On Fragility

Well, in a long conversation on the fragility of our civilization with commenter Boonton, one point of contention is apparent. Mr Boonton thinks that the “inflection point” in economic, i.e., the rise of technology in the late 19th century means that comparing today’s culture and civilization to those before is a apples/oranges comparison. Now, everything is different. I demur.

What features characterize today’s technological culture:

  • It is highly interconnected.
  • That interconnection is fueled and aided by high speed cheap transportation.
  • Continued technological advancement is essential.
  • Population levels are staggering when compared earlier eras.

Western Rome fell. It was highly connected and had, for its day, cheap transportation with the Roman road system. Yet it fell, and standards of living and population levels dropped precipitously. The statement “standard of living dropped” this cannot be emphasized enough. Roman era was quite wealthy. Technology that existed, for example examining simple wares like fine china was not eclipsed until the 18th or 19th century. Literacy was almost universal in Rome, even the poor and the slaves could read. Charlemagne was illiterate … and a king, the first “Holy Roman Emperor.” Literacy levels of the Roman era were also not eclipsed in the West until … the 18th or 19th century.

Examine the pottery situation for a moment in the Roman era. Pottery shards happen to be a refuse item which survives for archaeologists to find. In Britian, after Rome retreated something quite surprising happened. Pottery vanished. A potters wheel is conceptually quite a simple thing. But it takes a little time to master. It takes just a little infra-structure to maintain. But … the culture that survived in Britain in the post-Roman times had not the wherewithal to do so.

The only holdout and exception then is technology. How fragile then is technology. It is assumed by many that text and our written records, which are in fact robust and repeated and kept in many places, will insure that our technological advancement and prowess is secure. Things however may in fact not be a secure was we imagine. For it is not the written record on which most of our technology rests but instead of on the unwritten and ineffable expertise of those keeping industrial technological machines running and improvements coming. Michael Polanyi notes the example of the German sale to Hungary of a light bulb manufacturing process. The machines were duplicated, the process written down, and training was completed. Two years after the installation was completed … the machine still had yet to produce a single working bulb. Why? Because the people running the machine were not able to transfer the knowledge of how to run the machine elsewhere.

Our industrial processes and indeed our academic scientific culture is ineffable. It is a culture transmitted by master to apprentice. It depends not only on the skills transferred but cultural norms and values which have to be assumed successfully by the student in order for the continued progress of technology, of science, and academic excellence.

Additionally there are hundreds of thousands, if not many milions, of interlocking industrial components which are required for our civilization to continue. Most of these have multiple sources. Many of these (thousands) are essential, the loss of just one, for example high power/voltage step down transformers, would spell disaster. It is likely that many of these thousands of essential cannot-live-without components, of which we are not really aware in our daily lives, depend on just a few experts to continue their production maintenance, and improvement. One pandemic could wipe out a number of experts in many of these components and … it is not implausible that for some few components the expert base might be lost. Then the social unrest of the pandemic would be acerbated with a failure of one or more key infrastructure components keeping things running. Which in turn causes, because of our very high population levels, starvation and deprivation … which causes the loss of more components and bam! Most of us, just like the survivors of the Western Roman region will be back at pre-civilization early iron age levels.

It might not be a pandemic of course. Our worldwide economies are tightly linked. A monetary crises might cause civil unrest. The resultant violence might leave us missing the people needed to replace the lost infra-structure in the wake of just that. Right now there are some who suggest that the academic industry is the next bubble, which might pop under the stress of the current economic woes. This might not leave the scientific culture which in part depends on university cultural elements intact. If advancement of technology ceased … do we depend on continued technological improvement or not? Our culture is dependent on cheap oil. While it is a matter of debate how long cheap oil will persist … it is not really a debate over that it will at some time cease to be cheap. When, is debated. That it will become dear is not. The unrest that might arise on transition from an oil based civilization to a petroleum-is-expensive one, like the other events noted above could be the proverbial straw, breaking the back.

The point is that there are still striking similarities between our culture and the Roman one. It failed … and perhaps a lesson there to be learned is that our time of peace and prosperity is not likely to be as permanent, nor is as robust as we pretend.

Leave a Reply

Your email address will not be published. Required fields are marked *

14 comments

  1. Boonton says:

    I would add a difference in that the Malthusian trap was passed over after the ‘inflection point’. Unlike all previous periods, population did not immediately expand to consume a new source of production….or instead population did increase but production increased so much faster that a portion of new income could be used for additional investment rather than immediate consumption.

    Western Rome fell. It was highly connected and had, for its day, cheap transportation with the Roman road system. Yet it fell, and standards of living and population levels dropped precipitously.

    I refer again to the graph at http://krugman.blogs.nytimes.com/2009/07/01/a-bit-more-on-malthus/

    Western Rome was great for its day. As you can see going from 1000 BC to today Rome was barely a blip. The argument isn’t whether or not you can tell a difference between Europe of 50BC and Europe of 900 AD (you can), but does that difference make a hill of beans compared to say 1700 versus 1900? It doesn’t.

    Additionally there are hundreds of thousands, if not many milions, of interlocking industrial components which are required for our civilization to continue

    And interestingly this is what makes civilization non-fragile IMO. If one bulb factory blows up others are in place. If all bulbs are suddenly found to be highly toxic, well there’s led bulbs, and flourscent bulbs and so on. Yes plop 50 people on an island and they will have a rough time of it. 500 million people, though, are very difficult to send back to the dark ages.

    This is not to say it couldn’t be done. I’m sure a wordwide extinction event, for example, could do it. But that’s not what we are talking about when we discuss fragility. Fragility means easily destroyed or damaged. The opposite doesn’t mean indestructable.

    One pandemic could wipe out a number of experts in many of these components and … it is not implausible that for some few components the expert base might be lost.

    A pandemic that targets only people with a certain knowledge base? That makes for an interesting sci-fi novel….say a virus that kills all C++ programmers (or bloggers for that matter)….but since we have so many different types of people it is unlikely to happen unless you have a pandemic that wipes out huge portions of the population….but that’s not a ‘fragile’ type event.

    It might not be a pandemic of course. Our worldwide economies are tightly linked. A monetary crises might cause civil unrest. T

    Possibly but we had numerous hyperinflations worldwide and while they do cause turmoil civilization itself seemed to hold together. The Great Depression, likewise, did not cause civil wars in countries like the US, UK etc. Evidence of fragility remains lacking here.

  2. Mark says:

    Boonton,
    I don’t think that plot you point out is credible. It doesn’t mesh with data I’ve seen elsewhere. For example, the The Fall of Rome: And the End of Civilization I cited earlier cites archaeological comparisons of cow bones dated before during and after the Roman era indicate that the size of the cows (which is related to the resources and knowledge available to care for them) after the fall sank to levels comparable to either late iron or bronze age levels (I’ve lent my copy out). Essentially there was a drastic change in the economic well being of everyone during this period. You graph either is looking at populations that miscount and miss this, is framed in a way that minimizes these drastic differences, or is just based on faulty data.

    500 million people, though, are very difficult to send back to the dark ages.

    I disagree. You can’t just make unsubstantiated claims of your case and think your making a contribution to the argument. My claim is that in times of crises 6 billion mouths to feed in an short term economic crises that can only feed 1 billion will lead to a political situation that will make economic recovery impossible.

    A pandemic that targets only people with a certain knowledge base? That makes for an interesting sci-fi novel….say a virus that kills all C++ programmers (or bloggers for that matter)….but since we have so many different types of people it is unlikely to happen unless you have a pandemic that wipes out huge portions of the population….but that’s not a ‘fragile’ type event.

    Uhm, that’s not what I’m saying. C++ “experts” number in the many thousands so it is unlikely that C++ expertise will be a weak point. My point was that there are essential elements or components in which the expert number is in the tens not the thousands. And that the number of this sort of essential (for which there are no reasonable alternatives) but required components are likely in the hundreds. To restate there are hundreds of essential components to our civilization for which the number of experts sustaining our knowledge in that field are in the tens. A pandemic doesn’t have to “target” this type of expert. Just a normal statistical fluctuation would do to wipe one or two o these out.

    The Great Depression was not a global hyperinflation. I’m not sure why you count it as such.

    A great pandemic, as you know, many experts say is past due as a result of our population density and travel practices.

  3. Boonton says:

    The Great Depression was not a global hyperinflation. I’m not sure why you count it as such.

    You said monetary crises and the GD is seen by many economists as being caused by a contraction of the monetary base…hence it’s a monetary crises just as hyperinflation is a crises caused by too much money.

    To restate there are hundreds of essential components to our civilization for which the number of experts sustaining our knowledge in that field are in the tens.

    You need to find an example which is essential, has not substitutes and can be wiped out by the loss of a few hundred or thousand experts. Hungry, for example, couldn’t get the light bulb factory to work but they still had light and civilization. There are areas where the loss of a few thousand people may inhibit a technology for a generation…but I’m not sure its both an essential area and has no substitutes. The Space Shuttle, for example, might be grounded but its importance to civilization is marginal.

    A great pandemic, as you know, many experts say is past due as a result of our population density and travel practices.

    Note though that when European explorers made it to the Americas the native population suffered massive plagues. Europe and Asia had a lot more density and travel via the silk road. While that might have technically made a pandemic more possible it also made the population more disease resistant.

  4. Boonton says:

    I think we have to understand here that fragility is a quality. Something can be more fragile or less fragile than something else. The fact that Mark establishes that some civilizations fell, only demonstrates that civilization is not indestructible.

    Mark also identifies that as civilization grows bigger, the distance it can fall also grows. A super-pandemic that kills 99% of humanity will kill more when there’s 7 billion people than when there was only 50,000. That doesn’t tell us, though, that civilization is fragile. The small group of 50,000 can be wiped out by a pandemic that takes out 25,000 people but a civilization can asorb that much death without even noticing a blip in their statistics.

    Larger civilizations IMO are less fragile. We have one large meta-civilization that can be called the Eurasian civilization. For the last several thousand years, there’s always been a civilization here. There have been moments of fall (Rome, the middle east, India) but there’s never really been a period of non-civilization. In South America you likewise had a very large civilization that fell to another larger civilization.

    This makes sense IMO because human vices are rather predictable and stable. The motive for corruption and decadence is universal. The fact that we are ignorant but like to think of ourselves as smart is likewise true across the board. Huge civilizations have developed defenses against these weaknesses hence they are able to handle most of the blows that strike a civilization from the inside. There are outside factors like pandemics, comet strikes, invasion by outside forces….but here Mark has to confront a question: What is stronger defense or offense?

    As a civilization becomes bigger, one side of the coin is that there are more targets that could be hit to do great harm to the civilization. But the other side of the coin is that as a civilization gets larger, its defenses also get larger. For example, only 3% of our pop. provides agriculture. That may be a weakness (take out that 3% and the other 97% will have a hard time getting up to speed on how to operate a farm). But it also improves our defenses. Unlike having 90% in farming, we can have 5% of the population devoted to R&D against germs. We can have a full time military with highly trained specialized forces who do NOT have to take 6 months a year to work the farm or who do not have to be fed by pillaging other farms.

    As civilization grows, then, both offense (the capacity to strike down a civilization) and defense (the capacity to ward off such an attack, human or natural) grow. Fragility then can be thought of as a metric of offense/defense. That would make a fascinating graph. AS the ratio trends less than 1 fragility declines, if it is greater than 1 fragility increases.

    I wouldn’t be surprised if this wasn’t a straight line. I suppose there might be some sized civilizations that have fragility of less than 1 and then more than 1….perhaps if super large civilizations might revert back to less than 1 (a planet wide government? the Federation of Star Trek? The Galactic Empire of Star Wars?)

  5. Mark says:

    Boonton,
    Re Depression and hyperinflation … you linked in successive sentences hyperinflation and the Depression. That’s all.

    Hungry, for example, couldn’t get the light bulb factory to work but they still had light and civilization.

    Yes, the Hungary light bulb example is a demonstration of the ineffability of the knowledge that keeps us ticking not that light bulbs are an essential component or one that is sustained by a small number.

    Look at a genetic example … with a civilization as an evolving creature. More specialization is helpful. Brain cells can not digest or gather oxygen or excrete and so on. Yet, if a few cells in the islet of Langerhans fail … oops the creature dies. Specializations leads to hidden fragility. Looking at a human you wouldn’t realize that just a patch of cells here, a pathway there blocked and … boom! death. You can’t kill a slime mold so simply, each cell is less specialized. Many far far simpler organisms are less fragile than we and can survive a far greater range of conditions. Gould said that the environmental activists have it wrong and that life is not threatened by human activity … just that human life might be threatened.

    A human can do more things it is however less stable. If you want to maximize robustness of your society (as a politician) you’d make different choices. Progressives by nature don’t find robustness a feature on which they consider policy. Conservatives (at least in my case) do.

    It is a defining difference. And I’m not arguing that you’re nuts to disregard fragility when you consider policy. I just think you’re making an error of judgement and not one for which you have no reasons for making your choice. But … if you are honest for a moment, I think you (should!) admit that I have valid reasons for thinking fragility is important.

    I named a suggestion for a “crucial” element, high power high voltage transformers. I doubt the market is big or that there are more than a few hundred people with the expertise to manage the manufacture process or make improvements in the same. My point is that these elements are not obvious to us now … when things are working correctly, just like our insulin producing cells.

  6. Mark says:

    Boonton,
    I’d add I’m not a biologist, slime mold might not be the correct example of a very robust colonial species … of which I’m pretty sure there are many. If biology is in your expertise I’m sure you can supply dozens of good examples of a robust simple organism.

  7. Boonton says:

    Re Depression and hyperinflation … you linked in successive sentences hyperinflation and the Depression. That’s all.

    They are linked in that they are examples of monetary crises. A depression is caused by too little money, hyperinflation by too much.

    Look at a genetic example … with a civilization as an evolving creature. More specialization is helpful. Brain cells can not digest or gather oxygen or excrete and so on….

    In other words we have a trade off. Greater specialization offers the possibilities for better defense. Likewise it offers greater opportunities for a decapitation (offense). I would say that civilization tends to increase the former faster than the later increases.

    It is a defining difference. And I’m not arguing that you’re nuts to disregard fragility when you consider policy. I just think you’re making an error of judgement and not one for which you have no reasons for making your choice. But … if you are honest for a moment, I think you (should!) admit that I have valid reasons for thinking fragility is important.

    Old comfortable ideas can also IMO be a cause of fragility. The Decline and Fall of Rome is a romantic and melodramatic analogy but its over played (I spent this afternoon watching Bill Maher interview Gore Vidal who spun his own comparisions between the US and Rome, while he approached it from the left side I found it dull, hackneyed, and fundamentally uninteresting).

    I named a suggestion for a “crucial” element, high power high voltage transformers. I doubt the market is big or that there are more than a few hundred people with the expertise to manage the manufacture process or make improvements in the same.

    I’d be curious to know how easy such a ‘decapitation’ would be. Are these few hundred people all in the same area? Is it easy to get a list of them? Are there really a few hundred or is it more like a few hundred but there’s a few thousand who could get themselves online with them in a few months if there was suddenly a huge demand for such experts?

  8. Mark says:

    Boonton,

    I’d be curious to know how easy such a ‘decapitation’ would be. Are these few hundred people all in the same area?

    Two things. This is just one example, I figure there are hundreds. The approx. one hundred individuals don’t need to be scattered. The point is in the event of a pandemic a statistical fluctuation would be likely to eliminate all of the ‘experts’ in one or two such essential fields.

    And yes, a human has more defenses and capabilities than a slime mold. The point is that its continued existence is likely more fragile.

    but there’s a few thousand who could get themselves online with them in a few months if there was suddenly a huge demand for such experts?

    How in a seance? The point is the knowledge is ineffable and the remaining carriers of that knowledge are dead. The knowledge is transmitted by apprenticeship.

    In the time to “recover” from the demand, crucial elements of our society (power? food?) are lost. Panic and disorder and overpopulation do the rest to enact complete breakdown.

    Gibbons book may have been influential but it is outdated and contains many errors, not the least of which was his particular bigotry against the Eastern Roman empire.

  9. Boonton says:

    Two things. This is just one example, I figure there are hundreds. The approx. one hundred individuals don’t need to be scattered. The point is in the event of a pandemic a statistical fluctuation would be likely to eliminate all of the ‘experts’ in one or two such essential fields.

    1. Yea they kind of do. It’s conceivable that a pandemic might hit and kill everyone in a single building and if that building happened to house the 100 ‘essential’ experts we’d be in trouble. If those experts are scrattered the odds that all 100 will go down are a lot less likely (granted not zero). This is kind of like the story about the ‘secret formula’ to Coke that only two Execs know and they are never allowed to fly on the same plane. Yes maybe both planes will go down but scattering them reduces the risk.

    2. I would imagine that these experts would be scattered among several nations which would apply different policies to a pandemic or if nothing else a pandemic wouldn’t hit all countries at exactly the same time leaving some room for at least a few to take steps to protect the precious experts should it appear like a pandemic with a 1% mortality rate has, due to a ‘fluctuation’ killed off a huge portion of this small group of experts.

    How in a seance? The point is the knowledge is ineffable and the remaining carriers of that knowledge are dead. The knowledge is transmitted by apprenticeship.

    A general practitioner is helpless in addressing an issue normally handled by a cardiologist? Or for that matter is the cardiologist’s nurse helpless? Do these experts who keep our civilization going work alone in a hollowed out base 100 miles under the ground? Presumably these experts work with dozens of other people who, while not fully apprenticed in the job, have picked up at least a few key elements of it.

    See the issue with your ‘light bulb factory’ ancedote? Yes the factory could get blown up with everyone in it from the top scientist/machinest down to the janitor. Or yes a strange virus could kill all the top bulb machinests in the world. But if its the former, there’s plenty of other factories scattered all over the world. If its the latter there’s almost certainly other people who could step in and at least do a minimially acceptable job as they get up to speed. OK if a virus or bombing destroyed all the workers (and ex-workers) at all the bulb factories in the world at once we’d have a problem. That, though, does not seem to be a fragility but a rather exceptionally difficult strike to happen (either naturally or intentionally). This does not account for the ability of society to adjust its consumption of bulbs (rationing remaining ones, using gas based light, adjusting work hours to take advantage of daylight) to get through the ‘downtime’ as new capacity goes online. Certainly such a diaster would be expensive but it is hardly clear it would shatter civilization.

    Perhaps you should consider the history of warfar. Attacks on cities in WWII and after have deprived large portions of society with many civilized comforts like electricity, water, sanitation etc. Despite this the fact remains that efforts to ‘break’ a country by airpower have mostly failed…even with the exceptional ability the US has developed to do ‘surgerical strikes’ that can take out key targets like key electrical facilities, TV stations, even individual floors of a building….despite this ‘boots on the ground’ are still required to secure a victory.

    More Importantly, I have to raise this issue:

    WTF!!!!! You were talking about the dangers of ‘progressive policies’ like gay marriage and prohibition. Now we are imagining amazingly unlikely pinpoint virus outbreaks that take down key experts? This is like the butterfly effect gone made. So maybe prohibition in 1930 will cause a strange outbreak in 2095. It’s just as likely prohibition in 1930 caused such an outbreak NOT to happen.

    Gibbons book may have been influential but it is outdated and contains many errors, not the least of which was his particular bigotry against the Eastern Roman empire.

    You are probably right, although I have yet to have the opportunity to read him and despite him being outdated I’d probably be better off reading him rather than other more modern books on Rome. Nonetheless, its not the outdated parts that’s the issue its giving the idea too much influence. Yes the ‘fall’ makes a nice dramatic image which one wants to apply everywhere (fall of the US, the EU, the USSR, Microsoft, Enron….etc.etc.) Taken too far (and since Rome was taught to schoolboys in Gibbon’s day and before it was probably overplayed by then) and this becomes more lazy than insightful.

  10. Mark Olson says:

    Boonton,
    Statistical fluctuations require geographical continuity. I didn’t think statistics worked that way.

    More Importantly, I have to raise this issue:

    The statistical fluctuation I noted is just one of many ways in which our civilization is fragile. The point is if it is fragile … do you whack at it with abandon or not?

  11. Boonton says:

    The statistical fluctuation I noted is just one of many ways in which our civilization is fragile.

    You would have been more sensible if you pinned your fragility hopes on a UFO invasion.

  12. Boonton says:

    In other words, if you must have a crappy argument at least give us special effects.

  13. Boonton says:

    Perhaps a good proxy test for your fragility hypothesis is to look at large companies. How many examples of large companies can you think of who were sunk by a ‘statistical fluke’ that takes out a key employee or expert? Given that companies do not have the power to change laws or draft labor or tax or use force like gov’ts (well they usually don’t), they should be *more* fragile.

    To make it a fair test, though, the companies should be large and not premised on an inherently unsustainable business model (Madoff’s ‘fund’ for example or Enron’s systematic balance sheet fraud) and have some degree of diversity in its business model (in other words it doesn’t place all its eggs in one basket such as Martha Stewart’s company which puts its eggs in Martha being alive, healthy and having a clean name).

  14. Boonton says:

    A pandemic doesn’t have to “target” this type of expert. Just a normal statistical fluctuation would do to wipe one or two o these out.

    One final thought, what Marc is describing here is essentially a six or seven sigma event. To put it in proper persepctive, imagine saying there’s 100 exclusive experts who are dedicated to some essential aspect of civilization and one morning they all slip and fall in the shower resulting in death or coma for all.

    civilization is supposedly fragile because it can’t handle such an event. Even if it can’t I wouldn’t exactly say this is fragility. That’s kind of like saying your house’s foundation is fragile because it couldn’t survive a quantum fluctuation that caused 3/4 of the concrete to leap to the other side of the world.

    Of course, I don’t accept that 100 such exclusive experts exist. Even if such essential ‘experts’ were to suddenly be taken out I suspect co-workers who may not technically have all the qualifications of the former experts will be called into duty to help out. In the more recent post on healthcare Marc asks if we would want to see cosmetic surgeons doing emergency appendectomies. Well a cosmetic surgeon probably could do an acceptable if not optimal job of an emergency appendectomy if no one else was around. Likewise a nurse could probably deliver a baby if no Obgyn was available….if need be even a woman who has given birth might be able to do a good job if need be.

    The ‘Death Star’ model (one well placed shot blows the whole thing up) of civilization might be simple but I don’t see the evidence that it is accurate.