Regarding the sentence "One that he doesn’t discuss much but that is particularly pertinent to today, is the accretion of laws and regulations." which appears in the review: I think all these laws and regulations can be regarded as being part of the increasing bureaucracy that Tainter identifies as the cause of eventual collapse of complex societies.
One thing I would add, in addition to the diminishing marginal return of bureaucracy that Tainter mentions, is the tendency of contemporary administrations to create more bureaucracy even when their own stated goal is to reduce bureaucracy or increase efficiency, for example by trying to collect and compare precise numbers that show how they have become more efficient at a task that is in essence not really quantifiable. A related book I recommend particularly in that regard is Le Règne de la Quantité et les Signes des Temps by René Guénon.
> trying to collect and compare precise numbers that show how they have become more efficient at a task that is in essence not really quantifiable
For me this is particularly maddening as I run into it frequently. Managers/bureaucrats want to make decisions based on 'metrics', primarily because it gives them cover if it happens to be the wrong decision. But, many many issues where decisions are required are more subjective than objective. Meaning there really aren't any metrics that can reliably 'measure' them. Instead of just acknowledging this plain and simple fact, they just create more metrics and/or define the criteria with even more detail. It just makes more work but adds no further insight because ultimately, the issue is subjective and no volume of measurements will help. Think about coming up with a formula about how 'good' art is by measuring how many colors it has, how many curves, right angles, etc. It just doesn't work that way and but they keep trying.
The reason is responsibility, and not one wants to take it, especially not politicians or people working for the goverment. That is why they need metrics, so they cannot be taken responsible for the consequences. "It was not my fault. The numbers said it."
I've seen this in academia, for-profit, and non-profit sectors as well.
It's what happens when responsibility is not married with reward, or worse, is punished by those above you. Why take responsibility when all responsibility gets you is fired, scolded, blame, or a bunch of additional work but no accolades, pay raises, promotions, etc?
The only way to get around this, that has actually been proven to work so far, is to give a single body nearly unchallengeable power. Such that even if some mega-disaster happened because of a lack of quantification, nobody would ever dare punish, simultaneously, a super-majority of said body.
i.e. central committees of ruling communist parties.
Of course the downsides of such a system are familiar to many.
> Managers/bureaucrats want to make decisions based on 'metrics', primarily
because it gives them cover if it happens to be the wrong decision.
That might be true, but "making decisions based on metrics" is also the core idea of the meritocracy/technocracy. Rather than claiming that a person has special insight or superior intuition, and should thus control the decision, there's this contemporary (post-Enlightenment?) idea that we could instead systematize decision making and thus avoid or at least reduce cognitive bias, corruption and nepotism.
" Managers/bureaucrats want to make decisions based on 'metrics', primarily because it gives them cover if it happens to be the wrong decision"
My word, no, they want to make decisions based on 'metrics' because it validates the 'do or not decision'.
We make hardware based products and have a very real need to model component prices, fluctuations, trade/political risks etc..
Most things are not 'art'. If someone wants metrics on something so fuzzy, put the 'wide error' bars on the estimation.
Better yet - don't even talk about measurements, make it a 'risk assessment' exercise and identify the key issues that go into defining the unknown.
Ontario government is spending a huge amount of money doing 'environmental assessment' for high speed rail - it's better to do that than to start building and hope they don't come across some 'magical wetland' etc.
Granted - they are spending too much money on the assessment ...
I am in no way advocating for total `gut-based' decision making, because that would be an obvious disaster. I think gathering data, analyzing that data, and making a decision based on the results is a very good and important thing. My issue is that not ALL issues can be decided in this way. So while most things are not 'art', as you've said, some very much are like art and its for these types of issues the metrics model breaks down. It's just the wrong tool for the job. The `risk assessment' approach, I agree can also be a useful one in many cases but not all.
I think one example is when trying to measure, or more likely, rank, people for suitability for something. This can be more of an art. There are metrics that can be useful, but then Goodhart's Law [1] can kick in. And even when the metrics are useful, they simply can't account for, and will never be able to account for, the shear complexity and adaptability of actual people.
So yes, metrics can be good for some thing, risk assessment good for others, but that still leaves some problems on the table for which neither are sufficient.
> But, many many issues where decisions are required are more subjective than objective.
They're still objective, it's still objectively wrong to order your employees into a battle royale to the death or paint everything they can find teal as a response to pretty much any problem even when the$e are multiple right answers with a human component. It's just that measuring can cost more than the benefit the task brings, and change what is being measured to destroy anyvalue produced.
The belief that all problems can become objective provided enough data is measured is the problem. It's not anywhere close to practically true, and I doubt its even theoretically true even with all the assumptions of a frictionless spherical cow in vacuum type scenarios. This belief is what is driving the both the wasted resources in pointless data gathering, but also a paralysis of decision making by those in positions to make them.
The key above is 'all'. There are many, a majority even, of problems for which data gathering and analysis is absolutely the best the approach. These would be the objective problems. Not all problems fall into this category.
> The belief that all problems can become objective provided enough data is measured is the problem.
Reality exists. There are objective answers to management problems. There may not be a unique right answer, and which answers are objectively correct depends on the people involved far more than any oversimplified metric, but that does not make them subjective. It's objectively true that in any long term problem with difficult to measure outcomes giving people agency, allowing them to own their work, respecting them, and communicating in whichever way is appropriate to the particular situation is more effective than trying to treat them like a pocket calculator and maximizing some kpi that isn't even related to the output you want.
The term you are searching for is solvable by an algorithm or measurable, which they are objectively not. I'm fond of Goodheart's law as a rule of thumb as well as keeping in mind the maxim that any measurement both has a cost and is an interaction with the system being measured and thus must change it.
> There are objective answers to management problems
That is something you believe to be true. And something I believe to not be true. Have you ever managed large teams of developers? I have. For many years. And I was very successful doing it.
However there was no “objective” way for me to manage the teams. Humans are complex. And the interaction of team members even more so. I have never seen an “objective” metric applied to a team that actually worked without unexpected negative consequences and/or without being gamed by smart team members. If you have an example of an objective metric that has actually worked in reality without negative side effects and without being gamed I would like to hear it.
Objective does not reducible to a metric. It means when given an object (a team of developers and your relationships wih them) then there is a space of correct answers (ie. not reducing those interactions to a metric) and a space of wrong answers (ie. trying to start an orgy every day).
Your real name is subjective (it depends only on what you decide it is)
Your favourite colour is subjective
Whether you like a piece of art, think it's good or think it achieved what was intend€d is subjective
The meaning of a piece of art is objective but not reducible to a metric and maybe not unique.
This assertion that things are objective if and only if simple models based on overly reductive metrics are the only correct tool for thinking about them is incredibly wrong and destructive.
How the totality of possible choices are sorted into the correct and wrong answer spaces is subjective. Although there are many that will be sorted the same independent of the manager, there are others that will be sorted specifically because of the manager (the subject). One manager will decide one way because of their internal thought processes, and another may decide the opposite. This then means that some (but not all) management choices are in fact subjective.
That's a description of objective. Additionally there is a very simple test for whether your allegedly subjective edge case was objectively right. Did the thing you were trying to do get done? If the answer is no, then it might have been a reasonable strategy, but if there was a different strategy that would have achieved your goals (both stated and unstated) better then it wasn't the most correct strategy.
There are actions that lead to the things you want (whether profit, or a project, or a team that feels fulfilled, a sane and safe working environment, etc) being done, and actions that do not. Sometimes which are which are unknown or even unknowable ahead of time or the only known thing is that attempting to instrument humans and treat them as machines will not work. Sometimes they are very particular to the situation. But once your actions have an explicit goal then strategies for achieving that goal are objective.
Whether, after repeated projects, your team has succeeded more often than average at delivering the project with clients that either will return or help generate more clients, and approximately average (or less) resources as input, that the team are content, cohesive, and fulfilled with their work, and possibly that it was profitable. Ie. that the world you are in after having managed is one you like better than the one that you started in (and before you digress by saying that is subjective, the goal can be subjective but strategies for achieving that goal are objective once a goal is selected) or at least better than counterfactual worlds where you managed differently.
None of these things can be reliably turned into a metric that you can measure day by day without one or several highly fallible layers of abstraction. Any model complex enough to capture all the nuances such that you can blindly follow its output can just be called a manager.
So what you are saying is that you can only objectively measure anything after the fact. Making objectivity useless as a team management tool. My point exactly.
Not only that, even if your team was successful in the past, that doesn’t mean that doing what worked in the past will also make your team successful in the future. The context is important. And the context is always different when you start the next project. It might be a different customer, a different problem, a different team (people leaving/joining), a different world (politics/laws) etc.
So what you are left with is your experience, your people skills, and your ability to adapt to new situations. I believe that no “objective” tool will help with that. But I would be happy to be wrong.
So let me ask again: if you know an objective way to improve the likelihood of your team being successful then I would love to hear it. It would be great to have such a tool.
You just described objective tools after straw manning instrumentable problems as the only objective ones.
Your problem is you're conflating 'subjective' with 'not instrumentable'. Your name is subjective. You can be in an empty white room and choose it and it will always be correct. So is your favourite colour, or whether you think titanic was good. You can't have a subjective opinion and be wrong (although you can make an assertion about one and learn later that you were confused or that your opinion changed)
Once again, the strategy for achieving your goals once they are picked is objective. If you sit in an empty white room and decide the correct strategy for your software business to succeed in your goal of having a product used by a million people is to fire all your experienced employees, hire apprentice chefs, and instruct them to spend all of their time writing poems by hand you're almost definitely wrong.
The post hoc knowledge is proof that it's objective, not proof that it's reducible in its entirety to a simplified metric.
Metrics are a tool for dealing with some objective problems, so are "your experience, your people skills, and your ability to adapt to new situations". Measurements can also include having a conversation or reviewing someone's work but not trying to reduce it to a small set of numbers.
You need to make some measurements to know if you're in touch with reality, but throwing out everything except some tiny subset of measurements that fit in a policy doc or a kpi outline is just as stupid as making decisions from your white room with no feedback.
> which answers are objectively correct depends on the people involved
This is the almost the very definition of subjective.
However, I do take your point about problems that are objective but that are not measurable or solvable by metrics and algorithms. I would add that to my list of problems for which managers will use the metrics tool, even though it won't and can't work and causing frustration for everyone who is relying on them and/or have to collect the metrics data.
> This is the almost the very definition of subjective.
The implication is that the manager is the subject if the object is their interaction with the employee. Thus the answer would be fully internal to the manager if it were subjective.
Of course which answer is correct depends on which problem you are solving. By that metric every problem is subjective.
“Pournelle's Iron Law of Bureaucracy states that in any bureaucratic organization there will be two kinds of people:
First, there will be those who are devoted to the goals of the organization. Examples are dedicated classroom teachers in an educational bureaucracy, many of the engineers and launch technicians and scientists at NASA, even some agricultural scientists and advisors in the former Soviet Union collective farming administration.
Secondly, there will be those dedicated to the organization itself. Examples are many of the administrators in the education system, many professors of education, many teachers union officials, much of the NASA headquarters staff, etc.
The Iron Law states that in every case the second group will gain and keep control of the organization. It will write the rules, and control promotions within the organization.”
There is some insight in this, but it's also deeply cynical to the point of being very misleading and maybe a bit wrong.
Teachers and Administrators are equally as committed to 'keeping their jobs'.
It just so happens that 'teaching' is mostly 'teaching' and all you can do is 'teach' and so long as you have kids in front of you 'it's working'.
The scope of their roles doesn't allow for them to go too far away from 'the mission'.
Administrators are not default committed to the 'org' they're just in a position where their decisions and actions may or may not further the mission.
'Very Bad Teaching' is rare and usually self evident. With administration, it's not so clear.
NASA Engineers are not like teachers, in that, the could be wasting a lot of time and energy. Maybe the risk is worth it in some cases (?) in others, maybe not so much.
FYI - in teaching, the biggest costs are generally the 'teachers'. etc..
Regular Public Schools aren't a good opportunity so much to demonstrate this 'Law'.
A not-quite-independent entity like a Post Office might be. They often fight directly against innovation due to the risk it may create to jobs for rank and file, and don't quite have the same competitive basis as others.
Pournelle was not cynical enough. You only need to spend time in corporate America to watch the bureaucrats, administrators, and bean counters take over. It always starts small, but every day, while you are focused on making a better product, or a better process, or service to the customer, they are focused on control of the finances and core decisions, because it is in their nature to do so. And one day you look up from your work, and find that you are no longer permitted to make the decisions that are best for the company's goals. It is inevitable, because these people are like barnacles on a boat, or scum on a pond, or tarnish on silver. They dull and destroy every good thing they touch.
I've worked in Corporate America, consulted for government and at startups.
I know it's difficult to understand, but organizing people around large and complex goals is hard, and I wouldn't say that it requires a 'submission of autonomy' ... but it does mean that people's roles become much more narrow and that it's really hard to see every move in the big picture.
The best place to see it is in the military, where people are often thrown in situations with much higher risk etc. where the payoff might not be as much as we'd like ... but if we want to defeat Hitler, that's pretty much the way it's going to be. Soldiers in 'good armies' get to make decisions about the tactical bound in front of them and not much beyond that. No 'big picture' stuff.
Most of corporate America is highly inefficient but it's also not that bad.
Another way to understand it is that inefficiency means underutilization ... but we still show up and have meetings to fill time anyhow. Like some managers could work 25% of the time and do just as well, but their jobs are mostly still important.
It's hard to see that often from a 'single position' - how doing XYZ might seem perfectly obviously in the companies best interest but it may not be able to work that way. Though of course often it is.
I also focused in on this sentence. How difficult it becomes for an average person to start new things on his/her own nowadays! The only solution to the diminishing of widespread bureaucracy i can think of is war - but is costly, can take a long time, and most societies don't make it past a civil war. Like a forest fire, it clears away the hubris and overgrowth of process that accumulates over time.
What people arguing against accretion of regulations tend to miss is that this accretion (as currently experienced in the US at least) goes hand-in-hand with accumulation of capital and consolidation of industries. The large corporations can afford to comply with complex regulations. For them, maintaining complex regulatory regimes (through legalized bribery and regulatory capture) is part of the defensive fortifications that maintain their monopoly/oligopoly positions. As an optimist, I believe we can reform the system (e.g., through antitrust enforcement, which the FTC under Ms. Khan is doing more of than in recent decades), rather than burn it down.
I applaud your optimism, but don't let optimism cloud realism. There's comes a point when war is the only path that can relieve the accumulated stresses.
> comes a point when war is the only path that can relieve the accumulated stresses
This is historically untrue. When complex societies civil war, they tend to fall apart. (Wars of unification being an exception.) Their citizens’ living conditions permanently fell, often for generations, until an outside power subsumed them.
This trend looks to be strengthening as time goes on, probably on account of increasing species-wide firepower and mobility. Instead, peaceful (if politically violent) “revolutions” reboot the society from time to time. One could argue the Great Depression was the last such restructuring of the American system.
I don't think war is necessary (and looking at things like the "war on terror" that spawned a lot of rules, it might even do the opposite).
Societies can do "de-sprawling" of rules, but it isn't a frequent thing (the German civil code is kind of a reduction of lots of several codes into one, for example).
There is probably also an element of aging societies becoming more risk averse, hence more rules and regulations.
The "war on terror", like the "war on drugs", is a misnomer though; it's not really a war (although a couple of wars were waged as part of it, or alongside it).
Another interesting point is Switzerland after 1848.
They had a very small-scale civil war which made them rethink their political structure.
Or the Austrian empire in 1866, which lost a major war against Prussia and was forced to reform itself into a dual Austro-Hungarian monarchy. Which was somewhat more competitive and liberal than its predecessor and if it avoided other wars, it might in some form survive until today.
Neither Germany or Japan stand out as being societies with minimal bureaucracy and unnecessary/intrusive regulation. Singapore (as a state that had to rebuild considerably after war) is arguably an even more extreme example.
They were post-WW2. The German Miracle in particular came about when Germany went full free market as a way to recover from the economy being burned to the ground. This boom lasted until 1970 when the socialists were voted into power, and on came the taxes and hamstringing. The Japanese economy immediately after WW2 was run by American leftist academics, who refused to allow big business to operate. The economy flatlined. Until that was rescinded, and the Japanese free market economic boom began and ran up into the 80s.
That sounds very much like a conveniently right-wing potted history. I'd be willing to bet the truth is rather more subtle*. Either way, they were both examples of economies that only got off the ground because of massive government spending.
(*) a quick read of a wikipedia article about the Japanese economic miracle - which undoubtedly you'll consider leftist propaganda - certainly confirms this. At best relaxing anti-monopoly laws was an example of legislation reform that helped further boost the already impressive recovery that had occurred during the 50s.
Um, where did Germany get "massive government spending"? If you say "the Marshall Plan", look up the MP on wikipedia. Germany did get MP money, but far less than Britain and France did. The latter two did not have a "British Miracle" or "French Miracle".
I didn't claim the spending was the prime cause of any "miracle", but it's fairly certain it wouldn't have happened without it. France and the UK did have impressive recoveries too, that would have been impossible without the MP etc.
FWIW I would actually expect the fact that because the priority at the time was economic recovery before everything else, red tape wasn't allowed to get in the way to the degree it often seems to once prosperity had been restored. I'd even say that's probably a reasonable trade off, though it certainly doesn't justify governments letting growing businesses get away with whatever they feel like. Nobody likes bureaucracy or red tape, but almost always it's what we end up with as a result of repeated cases of corporate behaviour that's widely agreed to be unacceptable.
Leftists always mis-attribute the causes of prosperity. The wikipedia article cites causes that are commonplace in other countries that had no miracle.
Nobody here is arguing a fully socialist command-and-control economy is a better choice.
But you haven't made any sort of case that the German or Japanese "miracles" were largely due to raw unrestrained capitalism. There are way too many factors that made those particular countries special cases at the time to pinpoint all the primary causes, let alone tease out various secondary factors some of which probably tempered rather than accelerated growth.
German ordoliberalism and the social market economy are certainly not full free market, i.e., Germany never went full free market after WW2 and there also would have been no political support for such a move (from the left or right). The boom then ended in the 2nd half of the 1960s.
A country does not have to be "full" free market to enjoy the benefits of the free market. The more free market it is, the better the results. It's one of the beauties of the free market.
German industry was very interconnected with limited domestic competition in those days - a free market it was not by a long shot.
Edit: It's fine to make broad claims about free market and growth, but please provide some citations to back up sweeping statements (and also to allow to go deeper into definitions).
Do I really need citations for free market growth? Every country that tries it gets good results - the more free market, the better.
Communist economies always wind up facing famine.
Communist China, for example, switched from a communist economy to a free market one. Look at the results. Korea split in two, one communist, one free market. Which one prospered? East vs West Germany, same story.
The US tried central economic planning for gasoline in the 1970's (repealed by Reagan as his first EO). Disaster. The US tried central economic planning for airline routes, fares, and schedules up until Reagan ended that. The result was a huge increase in efficiency and low fares.
Of course you do! Statements like "The more free market it is, the better the results" are not just about things at the ends of the spectrum, so that needs proper backing not handwaving about communist vs. capitalist economies. For example, is Norway more free market than Iceland? What are the criteria?
Perhaps but war was posited of a method of "clearing the books" and allowing significant reform. If that were true you'd expect such societies to be less weighed down by bureaucracy than others that had no such opportunity.
Germany and Japan have had reputations as being somewhat bureaucractic for decades though, which says to me war had no lasting impact, if any at all. I.e. a very expensive way to achieve nothing.
Notionally the "wars" on drugs and terror are more akin to the Islamic principle of jihad, in the sense of a struggle, rather than of military domination, occupation, and conquest.
In the case of World War II, numerous governments were created from whole cloth following the war, amongst conquored territories, in newly-created states, and even amongst some of the winners.
Notably, Japan had a constitution imposed on it by the United States, which largely exists to the present day. Some traditional institutions were preserved (e.g., the emperor), but in almost entirely symbolic form. Germany likewise was reconstituted under a new government, as was Italy. Even France saw a new government formed (the Fifth Republic) in 1958. The states of the Eastern bloc were reconstituted under Communist rule as Soviet satellites. India, Indonesia, the Phillipines, and several Arabic states received independence. The state of Israel was formed. Much of Africa decolonised in the 1950s and 1960s.
There was of course some continuity through all of this, but in large part, new governments emerged and old laws were often voided.
But through all of that the German civil code after WW2 still looked quite similar to the one pre-WW1, e.g, rules on how people transact or private property had huge continuity.
Isn't the question not that laws etc. change but if there is true reduction of the total?
Get some employees, have some of them turn out to be pain in the ass, then find out how annoying west coast labor laws are.
Hire some employees, spread them across multiple states, find out the complexities of employment law in multiple states.
Start up a health plan, same thing.
There a million examples, none of them are bad by themselves, but every aspect of a multi employee business has rules and regulations that if you run afoul of, there are consequences. No one gives you a manual of these, you find them out along the way and it costs you money and time.
Maybe the OP hadn't in mind the actual process of incorporating a company per se, but the process of "starting" a company as in getting it off the ground in terms of money coming in/revenues. And, imo, it's more difficult now to do that compared to 10, 20 years ago, and not only because of increased competition.
There are more opportunities for 'regular folks' to do something on a 'grand scale' than at any time in history, but not everyone can do it.
Paradoxically, starting your own business in the 'old time' way, i.e. plumbing or roofing contractors ... it's like guaranteed work for great income. Plumbers are the new Dentists.
That said, there are areas in which regulatory issues are obviously a problem.
Perhaps you know since you’ve read the book: Does the book _really_ decide on declining marginal returns to complexity as an ultimate factor? Or is that the interpretation by the reviewer? The bridging of sociology, history and economics was somewhat rare back in the 80s and before. (Douglas North a notable exception, obviously not alone.) If the author does, now I’m wondering why I missed this book & going to read.
First thing as a counterpoint that comes to mind is biological complexity. There are marginal diminishing returns to complexity in biology as well, but not a rise / collapse rhythm in biological complexity outside of external events that I know of.
On the bridging of economics and sociology, I'd strongly recommend Charles Perrow's Complex Organizations: A Critical Essay (1986), which is a tour de force survey of the literature.
Notably, Herbert Simon spanned multiple domains (economics, psychology, sociology, organisational behaviour, AI, mathematics, ...), though comes off somewhat poorly in the text.
On Tainter, he specifically assesses eleven factors: resource depletion, new resources, catastrophes, insufficient response to circumstances, other complex societies (e.g., warfare/invasion), intruders, conflict / contradictions / mismanagement, social dysfunction, mystical factors, chance concatenation of events, and economic explanations.
Chapters 4 & 5 specifically address marginal productivity. They are titled "Understanding collapse: th emarginal productivity of sociopolitical change" and "Evaluation: compelxity and marginal returns in collapsing societies".
Yes, declining marginal returns to complexity is Tainter's principle thesis.
This is a great point of comparison. I feel that societal systems (and associated complexities) are more rigid than biological systems in the sense that a biological systems evolve efficient solutions in response to changing internal and external stimuli while societal systems cannot/will not due to man-made laws, norms, expectations etc. The latter (biological systems) shows first order adaptation while the former (societal) shows, at best second and third order adaptations, if it even adapts.
> The bridging of sociology, history and economics was somewhat rare back in the 80s and before. (Douglas North a notable exception, obviously not alone.)
So Karl Marx and Max Weber are just chopped liver?
Ah true point. Perhaps the addition /in the modern scientific era/ would be in place. Begging the question when that starts of course. Let’s say that economics was earlier embracing quantitative techniques in the 20th century and that there was a rift between the fields especially when sociology was young that nowadays is less pronounced. It’s all interpretation.
I would have said that the 1960s and part of the 1970s were filled with attempts to synthesize, among other things the fields you've mentioned. It wound down with Reagan and Thatcher's elections, which may or may not be proximate cause - the new right at that time were certainly not fond of the humanities in any way.
There is also the phenomenon of regulatory capture, when the bureaucracy responsible for implementing regulations gets captured by the corporations and entities involved. Very relevant to what's happening today.
I can’t say much about the veracity of her arguments, but there is a journalist named Sarah Chayes that has been investigating the US failures in Afghanistan as, primarily, a reflection of America supporting (and helping to build) a culture of corruption that undermined our primary nation building objectives in that country. She extends this argument to American political and business culture that shares these same qualities of corruption (that we conveniently label as something other than “corruption”).
In the book A Fire Upon the Deep, Vernor Vinge outlines a kind of bleak future for humanity in which societal collapse is inevitable. Despite interstellar communication and wild technology, there always comes a point at which the machinery of a given society is too complicated and heavy handed and over-fitted to handle some change in circumstances. The result is the survivors are thrown back to pre-industrial society. Of course in his book the machinery of society is more intrinsically enmeshed with high technology—a failure in supply chain could result in a failure of resource allocation or governance.
I found the prequel, "A Deepness in the Sky", even more compelling. The book's theme is the inevitable collapse of legacy software under the weight of its own complexity. And the book views codified government as a form of legacy software.
> the tendency of contemporary administrations to create more bureaucracy even when their own stated goal is to reduce bureaucracy or increase efficiency, for example by trying to collect and compare precise numbers that show how they have become more efficient at a task that is in essence not really quantifiable.
Imo it is not the existence or ineffectiveness of complex laws/regs that leads to problems. Rather, their lack of understanding among the population, and difficulty of execution/implementation.
I actually had a startup project the I shelved specifically aimed at addressing this. Wonder if I should revisit it...
Not sure you had to be downvoted for stating a simple opinion. While I don't believe you are right, I'd like to ask for the reasoning that led you to this conclusion.
> One thing I would add, in addition to the diminishing marginal return of bureaucracy that Tainter mentions, is the tendency of contemporary administrations to create more bureaucracy even when their own stated goal is to reduce bureaucracy or increase efficiency
> “Rules and regulations are immortal. They don’t die … there’s not really an effective garbage collection system for removing rules and regulations.” Piling rules upon rules instead gradually “hardens the arteries of civilisation, where you are able to do less and less”.
> First, there are way more examples of complex societies collapsing than I was aware of.
To add to the discussion, I submit this entertaining docu-series[0] describing the collapse of Bronze Age civilizations around the eastern Mediterranean.
The authors proffer that one possible cause for their near-total collapse could have been a breakdown within Bronze Age society's well-structured and complex systems from which recovery was simply not possible.
Apologies to the history purists for the inevitable inaccuracies to be found herein but this was an approachable series for me as a codemonkey.
I can also recommend the FallOfCivilizations channel on YouTube.[0] It's basically a podcast that later started adding video footage to its episodes. As the name suggests each episode deals with the rise and fall of one specific civilization.
I thought about posting this exact video, but you beat me to it ! Extra credits really has some very well done entertaining historical docu-series .... I recommend other historical buffs to subscribe and go through their videos
Diminishing returns on complexity might just mean that when thing get complex, they get harder to figure out, so a larger percentage of attempts at improvement (even attempts at simplification) are likely to fail. Given that complex societies take a long time to rise, the perpetrators thereof may see the long-term upward course as a law of nature, and become a little over-optimistic about the advantages of change.
The most universal law of nature: Periods of exponential growth always end.
I am reminded of software that has entered its "Big Ball of Mud" phase, where it has become impossible to understand, even to debug or simplify, and so it can no longer evolve to meet changing needs, and is eventually replaced with something newer (and, for the moment, simpler).
I'll agree that complexity is part of the Ball of Mud phenomenon. The worst I've seen was attributable to what used to be called 'coupling', mostly accidental, incidental, or pathological, between different parts of what was said to be a 'system,' mostly for lack of a better word. With all the incomprehensible consequences of any software changes, bug fixes of the buggy code were discouraged by management, because the average number of new bugs caused by each repair was approximately equal to 1 + the number of programmers involved in creating the fix.
>> a larger percentage of attempts at improvement (even attempts at simplification) are likely to fail
If we are talking about laws, there is no "fail" - no politician ever will recognize a mistake. Failed laws are never revoked - they are either ignored, unenforced or continue to do harm.
Came here to mention the interview by Nate and am happy to see you did.
In addition, and on a similar conceptual basis, I'd recommend:
The Great Simplification [1]
The Century of the Self [2]
Can't Get You Out of My Head [3]
The Resilience web site content [4]
The Consilience Project [5]
The Dawn of Everything has to be one of the worst books on anthropology out there. It's a thing disguised polemic dressed up as archeological history.
I find Yuval's work to be more digestible, in part due to better writing, and in part due to less pretense.
I would also suggest Susan RiceBauer's The History of Ancient Civilizations, Jack Weatherford's Genghis Khan and the making is the modern world, and Lars Brownworths Lost to the West: the forgotten Byzantine Empire.
I think the mechanism is simple "imperial overstretch." Societies that go to war give a lot of their resources to the people who produce military material. Those people's disproportionate control of resources gives them a disproportionate influence in politics, which they then use to encourage more wars, creating a virtuous circle. As long as war delivers plunder and expansion, society has the resources to continue this.
But the incentive of war suppliers to encourage war is not the same as an incentive to win wars. War suppliers profit more from wasteful use of the military, and endless extension of wars, than wins (which replace them with peacetime producers.) This slows down the rate of plunder and expansion until the resources expended on war material become a huge drain on the rest of society.
This brings general unrest as the lower classes become desperate, civil war when different factions argue about which wars to pursue or end and on strategy, and secession at the borders, especially the areas that were the fruits of expansion. This is weakness, and invites invasion by a neighboring society in the growth phase of the same process.
I think the association with complexity is just because complexity in society rises with the size and organization of the military and its infrastructure.
Does this cover the Bronze Age collapses, triggered as societies moved to iron? That could be a counterexample: the long-range trading for copper and tin was no longer required, and that decrease in complexity (combined with the increase in killing power of iron over bronze) led to multiple collapses across the Mediterranean world.
The historical cases covered are (chapter 1): The Western Chou Empire, the Harappan Civilization, Mesopotamia (includes the Bronze age collapse period, though not detailed as such), The Egyptian Old Kingdom, The Hittite Empire, The Olmec, The Lowland Classic Maya, The Mesoamerican Highlands, Casas Grandes, The Chacoans, The Hohokam, The Eastern Woodlands, The Huari and Tiahuanoco Empires, The Kachin, and the Ik.
These are brief accounts (14 instances in 13 pages), but common themes are highlighted.
I think that's a very good point. Maybe not that we don't have multiple societies, we definitely do[1], but that no one is willing to take the wheel of the so called "globalized" part of the world.
[1] See e.g. China, Mongolia, Afghanistan, large swaths of people without internet. The list goes on and on.
While differences do currently exist, China is increasingly similar to the west in many ways (source: lived here on and off 20 years, also lived AU/UK/US). One of the main differences is entrenched centralized political control not subject to electoral terms, although that comes with really significant downsides it also creates a truly impressive potential for rapid change on critical issues (eg. environment). The disconnected peoples are effectively systemically marginalized and increasingly irrelevant as "societies". They will likely suffer from climate change, sea level change and other coming catastrophes more than the global society. I see only one path forward, a hegemony of global capitalism unbound by political will and environmental destruction. One of the few avenues for potential resolution is private sector mitigating technology development. If only people would stop going to work for FAANG and selling ads.
> easy to see how over time their benefits decline and their costs rise when you only ever add laws and regulations but never do a partial or complete rewrite
written by someone who has thoroughly avoided learning even superficially about how laws and regulations are actually being created and curated.
or even just thinking it through. how would append-only regulation even work? you'd have multiple slightly differently worded paragraphs right below each other, and only the last one counts? that's clearly not how it's done, also not in the bigger picture.
You’re missing the point completely. It’s not about law and regulation literally being append only, it’s rather about the number of laws and regulations consistently growing over time, with very little repeal of old regulation, which are mainly removed only to be replaced by more complex and extensive ones on the same topic. If you even superficially learn about how regulations and laws are created and curated in the US, you will find that this is exactly how things work here.
> arguably led to the greatest financial crisis in a century
The act did not ban securitization or buying derivatives. Even wiki can't find any justification for this argument beyond an unsourced 'cultural shift'.
The argument has its own wikipedia article with numerous citations, including from Economics Nobel (Memorial) Prize winners https://en.wikipedia.org/wiki/Aftermath_of_the_repeal_of_the... saying that wikipedia makes 'unsourced' claims there is acting in extremely bad faith
And even if you read all those citations and didn't find them convincing (or found the cited counterarguments more convincing), then point a) above would still stand - claiming that regulations are never revised or repealed is an uninformed stance
I referred to a particular point in the original cite[0], which is unsourced. Taking that to refer to my entire argument is, well, bad faith.
As for your new cite, I'd say it's the arguments in favour that are most damning. They tend to eventually concede the point that G&S did not and would not ban the behaviour leading to the crisis; They just lament that their policy choices weren't implemented at the time or lament some 'cultural shift' with no evidence to substantiate it (IMHO, everyone looking could have seen the culture shifting at least a decade earlier).
Aside, the argument was never that regulations are never repealed - but that the natural overall direction is ever expanding regulation. This thread is arguably somewhat of a confirmation, since people are arguing for a regulation which had little to do with the crisis...
[0] "The perception is that the Glass-Steagall Act created a sense of accountability among investors within the financial management industry... Thus a cultural shift was certainly in order after its repeal regardless of the loopholes that existed prior."
This has no source, and violates wiki's ban on original research.
The repeal allowed banks and investment houses to be run under the same roof, changing the aggregate risk profile of financial institutions greatly, in the direction of significantly higher, and now correlated risk.
Ordinary 'investment banks' were perfectly capable and allowed to hold securities under G&S, in fact they issued more than any 'same roof' banks had (and as the wiki cite in the other thread shows, there was little structural change following repeal).
It’s my opinion, having worked in IT in various roles, that there are 3 kinds of systems. First, the informal systems, which are just everyone doing whatever works for them. Those systems are flexible, simple, and inefficient. Second, the rigid systems, which tend to crop up when the org grows big enough that the informal systems start getting in their own way. Those are simple, inflexible, and efficient. Third are the complicated systems, which happen when the simplifying assumptions in rigid systems become problems. These are complicated, a bit less efficient than rigid systems, and a bit less flexible than informal systems.
Now, the problem with the complicated systems- which tend to become more and more complicated over time- is that they require a dedicated staff to maintain, which tends to grow over time, and so the organization has to be big, and needs to have access to a big profit margin/energy surplus. If that goes away, then yes you have to collapse down to rigid or informal systems.
The same process no doubt plays out on a civilizational level too.
>"While these are essential tools for maintaining complex societies it is particularly easy to see how over time their benefits decline and their costs rise when you only ever add laws and regulations but never do a partial or complete rewrite.
It is the societal equivalent of the accumulation of technical debt in startups."
What a great analogy!
At least for people that truly understand what technical debt is...
there are many analogies that could be made comparing software development to society.
for example, distributed vs centralized is akin to communism vs capitalism.
web-development, like modern-day societies, only want to "scale", but forget that scaling goes both ways, up and down.
Yeah, I know. People without a functioning moral compass, to whom all the pain and death involved mean nothing? Romantic dreamers, who think that it will make everything better? Short-sighted escapists, who think that to not have their current set of problems means not having problems? Revolutionary true believers, who think that we will inevitably wind up at their version of The Way Things Should Be? Egotists, who think that sure, there will be lots of pain and suffering, but they will be fine? A mix of all of the above?
Whatever, it's a view that seems blind to all the death involved. Do they just think all that death won't happen? Or do they think "it will be worth it"?
I am profoundly uneasy with accelerationism. It seems like a view for sociopaths.
Well... let's say that as a child, my family struggled with hunger. There often wasn't enough money for food. I grow up, I get a good job, now I don't have a problem with hunger. So I eat a lot. But obesity, arterioschlerosis, heart disease, and diabetes don't mean much to me. They don't have the emotional punch that hunger does, because I lived hunger, whereas obesity is more of an abstraction, at least until my first heart attack.
I think there's a heavy emotional component. Escaping from the problem that I've carried the emotional weight of, for years, is a big deal. It feels worth it, even if I trade it for another problem that is objectively just as bad.
But in context, "just as bad" doesn't apply. Collapse is objectively worse than sitting in an office at a meaningless job, or poverty from student loans, or inadequate health insurance. People cheering for it are short-sighted, no matter how bad and how emotionally scarring their current experience is.
in addition to your enumeration there are the religious fanatics who take current problems as the "signs" for the "last days" and who are longing for some "second coming". some of them are fatalist some of of them actively long to accelerate the process.
That's dispensationalism.[1] It's a weird mixture of Jewish tradition and evangelical Christianity invented in the 1840s. It's tied in with "Left Behind", the building of the Third Temple, the second coming of Jesus, and Jerusalem as the world capital. Wikipedia has a good summary.
I am highly skeptical there'a any clear causal relationship between complexity and collapse, but let's say there's some. I would rank nuclear war and a few others (climate change, or more broadly environmental imbalance caused by humanity) much more plausible and realistic cause for the collapse, way higher than what meager change this has, and thus this is mostly a distraction and intellectual pastime than something that is worth our collective attention.
Long, slow, persistent frictions and rot tend to be more destructive over time than sudden attack or incident, though the rot may enable a specific incident to become a determinative trigger.
C.f., Bonhoeffer on the danger of stupidity --- unlike determined malice, it never rests and knows no bounds or restraints:
Stupidity is a more dangerous enemy of the good than malice. One may protest against evil; it can be exposed and, if need be, prevented by use of force. Evil always carries within itself the germ of its own subversion in that it leaves behind in human beings at least a sense of unease. Against stupidity we are defenseless. Neither protests nor the use of force accomplish anything here; reasons fall on deaf ears; facts that contradict one’s prejudgment simply need not be believed- in such moments the stupid person even becomes critical – and when facts are irrefutable they are just pushed aside as inconsequential, as incidental. In all this the stupid person, in contrast to the malicious one, is utterly self-satisfied and, being easily irritated, becomes dangerous by going on the attack. For that reason, greater caution is called for than with a malicious one. Never again will we try to persuade the stupid person with reasons, for it is senseless and dangerous. ...
Excellent read, thanks! I guess this is what we call populism today. Relatively smart people filling the minds of the populace with stupidity, slogans and catchwords just to do their bidding when the time comes.
It could be a toddler or child near dangerous equipment, it could be a primitive society, or an arrogant colonialising force ignoring local wisdom. It could be management interfering with engineers, or engineers with management. It could be the general public trying to address complex topics. It could be specialists presuming greater understanding over complex issues. It could be an individual or group blinded by panic, confusion, injury, or disease. It could be technologists attempting social fixes, or social practitioners addressing scientific ones.
The common thread is that it is the ignorance itself, and the stupidity (perhaps domain-specific, temporary, contextual, or emergent) which matters.
Arguably our inability to reform our economies to be more compatible with ecological sustainability is partly due to the level (and consequent inertia) of complexity in our systems though. Though I'd say the lack of motivation is largely due to human nature (there's presumably never been sufficient selective pressure to favour genes linked with the ability to properly plan for futures beyond our own lifespans or to be concerned by global-scale long term threats).
The chance of nuclear war in any particular year is low, but I wouldn't say that the chance of a nuclear war is low. We were seconds away from midnight multiple times in the twentieth century.
The political and technical failsafes against it need to work every time.
D-D-Don't quote me regulations. I co-chaired the committee that reviewed the recommendation to revise the color of the book that regulation's in... We kept it grey
I notice that the people who tend to be pessimistic more often than not are those that don’t have much direct exposure to technology (whether through work or interest) in their daily lives.
I personally would be somewhat pessimistic as well if not for technology (energy generation/storage and AI primarily in terms of resource creation, though improved AI without a corresponding implementation of some form of basic income would be worrying).
Just a very superficial comment/thought after reading the review:
There seem to be an underlying implied correlation between complexity and failure (or collapse).
If that premise is, indeed, part of the book's thesis then, in this context, would be fair to say that "simple" societies are less prone to collapse?
Primitive societies do not collapse or grow much. They just stagnate. It might be nice living (I would surely enjoy some hunter-and-gatherer lifestyle), but there is basically no progress over centuries or even millennia.
To look at one extreme example, Neanderthal stone tools (Mousterien) haven't changed very much over 100 000 years of their existence.
It's profoundly incorrect to say that nomadic societies "stagnated". They adapted to incredibly variable and harsh conditions from the equator to the Arctic and invented much of the technology of the modern world like agriculture.
Archaeological horizons and the lack thereof are only partly related to the societies that produced them.
If you're just getting into the subject of what ancient societies looked like and how diverse they could be, I'd recommend Dawn of Everything as an approachable instruction to the topic.
The delta in comparison to the timescale was extremely small, though. Stagnation does not necessarily mean "nothing ever changes at all", just a very small rate of change compared to the modern era.
A life of a person in 8000 BC was not that different from life of a person in 7000 BC. For a person from 1000 AD, the world of 2000 AD would be wholly unrecognizable.
Highly organized societies tend to transform their environment very thoroughly - just look at all the Roman buildings and facilities that still dot the landscape from Spain to the Levant. A rural society wouldn't be capable of building something like that.
Don't mistake the limitations of the archaeological record for the underlying reality. Humans are distinctive from most animals precisely because they have adaptations like culture that are perfectly suited to high speed adaptations to changing conditions. Someone born thousands of years would have found quite a lot different, just like you would if you were born in the Roman empire.
Suitable, yes. I don't doubt that cultures can develop extremely quickly when the conditions are ripe. The Japanese dismantled their feudal structure and transformed into a major industrial and military power within a single generation.
But that happened under a strong, merciless pressure from other industrialized nations. Nature acts way more slowly that Commodore Perry did and the adaptations are much slower, too.
> A life of a person in 8000 BC was not that different from life of a person in 7000 BC
What about Çatalhöyük? 7000 BC puts you around the zenith. Not far away in Mesopotamia, 7000 BC would also be about the start of the Hassuna period, the first development of pottery in the region [1]. In general, 8000-7000BC is seeing major progression in the Neolithic Revolution, so the life of a person in 7000 BC is perhaps as different from 8000 BC. as 2000 AD would be for someone in 1000 AD.
[1] And it's worth noting that pottery is, archaeologically speaking, basically the best tool we have for identifying changes in illiterate societies. Pre-pottery, there usually isn't enough surviving remnants of society to make good guesses as to the cultural practices of that society.
There is nothing colonialist about the fact that less complex societies don't develop as quickly as more complex ones.
It is not even about ethnicity. For example, the Incan empire developed very fast and reached Roman-like levels of complexity and political power in mere decades after its founding, in fact much faster than the Romans themselves once did.
In the meantime, hunters and gatherer's way of life in other parts of the world, including Europe (such as the Sami in Lappland), barely changed over centuries.
Is it a "fact" though?
Human civilization now is arguably more complex than it's ever been in the past, yet I'm not sure we could argue the pace of change in the last few decades is the most rapid/transformative it's ever been (which Tainter would presumably see as evidence of the reducing gains of increased complexity). Or perhaps like in politics, there are decades where nothing happens and weeks (well, years at least) where decades happen...
And we should definitely question the assumption that a faster pace of technological change is actually a good thing.
To be more exact, primitive societies can still collapse demographically (e.g. through a widespread plague or famine), but they don't have any extra complexity to lose.
I think there might be room for misreading/misinterpretation of that part.
I read that part of the review as saying complex societies have 'bucked the trend' by not being -as- prone to collapse; but with a big caveat that it looks like complex societies may operate on longer time scales, so the collapse might be staved off for longer. It's an open question about what 'long enough' would look like though, one that's probably beyond the author's scope.
It's also one that emerges throughout history. The term "Seneca Cliff" refers to an observation by the Roman rhetoritician. Apocalyptic concepts are common in the Bible. Eastern religions tend to take a more cyclical view, of recurring expansion and collapse. Oswald Spengler's Decline of the West is a classic. There's a modern literature largely spawning from the Club of Rome's Limits to Growth (1972) of which I'd include Tainter. James Howard Kunstler's The Long Emergency is a more recent take on that.
Broad-brush observations and gut feelings tend to be less substantive and articulable than specific indicia and mechanisms. Tainter's work is of interest in that he does point to a general rule that is common to multiple instances.
The world seems to be entering a crisis, but I don't see any overall trend over "advanced economy countries" (you mean rich ones, right?), nor I see reasons for this specific crisis to be very different from any other.
You may be thinking about one or two countries, and overgeneralizing.
At the risk of mentioning Trump, I thought this was what executive order 13771 was supposed to alleviate: make the regulatory state self-pruning by forcing it to repeal two regulations for every one that was passed. It was environmentally controversial, certainly. I don't see how the government will ever reduce its complexity unless it has a similar mechanism in place to force existing regulations to be removed to make room for new ones.
Society is a dissipative structure, and there is a thermodynamic logic that applies to it. Energy goes into a society, and outputs are metastable systems and the inevitable entropy. And sooner or later the amount of energy needed to maintain that metastability will exceed the energy available, this is in rough analogy to the Carnot engine efficiency, where efficiency is proportional to the temperature differences along the system boundaries. Everything makes perfect sense if you cast it in this heterodox framework.
> Second, Tainter proposes a very simple and general mechanism leading to collapse: declining marginal returns to complexity. Over time the benefits of complexity diminish and its costs increase.
This makes me think of the US Federal Tax code, the pure opportunity cost of the entire tax industry and legislation that could be otherwise spent if we just had a flat tax with it's full regressiveness accepted. Any charity, favoratism, etc from the government should come in the form of grants etc from some other agency--keep the taxes simple.
If you are arguing for a flat tax, then simplicity is not enoigh to make up for the obvious downsides. Particulalry as fixing them isnt conplicated.
If you're arguing for simpler tax systems, which I think I agree with, mentioning flat tax ideas in the same post is only going to put people off the idea, and see it as a scam to reduce taxes on the wealthy and undermine democracy.
> and see it as a scam to reduce taxes on the wealthy and undermine democracy.
While the wealthy do pay the majority of taxes - there's a shocking number of absurdly wealthy people with low tax rates (their overall tax bills are still eye watering to a normal pleb).
So it's hard to argue for a flat tax without noting that it will lower the average wealthy person's tax bill considerably. At the same time, most people who argue for a flat tax think it will actually increase wealthy people's taxes!
Most people are aware that the rich "cheat their taxes". They aren't aware that the vast majority of them are still paying above median tax rates.
There's no way around it. The tax bill, currently, is heavily footed by the wealthy. If you go to a "flat tax" it's going to reduce taxes on the wealthy and increase them on the majority.
The UK and the US differ dramatically in their use of the tax code to create behavioral and economic incentives. Most of the complexity in the US tax code stems from the government's repeated, endless use of "tax breaks" (credits, deductions and more) to encourage various things, whereas most other governments dispense money to people who participate in whatever is supposed to be encouraged. In theory the effects are similar; in reality, the implications are wildly different.
There should be some mechanism that measures the existing inequality and automatically adjusts tax progressiveness so to keep inequality at some constant level.
So, real question, why does the tax code need to be -that- simple? Like, we've got computers and stuff to do the math these days? It sounds like an easy way to say that one is for 'a simple, common sense solution' while papering over the very real harm it's going to do to at-risk and impoverished communities.
To contribute, why should those communities that a flat tax harms be forced to bear that weight of simplification?
It looks like you'd be ignoring the reviewer's second point: that the other factor in societal collapse is when those portions of society that bear undue weight no longer can bear additional burdens.
I’d say when it’s that simple there’s nothing to argue with and no major brain power demanded, let alone a computer program needed to crunch the numbers and optimize in some highly complicated multidimensional space. There’d be no competing interpretations to battle out between lawyers and CPAs. The only thing that would be argued about is at what tax rate.
I’ve often dreamed about what the tax formula would look like in its full gory details as an unevaluated expression full of variables and piece wise functions and all that.
To the second point I’d say why wait until collapse to trim off unproductive regulatory churn. I guess that’s the fundamental thing—is it a productive pursuit on net and compared to what exactly. It’s hard for me not to imagine the free market filling the void with something more productive even though I can’t single out anything specifically.
How can you just have a "flat tax" if you pay employees, or have inventory costs? You need deductions so you need to track everything anyway. That is the hard part.
So having a flat tax or not barely saves any complexity at all.
Adding a simple floor to the tax rate would solve much of the regressiveness. Nobody would pay taxes on their first $40k (or whatever number works), and then a flat rate above that. The floor could even be done as a prebate to gain some social safety net benefits. Every citizen is given $tax_rate * $40k every year, doled out as monthly payments, and pays a flat tax rate on all income.
We could retain a progressive tax system with or without the use of tax incentives. I would personally favor the removal of all tax incentives to be replaced as your suggest with grants, but would combine that with a continuously valued formula for tax rather than tax brackets. It's very simple, except that most people don't understand "equations".
This reminds me of a commentary on the new megalaw - the Inflation Reduction Act - that it succeeded (got passed) because it wielded carrots rather than sticks.
This approach is compatible with your idea - to keep the tax side of the budget super simple, and put all the incentives and giveaways in the grants side (fairly transparently).
In every complex societal collapse prior to the present day, subsistence agriculture was either the most common occupation or well-known to a substantial portion of the population.
I caution against turning doomerism into an intellectual fetish - unlike past societies, we don't know how to feed ourselves without "the system".
yes but that was because the lower class was largely responsible for food generation. today they are responsible for so much more. Technology, especially hi-tech contributes to our current complexity.
Albert Wenger is a partner at Union Square Ventures (USV), a New York-based early stage VC firm focused on investing in disruptive networks...Before joining USV, Albert was the president of del.icio.us through the company’s sale to Yahoo.
Honestly Albert, I'd like to see you working on bringing del.icio.us back.
Regarding the sentence "One that he doesn’t discuss much but that is particularly pertinent to today, is the accretion of laws and regulations." which appears in the review: I think all these laws and regulations can be regarded as being part of the increasing bureaucracy that Tainter identifies as the cause of eventual collapse of complex societies.
One thing I would add, in addition to the diminishing marginal return of bureaucracy that Tainter mentions, is the tendency of contemporary administrations to create more bureaucracy even when their own stated goal is to reduce bureaucracy or increase efficiency, for example by trying to collect and compare precise numbers that show how they have become more efficient at a task that is in essence not really quantifiable. A related book I recommend particularly in that regard is Le Règne de la Quantité et les Signes des Temps by René Guénon.