. Inflation results when abundance facts are misunderstood by scarcity theory. North American universities went through this stage with a peak in the 1980s to 1990s. Popular genre book publishing - and its corresponding fandoms - did the same a few years earlier.

Dissipation is the converse situation of scarcity facts misunderstood with abundance theory. Some North American universities may be entering this phase in the 2020s, though it is not yet established across the board. The World Health Organization and the US Center for Disease Control are examples of institutions in a state of dissipation as of 2021 - probably having entered that state earlier, but it became obvious with the spread of SARS-Cov-2.

Scarcity, inflation, abundance, and dissipation seem to fall into a natural cycle as the factual situations change and the institution changes its theories to keep up. I am especially interested in the inflation stage, associated with so-called elite overproduction, because of its externality: the institution in inflation is breaking promises to individual promotion candidates, who burn out and leave it, entering the broader society under an imperative to take back what was taken from them.

Finally, I've discussed some possible links between this model and other work.

Bibliography

Online sources are hyperlinked in the text. For printed books, I decided not to use formal in-text citations because I don't think they would be very useful to readers. Readers interested in these books will more likely want to simply read them, not look up individual factual points in my article. I also want to discourage Wikipedia-style bad faith discussion tactics.

Burns, Michael. Dreyfus: A Family Affair, 1789-1945. HarperCollins, 1991. Narrative history of the Dreyfus family, spanning the Affair and giving it a lot of personal context.

Doyle, Arthur Conan. The Hound of the Baskervilles. First appeared in The Strand Magazine starting 1901. Available online through Project Gutenberg. Sherlock Holmes novel highlighting the public fascination with forensics at the time.

McLuhan, Marshall. Understanding Media: The Extensions of Man. Routledge & Kegan Paul, 1964. Short interrelated essays on specific media and concepts in media studies; notable for introducing the "hot" and "cool" media concept.

Skala, Matthew. Shining Path. Self-published through Lulu, 2020 (print version). Online and EPUB versions at https://shiningpathbook.com/. A novel of the near future, written in 2009, covering some of the same issues as this article in the form I understood them at the time.

Whyte, George R. The Dreyfus Affair: A Chronological History. Palgrave Macmillan, 2005. Detailed timeline of the Affair with pointers to other sources; many appendices containing background material, some of it newly translated.

Wilson, Robert Anton. Prometheus Rising. Hilaritas Press, 2016 (second edition; main text is from 1983). Summarizes Timothy Leary's eight circuits psychological model, with an emphasis on hypothetical future human evolution.

Appendix A: Graphology and document forensics

The Dreyfus Affair deeply involved documents: their content, authorship, and significance. Most of the legal and factual questions considered by the courts-martial and other tribunals were questions about documents. Dreyfus's job as an officer in the General Staff was largely a job concerned with handling documents. The crime of which he was falsely accused boiled down to selling documents to the German military.

The documents that became significant in the case had French nicknames that sound precious to today's English ear. There were the bordereau, the faux Henry, the petit bleu, and many others. Some important documents were forged; some were fraudulently altered to change their interpretation; some never existed; and some mysteriously went missing.

Although the issues got steadily more complicated over the 12-year period of the main events, everything revolved around the central question of whose handwriting was on a document called "the bordereau." The bordereau was or appeared to be a list of other documents containing information about the French Army, that a French officer was offering to provide to Colonel Maximilien von Schwartzkoppen, military attaché to the German Embassy in Paris. Schwartzkoppen tore up the bordereau and threw it in a wastebasket at the Embassy, and then a cleaner who was spying for the French General Staff collected the pieces and turned them over to her employers, who took it as indication they had a traitor in their own ranks and began searching for him.

Some controversy exists (and existed) as to just how secret the documents offered in the bordereau actually were. At least some of them may have been public information anyway, that the Germans could have obtained through above-board research without espionage. But it was generally accepted that a French officer who wrote the bordereau would be guilty of treason. Dreyfus was accused and convicted fundamentally for the crime of having written the bordereau. It was later solidly proved that he was uninvolved.

In the late 19th Century, forensics - that is, the scientific or pseudoscientific examination of physical evidence in crime investigation - was a new and exciting thing. The Lafarge poisoning case of 1840 had attracted public attention throughout France, had made famous the Marsh test for arsenic introduced a few years before, and was the first significant conviction of a murderer primarily on direct toxicological evidence. The first Sherlock Holmes story was published in 1887 and was a sign of the times. At the end of the 19th Century interest in the application of scientific methods to criminal investigation and the justice system was at a high pitch, not only among the police and the courts but also in the general public.

In the 1880s the French police adopted the anthropometric identification system of Alphonse Bertillon, which involved measuring different dimensions of a man's body, like leg and arm lengths, and using the numbers as indices into a filing system, confirming matches using mugshots. (I write "man" because it was only solidly applicable to adult males.) Although photography of arrested persons existed already, Bertillon invented the mugshot in its modern systematic form. Anthropometric identification allowed the police to recognize whether someone had been arrested before without needing to know his name, nor do an exhaustive search of many thousands of files. It was a great advance in police methods, even if supplanted only a short time later by the use of fingerprints for identification - in which Bertillon also did pioneering work.

Graphology (graphologie) was the then-current term for the scientific or pseudoscientific study of handwriting. With a high-profile criminal case involving handwritten documents breaking at a time and place where forensics was a big, exciting thing, graphology was highly salient. When the Statistics Section first became aware of the bordereau, they circulated copies of it among senior staff asking if anyone recognized the handwriting, without positive result. They identified Dreyfus as a suspect because he was one of the few officers thought to have access to the items listed in the bordereau, and because he was Jewish. They obtained a sample of his handwriting and asked an officer who had an amateur's interest in graphology to compare them. That was Major Armand du Paty de Clam; he thought there was enough similarity to warrant a further investigation, so experts were called in - including Bertillon.

Bertillon did not have professional qualifications in the field of graphology in particular, not even by the standards of the late 19th Century. But I want to emphasize that he was the leading expert of his time, by the standards of the time, with respect to forensic science in general. In The Hound of the Baskervilles, published starting in 1901, Sherlock Holmes is annoyed to be described as "the second-highest expert in Europe," after Bertillon; and that mention in fiction conveys how Bertillon was regarded. His evaluation of the handwriting samples was that they were definitely a match, despite surface differences. A professional graphologist named Gobert, from the Bank of France, was also consulted at the outset of the investigation. He gave a more equivocal opinion. It appears that the senior staff followed Bertillon's opinion instead of Gobert's because it was the one they wanted to be true; and that they had already committed themselves to arrest Dreyfus anyway, before they received either report.

Bertillon became a star authority on the anti-Dreyfus side of courtroom expert-witness battles and debates in the popular press that spread out over a period of years. Several other experts also became involved; and although the bordereau was always front and centre, graphological examinations of other documents also became relevant. Henri Poincaré, for whom the Poincaré conjecture is named, was briefly involved at one point, along with two other mathematicians, to give a (negative) mathematical assessment of Bertillon's use of probability theory.

As the reasons to believe that Dreyfus had actually written the bordereau evaporated in the face of other evidence, Bertillon stuck to his initial assessment. He formulated an intricate theory of "self-forgery," which held that Dreyfus had written the document, but not in his own natural handwriting. Rather, he said Dreyfus had imitated what a forger would do to forge Dreyfus's handwriting, so that it could subsequently be claimed that the bordereau looked like a forgery; and he had done this unconsciously. When later given the opportunity to compare the bordereau with a handwriting sample from Esterhazy (its true author) that professional graphologists and casual observers considered to be a perfect match, Bertillon judged the Esterhazy sample to involve a further layer of forgery - someone expertly imitating the double forgery in the bordereau.

It seems significant to me, as indicating how forensic evidence was processed at this point in history, that Bertillon's graphological theories, as applied to Dreyfus and especially the bordereau, were primarily attacked for being "ridiculous" or "absurd"; not directly for being wrong! It is clear from our 21st Century vantage point that his claims in fact were wrong, but that was not the form in which he was criticized. The standard for assessing results in a nominally scientific discipline was laughter, not experimental evidence.

Even in 21st Century sources, the extent and nature of criticism of Bertillon's graphology usually amounts to little more than pointing and laughing at the intricate diagrams he drew in failed attempts to convince others of his theory, as well as his obviously incorrect factual conclusion about who wrote the bordereau. And other graphological results that were more accepted by mainstream science of the time, also lacked what in the 21st Century would be considered solid grounding in empirical facts. They were based instead on what deep thinkers considered to be intuitively reasonable ideas. Mainstream graphology at the end of the 19th Century, supported by theoretical claims from psychology as that field existed at the time, was often wrong.

That is why I describe graphology in the Dreyfus Affair as scientific or pseudoscientific. It was not scientific in the sense of that term used today. Today, after more than a century of additional research progress, a distinction is usually drawn between "forensic document examination" to answer questions like who wrote a document, thought of as scientific; and "graphology" as such, which answers questions about a person's psychology from examination of his or her handwriting, thought of as not scientific. Even the more scientifically-grounded "forensic document examination" to identify handwriting depends heavily on subjective judgments, and it is only given limited weight by courts today.

The distinction between these as two different disciplines was not made at the time of the Dreyfus case. Both kinds of questions were at issue, and it was generally accepted that experts in a single discipline which was called "graphology" could give probative answers to both. Bertillon was using graphology in the less-scientific psychological sense to support his conclusions about the more-scientific forensic examination questions, but all the other expert witnesses also mixed these two kinds of handwriting analysis.

Appendix B: Dreyfus's years of service

When Alfred Dreyfus retired from the Army in 1907, his pension was apparently determined by his number of years of service. Both Whyte and Burns report that he did not receive compensation for his years of imprisonment. However, the specific numbers these two sources give do not agree exactly with each other (Whyte says "30 years, 10 months, and 24 days"; Burns says "nearly 30 years of service"), and both agree approximately with the actual elapsed time between the dates shown on his service record for his start as an officer cadet at the Ecole Polytechnique (1 October 1878) and his 1907 retirement (25 August 1907), including his years of imprisonment and military non-active status.

A scan of the service record is available online from the French Ministry of Culture; my clip below is colour-enhanced.

I reckon 28 years, 10 months, and 24 days between the dates on the service record. That agrees with Burns's rough statement, and is exactly two years off of Whyte's more precise statement.

It seems that if we believe the number of years quoted by Whyte or Burns, and if that number of years was the sole factor determining Dreyfus's pension, then Dreyfus actually was given credit for his years of imprisonment after all; but they both say he was not.

And why do they disagree with each other? Burns's statement is vague, but to its implied precision of perhaps two years or better, it agrees with the dates on the service record. Those in turn are consistent with Dreyfus's age and with the dates given by other sources for important events in his life. Whyte's statement is more precise. It disagrees with the service record but only by two years, which is less than the claimed loss of five years, and the disagreement is in the wrong direction to be explained by his imprisonment anyway. Dreyfus himself complained of having lost five years, but that was in the context of his promotion seniority, not his total length of service for pension purposes, and it was based on the promotion dates of captains junior to him, not on counting his time of imprisonment.

I think Whyte may possibly have added up the durations shown on the service record instead of using the dates, although by doing that myself I still cannot reproduce his numbers exactly. Some of the durations appear to be incorrectly computed in the original document, as well as being ambiguously handwritten and liable to subsequent misinterpretation. In particular, Dreyfus's stint at the Ecole Polytechnique seems to be listed on his record as 1 October 1878 to 4 October 1880, and also as exactly four [sic] years, no days. The total shown on the service record (which also includes his service in the First World War, subsequent to the 1907 retirement of interest to me) seems to be a correctly-added sum of the durations shown there, including the incorrect durations as they are written and using the assumption that a generic month is equal to 30 days. Perhaps Whyte did a hybrid computation using the durations shown on the record but then comparing the dates to adjust the month and day counts, in order to avoid the thorny question of how to convert days into months when adding durations. That could possibly give the result he reports.

It is also possible that the durations shown are not incorrect, but are calculated according to rules or policies of the Army unknown to me that would specify something other than the simple elapsed time between starting and ending dates. For example, there could have been a rule that the program at the Ecole Polytechnique would always count as four years of service regardless of its literal duration.

It may be that the years of service included Dreyfus's imprisonment, but the pension calculation did not. The phrasing in Burns seems compatible with that interpretation. It also may be that this is getting back to the promotion-seniority question, with years as a major counting more than years as a captain, so that the late promotion effectively counted against him. The sources I have read imply that the pension amount was determined solely by duration of service from when he entered the army to when he left, but I have not found a solid, authoritative explanation of what rules applied to this matter in the French Army at the time.

I should repeat that Dreyfus throughout the Affair refused to accept money as compensation, so even if the Army had offered him more than the pension he would have had due according to the standard formula, or back pay or similar for those years in prison, he might have disdained to cash their cheque. I only include this Appendix at all because the apparent disagreement between my sources regarding the year count, seemed interesting.

2 comments

This was a very long but fascinating read. I'm left with a daunting list of further articles to look at that you've linked to, and a lot of questions to think over. It is comforting to have a contending theory that would explain some of the insanities of the present society - even if there's some flaw, at least it's a start.

And I suppose I read through to the end because I recognize my own situation in what it describes (who else would spare the time? Aside maybe from the very few people whose job it actually is to solve these issues, if they even exist).

It is similar to something else I've seen:
Hard times create strong men.
Strong men create good times.
Good times create weak men.
Weak men create hard times.

The main issue I see is the one you've already mentioned - what if it's NOT a cycle? What if this is actually the journey from birth to death?

What if, once an institution is even allowed to enter the abundance phase, there is no going back, and dissipation and dissolution will always follow?

Would not the best remedy be to prevent a state of "inflation" from ever being reached by having a form of societal organization with good feedback mechanisms that act proactively to stop any more people from being trained than there are positions?

Yes, if you're an organization and you're faced with "abundance", it is indeed more logical and moral to either stop hiring entirely, or use arbitrary hiring criteria that don't waste people's time or give them false hope. But is this very situation not a sign that the government has, either through neglect or incompetence, failed in its duties?
Esn - 2021-12-20 19:13

Thanks for paying close attention!

I think it may be very difficult, perhaps impossible, and also perhaps inadvisable, to try to prevent institutions moving from one stage to another. Benjamin Ross Hoffman wrote a very interesting reaction to Chapman's "Geeks, MOPs, and Sociopaths" article here (sorry this won't be clickable; the Web log software doesn't create links for anti-spam reasons; but you can cut and paste the URL):

http://benjaminrosshoffman.com/construction-beacons/

He makes the point that the "sociopaths" *perform a necessary function*; geeks may not like all the consequences of their presence, but do desire some of those consequences. Just not having them, may not be a realistic option.

Then if it's not a cycle, the last stage I've described can't really be the last one, and it may be that whatever comes next after abundance, if not a loop back to scarcity, is actually preferable to scarcity. See Chapman again, paraphrasing Kegan in the "bridge to meta-rationality" article linked above. If "scarcity" is stage 4 and "abundance" is stage 3, maybe instead of cycling between the two with a painful and problematic transition each time, the real answer is to move on to stage 5, breaking the cycle entirely.

I'm not eager to pin the blame on the government in particular because I think the forces involved are bigger than any government and governments themselves are institutions subject to the same failings as, for instance, universities. If there could be a government that really had the understanding and the power to manage this kind of thing, it's hard to imagine that it wouldn't be terribly subject to capture, that is, corruption, in itself and end up being much worse than a weaker government that didn't even make the attempt. But even with real-life governments and their limited capacity for effective social intervention, maybe even just a *little* more government attention to "elite overproduction" would be worthwhile.

When I was in Denmark I worked for a public university which - remarkably to me - was the most "run like a business" academic institution I've ever seen. Even though I was theoretically a government employee (and I am now officially entitled to a Danish government pension of 200 kronor a year for my sins, once I reach retirement age, no matter where in the world I'm living at the time), apparently once every few years the management of the university would negotiate a contract with the government to educate such-and-such number of students for such-and-such amount of money. And the government would decide how many students to hire us to educate, and in what fields, based on looking at how many jobs they needed to fill in each field. Note that our students, in addition to getting their tuition paid, also received a stipend for going to school; the North American "student debt" thing was completely foreign to this system. That seems like the start of something that could reduce the "elite overproduction" problem. It did have its own problems, though.
Matthew Skala - 2021-12-22 07:31