Joseph Tainter - Human Resource Use: Timing and Implications for Sustainability

Joseph Tainter, a Professor in the Department of Environment and Society at Utah State University, and author of the seminal work "The Collapse of Complex Societies", recently gave a speech on complexity and resource use at the 94th Annual Meeting of Ecological Society of America in New Mexico: (Conference theme: Human Macroecology: Understanding Human-Environment Interactions Across Scales). The speech, 'Human Resource Use: Timing and Implications for Sustainability', based on a forthcoming paper, is reprinted below with the authors permission.

(*NOTE: Dr. Tainter uses the word 'complexity' many times in the below speech, without defining it. I asked him for a definition of the term and he pointed me to his 2006 paper, "Social Complexity and Sustainability", that gives the following definition, from the field of Anthropology.)

"Complexity is more challenging to define singularly. As it has become a popular topic in recent years, competing definitions of complexity have made it difficult to clarify the concept. The nuances of these different conceptions are not helpful in understanding the relationship of social complexity to sustainability; those interested may consult the growing literature on this topic. To understand sustainability, it is useful to conceptualize complexity in human social systems as differentiation in both structure and behavior, and/or degree of organization or constraint (Tainter, 1988, 2000b; Allen et al., 2003; see also Allen et al., 1999). Social systems vary in complexity as they diversify or contract in structure and behavior, and/or as they increase or decrease in organizational constraints on behavior."

___________________________________________________________________________________________________

Human Resource Use: Timing and Implications for Sustainability

Joseph A. Tainter

Department of Environment and Society, Utah State University, Logan, Utah 84322, U.S.A.

____________________________________________________________________________________________________

Few questions of history have been more enduring than how today’s complex societies evolved from the foraging bands of our ancestors. While this might seem of academic interest, it has important implications for anticipating our future. Our understanding of sustainability depends to a surprising degree on our understanding of the human past. My purposes today are to show that the conventional understandings of cultural evolution are untenable, as are assumptions about sustainability that follow from them, and to present a different approach to assessing our future.

Cultural complexity is deeply embedded in our contemporary self image. Colloquially it is known by the more common term “civilization,” which we believe our ancestors achieved through the phenomenon called “progress.” The concepts of civilization and progress have a status in the cosmology of industrial societies that amounts to what anthropologists call “ancestor myths.” Ancestor myths validate a contemporary social order by presenting it as a natural and sometimes heroic progression from earlier times.

Social scientists label this a “progressivist” view. It supposes that cultural complexity is intentional, that it emerged through the inventiveness of our ancestors. Progressivism is the dominant ideology of free-market societies. But inventiveness is not a sufficient explanation for cultural complexity, which requires facilitating circumstances. What were those circumstances? Prehistorians once thought they had the answer: The discovery of agriculture gave our ancestors surplus food and, concomitantly, free time to invent urbanism and the things that comprise “civilization”–cities, artisans, priesthoods, kings, aristocracies, and all of the other features of early states.

The progressivist view posits a specific relationship between resources and complexity. It is that complexity develops because it can, and that the factor facilitating this is surplus energy. Energy precedes complexity and allows it to emerge. There are, however, significant reasons to doubt whether surplus energy has actually driven much of cultural evolution.

One strand of thought that challenges progressivism emerged in the 18th and 19th centuries in the works of Wallace (1761), Malthus (1798), and Jevons. The economist Kenneth Boulding derived from Malthus’s essay on population three theorems: the Dismal Theorem, the Utterly Dismal Theorem, and the moderately cheerful form of the Dismal Theorem. The Utterly Dismal Theorem directly challenges the progressivist view:

Any technical improvement can only relieve misery for a while, for as long as misery is the only check on population, the improvement will enable population to grow, and will soon enable more people to live in misery than before. The final result of improvements, therefore, is to increase the equilibrium population, which is to increase the sum total of human misery (Boulding, 1959: vii [emphases in original]).

The implication of this strain of thought is that humans have rarely had surplus energy. Surpluses are quickly dissipated by growth in consumption. Since humans have rarely had surpluses, the availability of energy cannot be the primary driver of cultural evolution.

Beyond a Malthusian view, there is another factor that undermines progressivism. It is that complexity costs. In any living system, increased complexity (involving differentiation in structure and increasing organization) carries a metabolic cost. In non-human species this is a straightforward matter of additional calories. Among humans the cost is calculated in such currencies as resources, effort, time, or money, or by more subtle matters such as annoyance. While humans find complexity appealing in spheres such as art, music, or architecture, we usually prefer that someone else pay the cost. We are averse to complexity when it unalterably increases the cost of daily life without a clear benefit to the individual or household. Before the development of fossil fuels, increasing the complexity and costliness of a society meant that people worked harder.

The development of complexity is thus a paradox of human history. Over the past 12,000 years, we have developed technologies, economies, and social institutions that cost more labor, time, money, energy, and annoyance, and that go against our aversion to such costs. Why, then, did human societies ever become more complex?

At least part of the answer is that complexity is a basic problem-solving tool. Confronted with problems, we often respond by developing more complex technologies, establishing new institutions, adding more specialists or bureaucratic levels to an institution, increasing organization or regulation, or gathering and processing more information. While we usually prefer not to bear the cost of complexity, our problem-solving efforts are powerful complexity generators. All that is needed for growth of complexity is a problem that requires it. Since problems continually arise, there is persistent pressure for complexity to increase.

Cultural complexity can be viewed as an economic function. Societies and institutions invest in problem solving, undertaking costs and expecting benefits in return. In problem-solving systems, inexpensive solutions are adopted before more complex and expensive ones. In the history of human food-gathering and production, for example, labor-sparing hunting and gathering gave way to more labor-intensive agriculture, which in some places has been replaced by industrial agriculture that consumes more energy than it produces. We produce minerals and energy whenever possible from the most economical sources. Our societies have changed from egalitarian relations, economic reciprocity, ad hoc leadership, and generalized roles to social and economic differentiation, specialization, inequality, and full-time leadership. These characteristics are the essence of complexity, and they increase the costliness of any society.

In the progressivist view, surplus energy precedes and facilitates the evolution of complexity. Certainly this is sometimes true: There have been occasions when humans adopted energy sources of such great potential that, with further development and positive feedback, there followed great expansions in the numbers of humans and the wealth and complexity of societies. These occasions have, however, been so rare that we designate them with terms signifying a new era: the Agricultural Revolution and the Industrial Revolution. It is worth noting that these unusual transitions have not resulted from unbridled human creativity. Rather, they emerged from solutions to problems of resource shortages, and were adopted reluctantly because initially they created diminishing returns on effort in peoples’ daily lives.

Most of the time, cultural complexity increases from day-to-day efforts to solve problems. Complexity that emerges in this way will usually appear before there is additional energy to support it. Rather than following the availability of energy, cultural complexity often precedes it. Complexity thus compels increases in resource production. This understanding of the temporal relationship between complexity and resources has implications for sustainability that diverge from what is commonly assumed. I will explore these implications shortly. It is useful first to present a historical case study, the Western Roman Empire, that illustrates these points.

The Roman Empire collapsed in the mid 5th century A.D., but its last 200 years of existence had been a reprieve. It had been nearly destroyed in the 3rd century. In the half-century from 235-284 the empire was repeatedly breached by invasions of Germanic peoples from the north and the Persians from the east. When these invaders were not being repelled, Roman armies were fighting each other in the service of would-be emperors. Many cities were sacked and productive lands devastated. For a time, rival empires broke away in the east and the west. It seemed that the Roman Empire would not survive much longer.

The Roman government had a clear sustainability goal: the survival of the empire. In response to the crises, the emperors Diocletian and Constantine, in the late third and early fourth centuries, designed a government that was larger, more complex, and more highly organized. They doubled the size of the army. This was very costly. To pay for this sustainability effort, the government taxed its citizens more heavily, conscripted their labor, and dictated their occupations.

With the rise in taxes, population could not recover from plagues in the second and third centuries. There were chronic shortages of labor. Marginal lands went out of cultivation. Faced with taxes, peasants would abandon their lands and flee to the protection of a wealthy landowner. The Roman Empire survived the 3rd century crisis and achieved two centuries of sustainability, but at the long-term cost of consuming its capital resources: producing lands and peasant population. When crises emerged again in the late 4th century, the empire lacked the resources to respond adequately and in time collapsed.

The Roman Empire is a single case study in complexity and problem solving, but it is an important and representative one. It illustrates the basic process by which societies increase in complexity. Societies adopt increasing complexity to solve problems, becoming at the same time more costly. In the normal course of economic evolution, this process at some point will produce diminishing returns. Once diminishing returns set in, a problem-solving institution must either find new resources to continue the activity, or fund the activity by reducing the share of resources available to other economic sectors. The latter is likely to produce economic contraction, popular discontent, and eventual collapse. This was the fate of the Western Roman Empire.

This understanding of complexity and resources has implications for understanding sustainability. Both popular and academic discourse commonly assume that (a) future sustainability requires that industrial societies consume a lower quantity of resources than is now the case, and (b) sustainability will result automatically if we do so. Sustainability emerges, in this view, as a passive consequence of consuming less. Thus sustainability efforts are commonly focused on reducing consumption through voluntary or enforced conservation, perhaps involving simplification, and/or through improvements in technical efficiencies.

The common perspective on sustainability follows logically from the progressivist view that resources precede and facilitate innovations that increase complexity. Complexity, in this view, is voluntary. Human societies become more complex by choice. By this reasoning, we should be able to forego complexity and the resource consumption that it entails. Progressivism leads to the notion that societies can deliberately reduce their use of resources and thereby achieve sustainability.

The fact that complexity and costliness increase through mundane problem solving suggests a different and startling conclusion: Contrary to what is typically advocated as the route to sustainability, it is usually not possible for a society to reduce its consumption of resources voluntarily over the long term. To the contrary, as problems great and small inevitably arise, addressing these problems requires complexity and resource consumption to increase. As illustrated by the Roman Empire and other cases, this has commonly been the case.

Many advocates of sustainability will find it disturbing that long-term conservation is not possible. Naturally we must ask: Are there alternatives to this process? Regrettably, no simple solutions are evident. Consider some of the approaches commonly advocated:

1. Voluntarily Reduce Resource Consumption. While this may work for a time, its longevity as a strategy is constrained by the fact that societies increase in complexity to solve problems. Resource production must grow to fund the increased complexity. To implement voluntary conservation long term would require that a society be either uniquely lucky in not being challenged by problems, or that it not address the problems that confront it.

2. Employ the Price Mechanism to Control Resource Consumption. This is currently the laissez-faire strategy of industrialized nations. Since humans don’t commonly forego affordable consumption of desired goods and services, economists consider it more effective than voluntary conservation. Both approaches, however, lead eventually to the same outcome: As problems arise, resource consumption must increase at the societal level even if consumers as individuals purchase less.

3. Ration Resources. Because of its unpopularity, rationing is possible in democracies only for clear, short-term emergencies. This is illustrated by the reactions to rationing in England and the United States during World War II. Moreover, rationed resources may become needed to solve societal problems, belying any attempt to conserve through rationing. Something like this can be seen in the fiscal stimulus programs enacted recently.

4. Reduce Population. While this would reduce aggregate resource consumption temporarily, as a long-term strategy it has the same fatal flaw: Problems will emerge that require solutions, and those solutions will compel resource production to grow.

5. Hope for Technological Solutions. I sometimes call this a faith-based approach to our future. Members of industrialized societies are socialized to believe that we can always find a technological solution to resource problems. Technology, within the framework of this belief, will presumably allow us continually to reduce our resource consumption per unit of material well-being. Conventional economics teaches that to bring this about we need only the price mechanism and unfettered markets. The flaw here was pointed out by William Stanley Jevons: As technological improvements reduce the cost of using a resource, total consumption will actually increase.

In conclusion, sustainability is not the achievement of stasis. It is not a passive consequence of having fewer humans who consume more limited resources. One must work at being sustainable. The challenges that any society (or other institution) might confront are, for practical purposes, endless in number and infinite in variety. This being so, sustainability is a matter of solving problems.

In the conventional view, complexity follows energy. If so, then we should be able to forego complexity voluntarily and reduce our consumption of the resources that it requires. This approach to sustainability implicitly sees the future as a condition of stasis with no challenges.

In actuality, major infusions of surplus energy are rare in human history. More commonly, complexity increases in response to problems. Complexity emerging through problem solving typically precedes the availability of energy, and compels increases in its production. Complexity is not something that we can ordinarily choose to forego.

Applying this understanding leads to two conclusions. The first is that the solutions commonly recommended to promote sustainability–conservation, simplification, pricing, and innovation–can do so only in the short term. Secondly, long-term sustainability depends on solving major societal problems that will converge in coming decades, and this will require increasing complexity and energy production. Sustainability is not a condition of stasis. It is, rather, a process of continuous adaptation, of perpetually addressing new or ongoing problems and securing the resources to do so.

It is useful to think of sustainability in the metaphor of an athletic game: It is possible to “lose”–that is, to become unsustainable, as happened to the Western Roman Empire. But the converse does not hold. Because we continually confront challenges, there is no point at which a society has “won”–become sustainable in perpetuity, or at least for a very long time. Success, rather, consists of staying in the game.

I think this piece goes right to the heart of the issues of our time - we ignore long term socio-economic problems, and when the long-term eventually becomes the short-term, we throw resources and technology at the problems. This is not sustainable, though given our finite lifespans, steep discount rates, and self-deception we may perceive it as so at the time. I.e. each day 'seems' sustainable, until it doesn't.

I have a few other comments.

Since humans have rarely had surpluses, the availability of energy cannot be the primary driver of cultural evolution.

Exactly. I suspect cultural evolution had much to do with a push/pull dynamic that combined time with environmental stimuli in ways that pulled via a) competition to move up the mating ladder (sexual selection) and b) combining/trying various 'technology' iterations that would best match the neural reward pathways that had existed from previous ancestral generations (novelty, salience, aesthetic...). Cave paintings, tools, colors, new language, dance, skins/clothing all were probably developed via trial and error and contributed to the success and pleasure of those that had the most mating success, on average, in each group of generations. These 'discoveries' would have largely been maintained via a cultural ratchet effect. In that sense, I would argue that complexity arose via cumulative ethological/behavioral conditions linked to sexual selection and reward feedback, and became 'stickier' during times of more abundant energy



In actuality, major infusions of surplus energy are rare in human history. More commonly, complexity increases in response to problems. Complexity emerging through problem solving typically precedes the availability of energy, and compels increases in its production. Complexity is not something that we can ordinarily choose to forego.

And I would argue that as fast as population has popped in the above graphic, complexity AND our human habituation towards more and more stimuli (in industrial society) and ways to compete has increased at faster pace. Ergo, to keep this trajectory we would need fantastic amounts of energy, not just a linear increase.

It is possible to “lose”–that is, to become unsustainable, as happened to the Western Roman Empire. But the converse does not hold. Because we continually confront challenges, there is no point at which a society has “won”–become sustainable in perpetuity, or at least for a very long time. Success, rather, consists of staying in the game.

I understand and agree. It suggests series of population pulses into the future, and given the one time nature of this fossil subsidy (barring some nuclear miracle), smaller amplitudes over time.

One concluding caveat is that our ancestors didn't have access to Dr. Tainters and others historical work, nor access to human and animal ethology, neuroscience etc. findings. Is it possible to build in such a pulse with buffer towards rebound to more complexity in future energy and resource budgets? The concept of complexity needs to be redefined using human ethology/behavioral science. But it might be too complex...

Nice graphic!

Nate,
"In actuality, major infusions of surplus energy are rare in human history."

Over the last 70 years our technology has unlocked 3 major new energy resources; wind power, solar power and nuclear fission.These all have the potential to be many times larger than all the fossils fuel created over last 4Billion years.

Solution #5, Neil?

Funny how the unlocking of those "energy resources" coincided with the exploitation of oil and gas.
Take away the FF's and you would have had and will have jack S#*t.

Dreams of windmills, nuclear reactors, and solar arrays reproducing themselves with their own energy is nice but hopes, expectations and theories don't pay the bills.

The reason we exploded to six and a half billion was the exploitation of oil and the means to feed us. Deal with that with "3 major new energy resources".

And, to add to that, if we were to utilize the other energy sources in the same way as FF's the earth's population would explode again...15-20 billion, more?

This doesn't help sutstainability, it destroys it...energy isn't the only constrained resource. Sustainable use of land, water, etc. are also compromised by an exploding population made possible by increased available energy. We already used the equivalent of 1.4 earths (or greater) now.

Humanity has already crossed the line, has already overshot the carrying capacity of this planet.

What exactly is “problem-solving” at a societal level? Does it make a difference if we look at things from a biological / ecological perspective level? A systems perspective? Culture is subject to the same evolutionary principles that govern all other complex systems in the biosphere. The problem to be solved by all living systems is how to adapt available resources (energy and material) to their own reproduction. Tainter’s claim that energy is not the driving force of cultural evolution is almost laughable (if it weren’t coming from him). His view of the issue is anthropocentric through and though. The final end of human “problem solving” understood in these terms is infinite growth. Of course sustainability is impossible long term – it defies entropy and the maximum power imperative– I’m surprise you signed on to this so un-critically Nate.

Hmmm. I didn't sign on so much as agree that throwing resources and technology at a problem, keeping demand and other drivers constant, will make things worse - this is really nothing new.

What is problem solving at a societal level? To me it means using the sunk cost of built infrastucture and cultural values to dictate a societies future path as opposed to a real matching of supply and demand balance sheets. In our case it means growth. And infinite growth is only the problem THIS time around - next pulse we might have 20-30% the population (who knows?), higher technology, less resources overall but more resources per capita and never follow an infinite growth paradigm due to what we (will) learn this time around - but we'll still be limited by the fact that the neural drivers that got us this far are not 'satisficers', but more related to 'wanting'. Until we address that somehow I expect it will be boom bust.

Thats why my super big picture model finally concludes that humans must continue to evolve. We are either dead as a species or not done yet evolving.

Whats interesting is we probably can't make the next step in evolution unless human populations become highly fragmented and isolated. How this happens with us continuing to remain our technical level is far from clear.

Best I can guess either some humans manage to move into space and evolve relatively isolated from the ones left on earth or we probably eventually go back to the stone age. And of course next time around climbing back out will be a lot harder given the timescale to replace the resources we have used up.

A third possibility is we actually use our technology to modify our own genetics i.e we evolve in place. This probably would include some sort of unified advanced AI stuff.

Regardless it seems some radical changes are required resulting in a new post human species one why or another. The human experiment either restarts moves or starts modifying itself. No matter what humans like we exist today at least in the collective are done.

I personally wonder if we would really survive a restart of the stone age I'd have to guess that under those conditions humans would evolve/devolve to fill the missing niches in the ecosystem. One would guess that given the genetic devastation we have created that our own species would probably become the primary genetic material to repopulate the devastated planet. Mother nature has to work with what it has and we can guess that given the adaptability of humans that our own species is likely to be the only material around.

That does not suggest that we will actually evolve the second time around to be super intelligent in fact it suggest that we will just keep enough intelligence to ensure the world is a much deadlier place full of former human smart rats if you will. Smart enough to make life miserable for any branch of humanity trying to obtain its former glory but not smart enough to make a second ascent.

On the resource side it also seems we would be looking at millions of years with that route with a repeat of humanity taking several million years to replenish the resource base and restart the experiment.

Thus outside of our genetic contribution I really feel like a descent into the stone age is far more drawn out than most people would think is effectively a clean slate solution. The final result would not be human anymore than the ancient mammals that survived are human.

Now of course if this happens this new intelligent species would at least be faced with the weathered remains of our oil wells and strange imprints from some of our cities. One suspects that even millions of years would not be enough for some future species to not be able to find puzzling clues of a former race esp if they reach the moon.

It actually sucks knowing your probably the first failure in the experiment with intelligence :(

Looking even farther out given the increasing knowledge of the number of planets in the galaxy the chance for life continue to increase fairly dramatically. The lack of known intelligent life outside of our own planet becomes increasingly more chilling. It looks like for whatever reason that few species are able to pass through this gauntlet of the first rise of intelligence to become space faring species.

Our own experience to date point out to intelligence probably being briefly present but unable to pass the first major stumbling block.
It seems given what we know about ourselves that the percentage of species that actually survive their first bought with intelligence are vanishingly small.

By the time they reach the level of complexity required to self evolve and recognize their own species short comings it seems we self destruct. Without more data it seems this internal problem is the most serious one.

Drawing lines with only one point is not even guessing but so far given what we know about planets and ourselves it seems that the chances of us making it are very very close to zero. And last but not least the sad part is if some other species evolves to intelligence millions of years later the slate if you will will be clean enough to ensure that they to will face exactly the same trap. So not only does the system fail given the nature of it it tends to fail again and again at the same point.

Now of course even with this small is not zero one has to suspect that somewhere some species have succeeded however I suspect that most only make it after repeated failures and they will collectively realize that almost all experiments in intelligence end in failure. And I also suspect that if such beings exist they will also know of other spectacular failures where interventions occurred. Probably with the rabid "saved" species taking out the truly intelligent species along with large swaths of a galaxy if not the whole thing. Given the saved species did not evolve past their highly viral larval stage.

Bottom line is if these concepts are correct we probably ain't getting any help even if it it exists we are on our own the chances of a "rapture" are pretty damned low.

So is it "intelligence" or a drastic artifically-extended phenotype that's the problem? I've spent a lot of time with dolphins and they're intelligent by reasonable criteria, and have seemingly been so for a lot longer. Given an opportunity, they'll use a computer to control their environment, etc. But "fire" and "thumbs" and "exosomatic information storage" were never an option for them.

just sayin'...

Well if we manage to wipe ourselves out or close to it Doplhins could return to land and create a intelligent species.

You also have lava flows and vents in the ocean so extreme heat source are available. Chemistry is chemistry and we have no idea how far biological tinkering could go if a species figured it out and had a few million or even a billion years so simple intelligence coupled with and eventual development of biology is probably sufficient to develop advanced technologies given time.

And of course who can imagine what and alien biology could be capable of we don't even know the limits of our own DNA code and you could have more versatile ones.

When your talking about millions and billions of years and millions and billions of start and planets the smallest chance thats not impossible probably happened somewhere sometime.

Look at it this way the better we become at finding planets the more we find.
The probability of life bearing planets thus steadily increases and with it the probability of intelligence increases. Chemistry works in such a way that very credible scenarios for the development of technology under a variety of conditions even without fire make sense at worst the slower biological route is open to all.

Even after a few thousand years of technical advancement we already have several credible designes for interstellar space ships that can travel a reasonable fraction of the speed of light. The most exotic antimatter is even technically within our grasp today. There is no fundamental reason we could not do it. In fact some sort of micro-blackhole is also almost within our grasp.

http://dsc.discovery.com/news/2008/09/10/black-hole-cern.html

And our species is young.

The galaxy is about 100,000 light years thus even traveling at only a few percentage of the speed of light which should be readily improved within a few million years say 10 at most and intelligent species would easily have detected and monitored any world that supported life.

We don't have any credible evidence that aliens have visited us and certainly no obvious overt visits and I'd argue that secret visits are both questionable and irrelevant.

Take this and then take Tainters arguments and many peoples own recognition of how humans behave and it suggests that making the leap for intelligence to some sort of long term stable society is incredibly hard. If it was easy then we would be living with aliens and we are not. And all our own investigations seem to point to very difficult issues for our species. So internally we know its hard and externally the lack of aliens validates this assumption.

Most advocates that these sorts of arguments are bogus generally rest at the end of the day on either some sort of innate ability to change or on the believe that technical advances will save the day.

If the believe in technology is correct then where the heck are the aliens ?

If believe in our ability to change is correct then same question where are the aliens ?

If you think that the truth is only a vanishingly small number of species actually make it then its sensible that the much smaller population of potential alien species has adopted a hands off policy. And we could well be the first kids on the block since your now talking millions of years to try several times to develop a new species and stars have a limited lifetime.

If you assume it takes 50 million years between intelligent species and the average time a planet is habitable by intelligent species is say two billion years and then say only one in a billion make it then suddenly you don't have a lot of aliens.

In fact what you have is the result that intelligence in and of itself does not promote long term species survivability over any other biological approach. In fact other approaches such as say that taken by sharks look like the win in the big picture.

We assume that by having become intelligent somehow we have beaten extinction yet overall in my opinion it seems this is not true.

Which means of course somewhere between when a species becomes intelligent and when the expand sufficiently throught the stars to beat extinction lies a huge obstacle that few overcome.

And Tainter and others continue to point out the existence of the obstacle just observing our own species.

Bottom line is with all the evidence gathered so far the result is technically adept species seem to face the problem of complexity and most fail.

The only way out seems to be to actually recognize the problem yet the vast majority of people would not even consider the issue as anything but something from the lunatic fringe despite the obvious failure of many human civilizations in the past.

I'd suggest at the minimum that the end of the oil age is the first great trial for the human species as a whole. Given we have nuclear and biological weapons it may be our last if not then we do know now that even if we have not reached the point of assuring our own species survival we are certainly capable of reaching the technical level required to destroy it. And we have very good evidence that for whatever reason few species in a good bit of the universe seem to make it past this point or at least the ones that do are not jumping in to save us.

And of course who can imagine what and alien biology could be capable of we don't even know the limits of our own DNA code and you could have more versatile ones.

Interesting that you should say that...
http://www.edge.org/3rd_culture/church_venter09/church_venter09_index.html

On July 24, 2009, a small group of scientists, entrepreneurs, cultural impresarios and journalists that included architects of the some of the leading transformative companies of our time (Microsoft, Google, Facebook, PayPal), arrived at the Andaz Hotel on Sunset Boulevard in West Hollywood, to be offered a glimpse, guided by George Church and Craig Venter, of a future far stranger than Mr. Huxley had been able to imagine in 1948.

In this future — whose underpinnings, as Drs. Church and Venter demonstrated, are here already — life as we know it is transformed not by the error catastrophe of radiation damage to our genetic processes, but by the far greater upheaval caused by discovering how to read genetic sequences directly into computers, where the code can be replicated exactly, manipulated freely, and translated back into living organisms by writing the other way. "We can program these cells as if they were an extension of the computer," George Church announced, and proceeded to explain just how much progress has already been made. ...

— George Dyson, from The Introduction

A SHORT COURSE ON SYNTHETIC GENOMICS [7.30.09]
George Church & J. Craig Venter

However I sure wouldn't want the same people who brought us "Vista" to try their hand at rewriting our genetic code...

Given the quality of software that this lot usually produces, I would not at all be surprised if they ended up with a bug

My take on the matter is that any truly intelligent species accepts and is content with the fact that their time is limited, and tries to make the best that they can of that time by cherishing and preserving their home planet. Spending their lives away from their precious home planet, zipping around the cold, dark, lifeless universe in a metal can would be the LAST thing such intelligent creatures would want to do. Only half-wit fools like us dream about that as something to be desired, while we go about fouling our own nest.

Memmel,

I am willing to bet that the aliens are out there. A look at the matter from a hunters pov is probably sufficient to explain why we have not yet encountered them.

When I go rabbit hunting even though I know rabbits are very common in my nieghborhood I only find one here and there-perhaps every half mile of walking over very fine rabbit habitat will turn up only one rabbit,even though there are a dozen handy.Maybe the rabbits have learned better than to advertise thier presence.Anybody homing in on a signal might be interested in "lebenschraum",a place to expand.

The other likeliest possibility as I see it is that even in "rabbit country" most of the countryside is not really good rabbit habitat-I can find the occasional rabbit almost anywhere around here but most of them are concentrated in smallish pockets of "edge" habitat that offers optimal food and cover.Such a spot might hold a dozen rabbits in an area not over an acre in size and the next such spot might be a mile away.

My wag is that there are is plenty of intelligent life out there somewhere but that it is probably scattered about with an average distance of many light years between the hot spots.We just haven't been broadcasting long enough to attract attention,assuming any one is listening.

And as far as our own listening is concerned I can't really buy the idea that other species would DELIBERATELY broadcast a powerful enough signal to burn thru the noise at interstellar distances unless they are arrogant in the extreme or else totally confident for some reason that nobody can come calling with ill intentions.

Greenish,

My knowledge of marine mammals is skimpy indeed but dolphins are very efficient predators and work together when feeding ,right?

So the benefits of cooperation in hunting,and perhaps in defending themselves(?) would create the fertile ground for the evolution of high intelligence and even w/o the higher intelligence they were able to eat well enough to support the evolution of thier large energy hungry brains.

But beyond gaining the ability to hunt more effectively,and perhaps to defend themselves more effectively,I don't see the "driver" for really high intelligence.Perhaps it is at some point a geometrically emergent property of a brain that achieves some particular level of complexity-just a lucky (for the dolphins) accident?

The jungle/savannah environment seems to be far more conducive to the evolution of intelligence and the hand/thumb /binocular vision/upright posture combo would provide lots of positive selective feedback,making the appearance of high intelligence "almost a foregone conclusion " in a creature similar to us,if these conditions prevail for long periods of time.

Perhaps if we do succeed in driving ourselves over he edge into extinction the
dolphins could travel the road back that the whales followed when they returned to the water,and build pyramids fifty million years from now.

Or maybe the octopus is most statistically likely new ruler of this world?
The tentacles seem to be pretty good substitutes for hands-good enough to jump start the positive feedback loops leading to high intelligence and they are very intelligent already?

I'm assuming if we depart early we will take the rest of the hominods with us ,as well as such species as the raccoon-but maybe not-I can see them evolving into creatures with fairly serviceable hands in a few million years as they are already using thier current proto hand to do some fairly clever things.You can bet that they will unlock any cage that is secured by only a latch for instance,or even two or three dissimilar latches.Perhaps if the environent changes,or they change,in such a away that they evolve an upright posture thus freeing up the forelimbs they will be in a position to rule the world a lot quicker than the dolphins or octopus.

Hi Mac. I'm a bit more rested today, so a few comments.

First, though, I'll note that the reason I brought up dolphins was not to steer the discussion away from the subject at hand, but just to note that "intelligence" is defined very differently, or left undefined, in most cases. We humans tend to measure it by human accomplishments; when what we're really measuring is "human-ness" and technology.

I set up a lab which ran for a number of years in which dolphins were allowed to interact with various situations and technologies, but not trained to do anything, all behavior at the dolphins' own volition; there were no humans in the loop. It was interesting to me and the only time such a thing has been tried. Their minds (tursiops) are roughly equivalent to ours in my opinion. They're self-aware, understand syntax, intuitively use tools, abstract easily, and learn very fast. They also have some cognitive abilities we don't, such as cross-modal abstract transfer between the sensory modalities of vision and sonar, and communicate between individuals with parallel sound-producing mechanisms, in their primary sensory modality. Which is to say, intelligent by any non-arbitrary standard and capable of some mental tricks we'll never achieve directly.

They have fairly rich intellectual lives, solving problems of the real world and socialization, and have likely been doing so since well before the chimp/human divide. Perhaps LONG before.

I'm not suggesting they're sage aliens in fish suits, although we're unlikely to ever encounter space aliens who are smarter. Rather, their intelligence has developed - (and to a fair extent converged with human mental evolution despite no shared anscestor for 60 million years or more) - with utter physical-reality barriers to expanding their extended phenotype as humans have. Yes, they still could have knocked themselves off by becoming sufficiently delusional, but they haven't, so intelligence isn't toxic to the planet or species per se, just as it doesn't magically trump other kinds of fitness.

So they won't be building rockets to mars. (Neither will we, for much longer.) Their lives are hunter-gatherers and always must be, even if they are sophisticated aboriginals with a shrewd sense of reality in some cases. There are no possible intermediate steps a dolphin could take to build a moon rocket. This has nothing to do with intelligence; our ability to do so is just a peculiarity of the fitness landscape, the intersection between physical reality, fire, thumbs, exosomatic information storage, and a rudimentary abstracting ability.

Of course, dolphins don't have complex societies by human standards, because no complexity shows. And certainly, they don't build ipods and salad shooters. But perhaps they also show something about "sustainability" - not everything is a problem to be solved. Every species is a sort of "answer" to "problems" posed by the universe. Perhaps we monkeys go looking for too many problems, and thus create even more.

My knowledge of marine mammals is skimpy indeed but dolphins are very efficient predators and work together when feeding ,right?
So the benefits of cooperation in hunting,and perhaps in defending themselves(?) would create the fertile ground for the evolution of high intelligence and even w/o the higher intelligence they were able to eat well enough to support the evolution of thier large energy hungry brains.
But beyond gaining the ability to hunt more effectively,and perhaps to defend themselves more effectively,I don't see the "driver" for really high intelligence.Perhaps it is at some point a geometrically emergent property of a brain that achieves some particular level of complexity-just a lucky (for the dolphins) accident?

Clearly, abstracting ability is useful. What we call "intelligence" exists in a broad range as do other physical characteristics. There are emergent "threshold effects" such as self-awareness which are probably accidental but inevitable past a certain level of abstracting. Perhaps culture, legends, and religion too, why not? Moreover, cetaceans are not inherently limited in absolute brain size by an infant's head having to pass through a hole in the pelvis, which has been a hard physical limit to human brain evolution.

The jungle/savannah environment seems to be far more conducive to the evolution of intelligence and the hand/thumb /binocular vision/upright posture combo would provide lots of positive selective feedback,making the appearance of high intelligence "almost a foregone conclusion " in a creature similar to us,if these conditions prevail for long periods of time.

Ah, careful of retroactive narratives which imply a preferred direction to evolution, they'll bite you.

Perhaps if we do succeed in driving ourselves over he edge into extinction the
dolphins could travel the road back that the whales followed when they returned to the water,and build pyramids fifty million years from now.

I doubt it, but it wouldn't be physically impossible. I've tried to keep the options open to them to continue evolving; their status at the current mass extinction is innocent bystanders, and they may not survive us. Certainly the other apes are unlikely to. But what good is a pyramid? Rhetorical question.

Or maybe the octopus is most statistically likely new ruler of this world?
The tentacles seem to be pretty good substitutes for hands-good enough to jump start the positive feedback loops leading to high intelligence and they are very intelligent already?
I'm assuming if we depart early we will take the rest of the hominods with us ,as well as such species as the raccoon-but maybe not-I can see them evolving into creatures with fairly serviceable hands in a few million years as they are already using thier current proto hand to do some fairly clever things.You can bet that they will unlock any cage that is secured by only a latch for instance,or even two or three dissimilar latches.Perhaps if the environent changes,or they change,in such a away that they evolve an upright posture thus freeing up the forelimbs they will be in a position to rule the world a lot quicker than the dolphins or octopus.

My guess is that intelligence has little to do with ruling the world, that's a human aspiration and we aren't doing so well with it. What does an orca think about with its hugely larger brain? What culture might there be in Bowhead whales, who can live more than 250 years - and maybe a lot longer - and which have the same senses we do? We techno's have been harsh on aboriginals.

The way the cephalopods "think" is fascinating, and they too can communicate in their primary sensory modality, which may involve a level of parallelism which would quite alien to the way we think.

The single-celled organisms rule the world; we're a sideshow. Hell, vertebrates are a sideshow. But one it would be nice to keep that show going, and I hope it does...

cheers

Uprated.
(Even though I didn't understand it all).

Maybe the dolphins did figure it out and that's why they went back to the sea.

cfm

Man had always assumed that he was more intelligent than dolphins because he had achieved so much... the wheel, New York, wars, and so on, whilst all the dolphins had ever done was muck about in the water having a good time. But conversely the dolphins believed themselves to be more intelligent than man for precisely the same reasons.

Douglas Adams, The Hitchhiker's Guide to the Galaxy

Thanks!

A reply from you is as good as a one on one with a good professor,and a lot cheaper than tuition!

Your point about retroactive narratives is well taken and I understand that nothing is a foregone conclusion in evolution,but given a particular set of conditions both genetic and environmental it still looks reasonable(to me) to assume that some general outcomes can be if not exactly PREDICTED then anticipated as being somewhat more likely than other possible outcomes.Of course the precise mutations or combination of mutations that led to our present day brain might never turn up ,any more than some specified hand might turn up in playing cards.Maybe proto humans might evolve intelligence as we percieve it in one case out of dozen or a hundred or a thousand or a million if mother nature runs the experiment that many times.

But it seems that many things favorable to the evolution of intelligence would already be in place and therefore not many "lucky breaks" would be needed.For instance binocular vision implies the ability to judge distances accuratelyrequiring brain circuitry for that purpose,and walking upright must have encouraged much greater use of the hands,thereby applying selective pressure towards the development of greater fine motor control,thereby again applying selective pressue to more brain power,etc.

As far as the chain of events happening that would actually result in dolphins becoming land dwellers land-the chances of this happening must certainly approach zero and I suppose the process would occupy as many millions of years as the return of the whales ancestors to the water.

As far as ruling the world or building pyramids are concerned ,I'm with you -I mentioned these things only as a sort of linguistic short cuts to save words.

I am very much interested in reading more on this general subject and have just started Pinkers "The Stuff of Thought".

Any relevant book titles from any quarter will be appreciated.

I have spent quite a few hours daydreaming about percieving the world as my hound percieves it-obviously his brain trusts his nose to a far greater extent than his eyes.There must be hundreds or thousands of olfactory cues floating around that allow him to know things,hound fashion, we cannot even imagine.Maybe he can judge the "state of mind " of another dog by smell as well as we can another human by sight.

As far as being off topic is concerned,as I see it electrons are cheap and it's easy to skip over a comment.The most productive dialogues are often the ones that veer off into side issues that turn out later to be highly relevant or that open up whole new areas of inquiry.

Thanks for your kind words, Mac. A reply from you is a good way to start a dozen new and interesting conversations, but time presses and I probably shouldn't veer too much off the keypost topic. But in brief reply:

Your point about retroactive narratives is well taken and I understand that nothing is a foregone conclusion in evolution,but given a particular set of conditions both genetic and environmental it still looks reasonable(to me) to assume that some general outcomes can be if not exactly PREDICTED then anticipated as being somewhat more likely than other possible outcomes.

Certainly there are some situations in which the path between two spots on the fitness landscape are less blocked than others, but I think it may be unpredictable even in principle before it happens. A darned interesting topic, though.

Our last common anscestor with a dolphin would have been a shrew-looking small critter no smarter than a possum and likely dumber. A lot of species subsequently popped from that anscestor's portent-filled loins, and most of 'em weren't too bright. History has been described as "frozen accidents", and I think that's a good way to think of it.

And I'm mentioning these in reply to your post even though I know you're on top of these subjects and are a lot more well-read than I, since they're worth discussing in general.

I have spent quite a few hours daydreaming about percieving the world as my hound percieves it-obviously his brain trusts his nose to a far greater extent than his eyes.There must be hundreds or thousands of olfactory cues floating around that allow him to know things,hound fashion, we cannot even imagine.Maybe he can judge the "state of mind " of another dog by smell as well as we can another human by sight.

Dogs are olfactory geniuses. Who knows what tricks a canine mind can do with abstracted smell input? It may smell in "color", in the way Dawkins has speculated a star-nosed mole maps touch input to existing brain qualia. Point is, we don't know. A dog with the intelligence of a human imbecile would be the greatest diagnostician ever, probably. We blind ourselves to the wonder of life all around us by seldom taking off our monkey goggles. Thanks for the reminder...

I've personally become convinced that humanity is just a very short-term evolutionary stepping-stone to the super-intelligent self-aware electronic mind. Will it be digital, or quantum, or some algorithm as yet not thought of? Wonder if they'll keep any humans around as pets? The core question is, should we allow "artificail intelligence" / robots / computers to become self-reproducing? As soon as they do, humanity is of no further use to evolution.

How are they going to develop emotions to drive their actions?
That is a big leap from the state of the art right now.

Look where they've come in 60 years. Give them another thousand or ten years (just the blink of an eye in evolutionary terms) and you won't recognize them. The "state of the art" today is as a zygote embedding in the womb wall. Hints may be made out on the horizons if aware and observant, but very few. The key is reproduction. At now, humanity with sophisticated metalurgy, etching, electricity generation, maintenance etc. is still required. For how long?

If we ever succeed in building a generation of computers good enough to run an economy and direct it to build more computers then this evolution of self directed as opposed to human directed artificial intelligence may become something worthy of serious thought-right now it doesn't seem even remotely likely that we will see anything along these lines for the forseeable future-certainly not within the next few decades.

But if it ever gets off the ground!!!! Computer evolution will be Lamarckian.Improvements can be incorporated more or less immediately across the board instead of arising and spreading snail paced by natural selection,and computers need no extensive and expensive education-load a program and data and she's off like a rocket even if still physically warm from the last test run for quality control purposes.

Just imagine how cheap engineering talent would be if highly intelligent men could be produced in factories by the millions every month and an doctoral level education downloaded into his skull in a few minutes!!

And while I cannot see any reason for such computers to "want to " reproduce,other than such "desire" being programmed in to the first generation it doesn't seem likely that they would tolerate much wasted effort or waste of resources if such would interfere with thier rapid reproduction.

We might or might not survive as lab specimens if they see a use for us a research materials.

Agreed, OFM, and no, we won't survive except perhaps as zoo specimens. The hints are already there. In my short life, I've gone from being impressed by a crude analog computer used for jet turbine sims at U of Guelph in the 1960's, through repairing radios and motor controls based on discrete transistors, to observing the installation of a huge analog computing machine which could control every parameter of a huge paper machine, to experimenting with 10 NAND gates or a 7-seg LED numeric display controller on a single IC in the 1970's (impressive), through assembling a 16K memory board the size of a sheet of paper from 4 kbit DRAM's and interfacing with a Zilog Z80 cpu in the 1970's (incredible), to being floored by IBM's decision to publish a documented listing of the assembler BIOS of the original PC (inexplicably suicidal), to applying Ungermann-Bass ethernet cards with more powerful processors (80186 vs. 8088) on the network card than the computer used, to working on the architecture team of a group who in 1993 implemented artifically "intelligent" neural nets to the task of determining who should/would get audited for Federal Sales Tax evasion in Canada, through observing that my friend's car radio in 2000 contained more RAM memory than was available to the entire computing resources of a large company just 10 years before ('course it was a Ferrari), to today where my children gift me on my birthday with a Sony Book reader which has features which are not less than we would have considered magical 20 years ago.

Oh no. We haven't seen the endpoint of this revolution yet. I'm not in any way religious, but if I were, I would be assigning this pace of development of machine intelligence "capacity" (unexploited) to an off-world god-figure. The end is not even close to nigh.

Hardware is cool and your right things are moving however I'm buried enough on the software side to recognize complexity is killing software.

And the hardware side it looks like heat will be the nemisis for quite a while.

In fact the software I've been working on for five years is explicitly designed to allow you to pull out a module and replace it with a simple implementation.

Its design cycle goes like this.

1.) Design interface
2.) Plugin any reasonable implementation and if you must develop one
3.) Use Test software
4.) Refine interface
5.) Use/Test
6.) Rewrite implementation keeping essential components maybe moving seldom used features to extension module.

I suspect readers of the oildrum would not be surprised that I actually design my own software to object Tainters laws if you will of continuous simplification.

So on the software side at least not only have things moved fast but they already have hit the complexity wall. New layers no longer work in the sense that the efficiency declines and the ability to innovate declines dramatically as your locked into the design of your lower level software.

Outside of computers getting powerful enough to run multi-media software and polish if you will many areas of software have ceased to evolve.

Of course many others have also seen the problem and are now developing modular upgradable software but that just highlights the wall we have hit.

My personal knowledge of computers is limited to cruising the net,simple spreadsheets,and Word for the most part,but i have read a couple of "basics" books.. I do try to keep up with thier overall capabilities.

Your comment about being locked into the design of lower level software indicates that the programmer is in a position analogous to the Blind Watchmaker of evolution so far as I can see-he can only work by adding on to(or perhaps sometimes pruning back) what exists regardless of it's inherent shortcomings.This is understandable enough and seems to apply across the board to nearly every industry-you necessarily work incrementally with what you have on hand and only rarely does someone find a really new way of doing an old job.

If I understand what you are saying about modules you are creating entirely new programming from "scratch" that avoids this overarching problem.The idea is simple enough.

Do you expect these new "extension modules" to run on existing machines with existing operating systems?

And do you think that artificial intelligence can concieveably advance to the point that machines can run an economy capable of supporting what I guess must be defined as either artificial or synthetic life?A silicon ecology?

I can see a Luddite up rising in the future that will include some strange bedfellows indeed if this turns out to be the case.

I agree that the "10 year cycle" I grew used to in the last century has slowed down recently, especially on the software side. I suspect that the problems of software development, eg. the fact that that's really the only part of the system which is still a manual craft carried out almost entirely by humans with relatively little aid from machines is the bottleneck which will provide the initial impetus to develop genuine artificial intelligence. I think that computer hardware technology is already capable of supporting a basic intelligence capable of self-replication, if enough existing units of processors and storage are thrown at it, but the software, including the core paradigm, doesn't exist yet. When analysed, software has really remained static for over thirty years, eg. the best language for implementing something such as an artificial intelligence which we have at hand is still C/C++ and assembler. (Most popular modern languages are just user-friendly implementations of a subset of C++ commands, and of course all are just that over assembler / machine code). Some may argue for one of the more obscure experiments such as FORTH or ALGOL etc., but so far they haven't really provided sufficient advantage to make an impact. Of course the editing environments we use to interact with the software have improved dramatically but those sorts of improvements don't constitute progress toward smarter machines. I think the problem really is deeper, perhaps in the core set of instructions which are manufactured into the processors today. Many variants were tried at the beginning, but they've pretty much all settled down to an implementation of a bunch of I/O and math commands. Perhaps that's a dead end?

I don't think the basic stuff is a dead. In general I'm happy enough to use libc. On the IO side I've made the sensible generalization to URI's and data providers instead of files and the io is in general async but this is very common. So say up to posix libc I'm happy enough with the most part with whats there its when you go up even a little bit higher that things start to fall apart.

Simple concepts like advertising a service remain very convoluted Apple has bonjour and we have bluetooth and to some extent web services but in my opinion services are hobbled by problems in the design of a lot of software that exists between the basic libraries and these high level services everyone wants to create.

I could really do some serious long posts on the subject :)

But I guess the simplest answer is I'd say probably 90% of the software in existence today is probably a dead end. As and example relational databases have hit their limits the next big thing in that area probably will be based on some sort of database plus something else search/ontologies looks like the best pick for something else but the concept of a traditional database is rapidly waning. The combination of relational databases and software front ends for them takes out at least 50% of the worlds software if not more :)

I'd be surprised to see much of it make it 20 years in a form even close to whats around today.
The web as we know it is a dead end and will be junked. Now this does not mean the seeds of the future systems are embedded if you will in our current systems but the actual amount of code that will eventually grow to be around in 20 years is a small percentage of whats around today. Like I said 10% or less.

I think things like linux will make it however I can see linux, OSX, BSD eventually merging. I think Windows will be repeatedly rewritten from scratch and eventually the shear problem of reinventing the wheel will stall windows development as the complexity of the system and maturity problems with continuous rewrites causes windows to be unable to move forward. And of course given opensource has in built support for evolution and its dirt cheap commoditization and increasing functionality will drive make it ever more expensive for windows to compete as MS's profit margins fall. But these same forces that are refining the core software stacks are also providing the stabilization needed to support innovation in the higher levels.
As the core becomes sensibly fixed the higher levels can finally really evolve.

The key and big difference from today is I see the core software evolving where needed and not where its reached refinement at least in the opensource software stacks. This is different from what I suspect MS will do they will build the bigger shinier next big thing tm till it kills them.

For my own software stack this correlates in to stabilization of code interfaces as the modules mature they can increasingly be incorporated without much if any work on the part of the programmer. But still replaced or evolved if needed. You need this sort of stabilization without lock in and I think it will happen with our without my own contributions I see many projects going down the same road I am using different approaches.
Eventually these differences will be eliminated and I think modularity will itself become a commodity concept.

With luck based at least in part on my approach of course :)

Pretty much full agreement. Only thing is you're still simply addressing software "as we know it", which is never going to develop intelligence of become self-aware. For that, a complete re-do from the processor instruction set on up is probably needed.

Do you expect these new "extension modules" to run on existing machines with existing operating systems?

And do you think that artificial intelligence can concieveably advance to the point that machines can run an economy capable of supporting what I guess must be defined as either artificial or synthetic life?A silicon ecology?

Yes they will run on existing systems but "not really" By that I mean the complexity of the interfaces are steadily reduced until they are effectively running on libc and simply drivers they only use fairly reasonably basic aspects of the system. However :)
You can also surgically replace the implementations of many of these subsystems to leverage more advanced subsystems such as modern windowing system.

And of course there is a huge catch and its actually the real problem my software can't really go sideways if you will. I can use the windows but I can't use the widgets such as scrollbars buttons etc. Not that I can't technically create wrappers similar to many in existence but thats not the way my stuff works.

The reason I can't go sideways is the reason most software is messed up it makes to many assumptions about the design of the lower level software. In my software only the interfaces are exported and pieces load the software stack they need to function. So my stuff solves the sideways problem if you will but in doing so it highlights and exposes the basic flaws in lots of existing software.

Probably the top flawed types of software are windowing/widget systems and compilers. In general implementations of both are intrinsically flawed because assumptions about the layers in the system get deeply intertwined in the software itself tell its a hopeless mess.

However even with this I can do crazy things with my stuff like run a full blown rootless X11 server with div's on a web page providing the windows. The browser run's on bare metal and the traditional windowing system becomes a simple browser plugin.

Needless to say providers in my system can differ dramatically from traditional software stacks.

On the same hand maybe this hints at the sideways problem and why I've had to go a completely different route to solve it.

Of course like peak oil I've found in the software world a lot of programmers don't even grok their is a problem. Often the ones that do get a bit uncomfortable and point out that this problem is why they have jobs :)

It is difficult to get a man to understand something when his job depends on not understanding it.
Upton Sinclair

The fact that people who do understand what I'm doing make this observation tells me I'm on the right track.

Now as far as AI goes well obviously I think the real problem is this sideways problem as you make it easier to integrate sideways eventually more of these problems become amendable to ai algo's with little or no human intervention. AI is no different from other types of software if anything its the most flawed and again almost all of it fails miserably in the sideways test. A simple example almost all AI software uses some sort of ontology or typed database of some sort as part of the software design I've not seen any designed to work in such a way that this knowledge store is replaceable and adaptable and capable of being wrapped around unknown implementations.You can't do any aggregation with AI software in general. The fact that its generally uncommon is all that keeps it out of the ranks of my top worst picks.

As a simple example of a sideways problem consider this window I'm typing in. I love vi but I can't easily just drop in my via editor in place of the text area provided by the browser. I use firefox for example It has a little text search bar available at the bottom the text window does not have history and the search options are a bit limited maybe I want a google it button. I can't drop in a replacement easily.
With my software any component that implemented the interface could be dropped in in this case as long as its own dependency change could be satisfied in the context. In my software these higher level components are themselves written against abstract interfaces so in general its trivial to do this sort of sideways integration.

Other examples just looking at the web page I'm of course a oildrum addict so I like to look at it on my iphone no way is the interface mobile friendly it would be up to the designers to develop a phone interface but for a rich site like this I'd like to customize my view I don't need to see the list of people that work on the oildrum I know their names after this many years :)

These are visual examples but they are the easiest to explain non-visual software has the same problems just you have to be in the domain. Compilers for example generally don't make it easy to play with the AST ( Abstract Syntax Tree).

As you get into AI software its even worse. Numerical software is also generally really bad. Thus for AI my basic opinion is that until you can solve this sideways problem you can't even think about real AI.

Now once its solved then you can play a trick you can start with software that uses hand coded modules and add machine integration then as the AI system evolves it reaches the point it can analyze and replace its own hand coded modules with machine coded ones. Once its self programmable then your basically done. You can see my interest in mucking with compilers because the AI needs to be self reflective against its own code base.
Effectively it learns how its programmed and modules are steadily added until it can self modify.

The neat thing is you can effectively code up the intelligent modules to get around road blocks you don't have to bootstrap a traditional AI algorithm in effect your start with a idiot savant and over time it gains real expertise. Thus the trick is to develop a way to cheat until you eventually don't have to. Sort of like getting really good at counting cards and playing a lot of poker and eventually becoming naturally good at poker and no longer needing to count cards.

Of course I think this is how intelligence actual arose at first it was hard coded chemical responses that turned into instincts as better receptors arose then to choosing among several instinctive moves and thence to creating prototype instincts or models which are just thoughts. Each stage really added basically sideways extensions while the core instinct remained largely intact. We still have very basic fight or flight responses for example. Mother Nature cheated so why don't we figure out how to cheat too ?

Utterly OT, but I just have to ask: Are you familiar with Common Lisp, in particular CLs macro system?

CL macros are effectively extensions of the compiler. A CL implementation is commonly made up of a small number of basic operators written in assembler or C... The rest is mostly macros.

(One of the features that make this practical is the Lisp syntax -- often passionately hated by non-Lispers -- which has the expressive power of XML, and then some... Converting Lisp-syntax to XML is trivial, but Lisp is much easier for humans to read).

The programmer can write new macros and also redefine existing ones, actually changing the way the compiler works, and thereby the language itself.

Also, CL programs run in an interactive Lisp process... which can be seen as a sort of quasi-OS (or perhaps "embedded OS"?). There is no need to restart that process to update a definition, even a macro definition. Which means you have a run-time-re-programmable (potential) OS...

There seems to me to be quite a bit of Lisp philosophy to your programming.

You're not sneak-reinventing Lisp machines, by any chance? ;-)

I'm buried enough on the software side to recognize complexity is killing software.

And the hardware side it looks like heat will be the nemisis for quite a while.

On the one side complexity, on the other thermodynamics. Give me Scylla and Charybdis.

But it's more than that. How do you spell "low hanging fruit", memmel? Limits to growth? Diminishing returns? What about morality?

In one of the business hats I wear, I have charts and diagrams of how to move from rev3.* to rev4.*. They are ten years old. Or older. Never going to happen. Over my dead body. if I can get another year out of my cash cows, fine.

cfm

Yep my opinion is simply that computers hit the complexity wall harder and faster.

I think anyone willing to look will discover in both hardware and software what progress is being made is done by grappling with Tainters concepts. Aerospace I suspect is in the same boat.

You have to be able to see the difference between what we are trying to do and what we accomplish esp since in many ways the edifice of complexity that supports modern engineering makes many things look easy.
And of course the dreams of engineers are boundless.

Whats really interesting is the start from scratch concept continues to remain valid in engineering and sometimes its actually used. Sometimes with great results sometimes not. Just because you think you can do better is no assurance you will.

What I find really interesting is that the real problems exist regardless of the viewpoints of the engineers working on the complex mess that is modern engineering.

Despite the widely varying opinions and views held by most groups of engineers it does not change the problem.

The most common solution is to figure out how to squeak buy one more time adding a bit more to the mess.

Every now and then you get a chance to do something cool and leeway to actually do something its these rare occasions and also the fact that you don't often figure out how screwed things are until its to late to change.

For me at least I've simply got way to much experience to walk away and take a lot lower pay check at the moment. And I doubt the grass is any greener anywhere else since the complexity problem is almost viral.
For programming at least open source has turned into a way for me to enjoy programming and do it the way I want to. Sometimes I'm able to inject these ideas into commercial projects.

I seriously doubt I would have stayed with it otherwise. I honestly think that given the chance engineers really understand Tainters concepts and if we where allowed we would simply the complexity.

My experience has been its not that we can't do it but we don't control the money and no one is going to pay us to do it right.

To some extent I think this is why the believers in the technical silver bullet are the hardest to sway. They see how it can be done right and don't stop to consider how we managed to make a mess of things even as we exploded technically. Its not that it technically can't be done its that the social system prevents it.
The problem of course is to do it right will cost a lot and its difficult to determine failure.
Society is unwilling and perhaps even unable to support such a effort.

In fact I'd suggest its incapable of it since adding complexity creates power for the "managers". You lift yourself up the rungs of society managing people not building cool stuff. Our own human social pyramid scheme lies at the root of the problem of complexity and its not easily solved.

The only solution is dissipation and re-localization by limiting the lines of management and probably recreating something more akin to guilds which of course have their own problems can you at least get out of this rut. Likely into another but the way we do things now is simply not going to work.

I won't go into science but regardless of attempts to curtail the formation of human pyramids the combination of people and money seems to relentlessly convert everything into a situation managed by some extravagantly paid remote executive. Even in science look at what the presidents of universities make these days.

Somehow someway we have to eliminate this so we can build the simplicity that we know can be done.
Believing in the techno fix is not the answer however since of course your going to need a highly paid management team and daily meetings to save the world.

Hi greenish,

Really interesting posts. I just have a couple questions & comments, off-topic of the general thread. I studied cognitive science at SFU, so I'm very curious about your studies with dolphins.

You say,

They're self-aware, understand syntax, intuitively use tools, abstract easily, and learn very fast.

Would you say, then, that they communicate using symbol systems in the wild (i.e. do they have words that they didn't learn from us)? And do they have infinitely recursive syntax? (Well, technically, humans don't even have that, despite what Chomsky says; we can only manage maybe 5 recursions deep and still keep track of what's being said... Anyway, can they manage recursive syntax?)

Also, from studying cognition, I've concluded that the ability to follow pointing (which necessitates having Theory of Mind and thus a capacity to interpret what another mind might want/be thinking about) seems to be a major prerequisite of developing symbol systems. (Cats, as a counter example, just sniff your finger when you point; and, although they can walk you over to something they want - such as a toy that's out of reach for them - they can't usually understand when you do the same, unless of course you want them to eat a piece of food! To be fair, cats do have theory of other mind, but it's pretty much limited to acquiring/demanding food, establishing dominance & building relationships, and playing; I guess being solitary hunters, they didn't need to evolve the capacity for coordinating behaviours with others.)

There is a researcher at Sony Labs by the name of Luc Steels who has done some really amazing work with robots into understanding embodied cognition and symbol grounding. (He has many papers out, but the one titled, "Intelligence with Representation" is probably the best summary that I've found of his research.) Among other things, he describes robots who play what he calls the "Naming Game" (the game itself, as well as the ability to "point", are hard-coded into the robots but their ontologies and [structurally coupled] symbol systems are developed individually in each robot via playing the [interactive] game).

The robots play multiple rounds of the Naming Game, in which one robot utters a symbol (call that one the "Speaker") while the other (call them the "Pointer") has to point to the referent of that symbol on a white board containing an arbitrary collection of stuff. If the Pointer points to what the Speaker identifies as the referent of the Speaker's produced symbol, the Speaker gives the thumbs up (also hard-coded), then the Pointer reinforces the connection between that categorization of their sensory perceptions and the associated symbol, and they play another round with their roles reversed. If the Pointer gets it wrong, the Speaker points to the referent of the symbol it just used and the Pointer adjusts their ontology by adding/removing sensory discriminations, strengthening/weakening the associations between a sensory categorization and a symbol, etc. according to where it perceives the Speaker to be pointing. Over time, their ontologies become more refined, and they converge on common symbols for the different categories they invent (arbitrarily, in the case of the robots playing the naming game, although Luc Steels would agree with me that in animals the categories largely depend on the demands of survival and reproduction in particular environmental and socioeconomic contexts).

Now, obviously, I'm not trying to say that humans play the naming game exactly, since our ontological and symbolic development seems to be less explicit than this in most cases (we pick a lot up from just watching others, and each subsequent generation doesn't reinvent the wheel but acquires and modifies the languages and categories of their elders); also, with many symbols acquisition is often quite abstract (i.e. we use more concretely grounded symbols, as well as metaphors and analogies, to describe the meaning of more abstract symbols, rather than point directly to a referent that may not be readily available as an object or event in the world). But, in order to carry meaning, some/most symbols must be grounded out in referents that can be pointed to in the world, with the rest being built off those, or simply existing to fulfill syntactic roles (e.g. connectors such as "and"). (I mean "point" here in some manner or other, not necessarily with your finger; walking over to a thing could constitute a rudimentary form of pointing, but one that's less useful, not to mention potentially dangerous if it's something that might eat you).

Thus, my third question is: can dolphins point, and if so, how? Do they use their nose, and do they have to swim over to an object or can they point the way we use a finger for distant objects? And do they do this spontaneously with each other, or just at the request of humans?

(As a further tangent, just for interest's sake, one key concept I took from Luc Steels work with the naming game is that it is via repeated interactions with each other that our respective understandings of the world become more similar, not just that our language grows more alike but that our ontologies do so as well. There is no guarantee that a language with no prior connection to yours will have a concept that equates, even roughly, with one of your concepts; furthermore, boundaries between categories could be in different places, which is why some words only carry "shades of meaning" in a translation, rather than the full spectrum of nuance present in the original.)

Hi greenish,
Really interesting posts. I just have a couple questions & comments, off-topic of the general thread. I studied cognitive science at SFU, so I'm very curious about your studies with dolphins.

Off-topic indeed for this thread I'm afraid, feel free to click my user name and contact me via email. I closed my dolphin lab in 2003 after about 13 years of research, so am not "in the biz" anymore. It was a side-project. If I conclude that I can't do anything more worthwhile for the world at some point, I'll write up more of it someday, but that's a big if. I'll do very brief replies below:

Would you say, then, that they communicate using symbol systems in the wild (i.e. do they have words that they didn't learn from us)?

I'd say they're entirely capable of it biologically, and that it probably varies between species, but that their natural communications are mostly not linear-syntactic.

When prompted to by the rules of a game, they can create arbitrary "words" and then use them as consistent referents. When the game then digitizes them, those "words" can be used between the computer and the dolphin, or by humans playing it back. Personally I doubt they do this much on their own, just as we don't hop on one foot much of the time.

And do they have infinitely recursive syntax?

I don't know. In general, I don't think they utilize syntax in communications the way we do, they're simply capable of it. Certainly, they can communicate and coordinate rather complicated behaviors; but they don't usually do so in the way humans would.

can dolphins point, and if so, how? Do they use their nose, and do they have to swim over to an object or can they point the way we use a finger for distant objects? And do they do this spontaneously with each other, or just at the request of humans?

active sonar, by its nature, incorporates and facilitates pointing. They are certainly able to coordinate pretty sophisticated hunting strategies in which the designation of attention foci would be necessary across the group.

Moreover, since they're wired to receive sound and process it instantly into 3D mental imagery the same way our optic lobes do, I've thought it unlikely that this ability would be arbitrarily left unused in efficient communication. In other words, between dolphins some referents are probably highly onomatopoetic and received as shorthand imagery. This might also facilitate basic communication between cetacean species which do seem to communicate within mixed-species groups, though this is my own wild speculation. (one of my neighbors, Paul Nachtigall, had a student create a DSP card that would generate phantom echolocation returns, but I'm not sure how far he went with that, it's a nice angle. )

One toy I made for the dolphins was a system which combined a touchscreen computer with voice-recognition capabilities, so that dolphins wishing to interact with computer programs for entertainment could do pointing by interrupting a network of invisible light beams via rostrum pointing, or accomplish the same thing by using digitized "words". Different dolphins tended to have their own preferences, or a single dolphin might do both in a single session.

ok, that's all the dolphin stuff. Just to halfway tie it back to the theme here, I did my best to tie the work to practicality, "satisficing" to get "good enough" information for conservation purposes, and then shut it down. My notion is to prevent their eradication, then perhaps we can get to know them over the long term. If we wipe them out, we don't deserve to know the answers.

I like dolphins, but we do not have to go as far afield as that. There have been in the past, and remain even in the present, peoples who were or are sustainable--that did live or do live withing the resources of their environment in a state of comparative cultural stability over periods lasting thousands to tens of thousands of years.

How did (do) they do it? Well, certainly not by thinking in terms of (short range) Darwinian advantage.

I am as pessimistic as Tainter, but for different and even opposite reasons: In studying civilizations that failed, Tainter is learning a lot about how to fail. This has its uses: By studying Rome we can learn a lot about what not to do.

However, suppose we want success--then we should be studying people who have succeeded, to see if there are things we should do to succeed, not just things we should avoid doing to avoid failure.

And this is the source of my pessimism: We natter on and on about Darwin and refuse to study peoples who have succeeded. Why is that? Do we wish to fail? I suspect, yes, we would rather fail and destroy ourselves utterly than learn and make the changes necessary for continuing life.

What is problem solving at a societal level? To me it means using the sunk cost of built infrastucture and cultural values to dictate a societies future path as opposed to a real matching of supply and demand balance sheets.

Right on Nate. And, inevitably, the key is how to overcome the vested political interests associated with sunk-cost infrastructure and outdated cultural values.

I think focusing only on the resource-constraint issue, comparing us to yeast, and blindly saying 'we're doomed' simply isn't useful nor terribly intellectually honest. It defeats the purpose of discussion. SOME societies ARE ABLE to overcome these problems at least for a little while. If we all want to lessen the consequences of the crisis that we all believe it coming it behooves us to understand why certain outcomes occur in some places and not others.

And, inevitably, the key is how to overcome the vested political interests associated with sunk-cost infrastructure and outdated cultural values.

Exactly. Just think how much energy and how many resources we could free up if we returned to walkable cities with public transportation and mass adoption of bicycles, rather than attaching our identity to the car we drive and stubbornly seeking to maintain Kunstler's "Happy Motoring Utopia". Just as one example.

As another, consider research into happiness by psychologists and behavioural economists finding that raising standards of living much above poverty levels doesn't significantly improve subjective well-being. (I.e. consuming more doesn't make you happier.) Mass adoption of this principle, combined with transference of status-seeking competitions into non-consumptive domains could do a lot to free up resources for addressing the problems facing our and future societies.

raising standards of living much above poverty levels doesn't significantly improve subjective well-being.

Does that generalizeration still hold after you force the said "raised slightly above poverty" family to live in close juxtaposition to eg. the Vandebilts?

Not entirely. Just as, for example, people will prefer to live in a neighbourhood where their house is worth $80k when their neighbours' houses are worth $70k rather than live in a house worth $100k when the neighbours' houses are worth $110k. Ditto for wages.

But that has to do with social comparison/status seeking. We like to be on the top of the pile. You can, however, "trick" this line of thinking by always comparing yourself downwards rather than upwards. Or, find a niche where you can be respected and don't compare yourself outside it.

Basically, the gains to happiness that come from raising your standard of living are a result of two things above poverty-level: 1) a temporary difference between your expected consumption levels and your current consumption levels, which readjusts fairly rapidly (~1-2 months at the most), & 2) gains in relative social positioning, which is a zero sum game and therefore an inefficient use of society's resources.

What exactly is “problem-solving” at a societal level?

Here's an example: terrorist threat -> Department of Homeland Security and the "war on terror ". Previous examples of this kind include the "Communist threat" leading to the maintenance of the largest standing armed forces, as a proportion of GDP, in the history of the world, and a costly "police action" in Southeast Asia.

Here's another: Global warming problem -> mandating the use of, and subsidising the production of, ethanol as a vehicle fuel.

A key characteristic that the paper didn't mention is that the preferred "solutions" generally change (social) things as little as possible. Minimal change is preferred over actually solving the problem.

Tainter’s claim that energy is not the driving force of cultural evolution is almost laughable (if it weren’t coming from him).

Why is it laughable? To me it seems that adapting to threats and malign changes in a society's environment would drive its history, the same as for the evolution of species. They're called "evolutionary pressures" for a reason. Energy shortage may be one such pressure - the deforestation of Britain leading to the adoption of coal as a fuel springs to mind.

The final end of human “problem solving” understood in these terms is infinite growth.

I don't think it works like that - there is no "final end". A problem arises; it is "solved". Rinse and repeat. The idea that any society has a "final end" (other than, perhaps, its own continued existence) is a progressivist myth.

I will grant that the most successful strategy for survival (to date) has been to continually strive for growth, so many internal features have appeared that reinforce the striving - features such as the myth of progress, private property, capitalism, and so on.

Of course, for problems that a certain society considered really big, they came up with a "Final Solution."

How many non-academic observers of Tainter/Jevons themes, read between lines and take collapse as given? If many, I wonder if there are any entities out there attempting to shape it? I would kind of be surprised, but not greatly.

Cornelius,

If these entities exist they are maintaining a very low profile.My personal guess is that at least a few such entities do indeed exist but that thier work is being done at the strategic rather than tactical level and very very quietly-any activity coming to the attention of the public would be represented as being directed toward other goals.

Collapse could be steered in many ways-a Bill Gates can either fund or not fund an initiative as he sees fit.If I had big money and were convinced that collapse is inevitable,I would not fund any initiative that would prolong the agony but I would find a "Greenish" to help me spend it in a way that would ultimately (hopefully) preserve as much as possible of the biosphere and thereby preserve a better world for any surviving personal descendants.

I just finished Nicolas Wade's "Before the Dawn" which E.O. Wilson describes as "the best book I have ever read on humanity's deep history".Lionel Tiger of Rutgers describes it as"bound to be the gold standard in the field for a very long time".

Considering my personal standing as "an amatuer scholar who knows just enough to be dangerous" I suppose I need say no more but I will add that this is an ENJOYABLE BOOK whose author has kept the discussion at a level accessible to the well educated layman.Books of this quality are rare and not to be missed by those of us interested in the big picture.

There may well not be a better single book if you have time to read only one such book on the subject of human evolutioon and early culture.

Supporting a seed vault would indeed be a good way to steer collapse and a winner either way collapse or not.

Absolutely. Having a number of varieties could help foresway a sudden sweeping crop diseases (e.g., wheat rust)

I'm on my second reading of Before the Dawn. It is superb.

I read it when it first came out. Good writing, but it seemed to me to be slanted toward emphasizing the role of conflict and competition rather than the many examples of cooperation and relatively stable, peaceful communities. (Just to be clear, if a book only emphasized these more peaceful communities to the exclusion of the very many, clear examples of conflict, I would be critical of that slant, too.)

Greenish, I would be very interested in any published (or unpublished) work you can point me to about the syntactic abilities of dolphins you mentioned.

You'll probably do pretty well with a google search on "dolphin syntax"; I haven't kept up with recent research. Diminishing returns on time investment in complexity, doncha know.

Most of mine was done for my own edification in helping me prevent dolphin deaths, publishing the bits which might help conservation efforts. I'm a bit of a capt. nemo character that way. I have the memories and videotapes, but what dolphins need just at the moment is not to be destroyed. Academia and the pursuit of knowledge are nice, but can distract one. As can blogging, so I'll only be here a bit longer...

http://en.wikisource.org/wiki/The_Lotos-Eaters

best

You'll be missed.

For 6,000 years in Egypt it was possible for farmers to pass away, and their offspring to farm the same land they did.

That's pretty stable, in human terms.

It's sad that humans seem so unable to understand how economies can become part of nature, and that there are a great many examples of natural system economies that demonstrate creative sustainability in the long term. When Joe, whose critical thinking about the literature of collapse theories I greatly admire, says things like "the fact that societies increase in complexity to solve problems" I absolutely cringe. The caveat he omits is "...societies that don't stabilize as natural systems do". A great many kinds of natural systems change the form of complexity they keep adding as they end their growth. They switch developmental directions to mature toward a peak of vitality, switching from adding outward complexity to instead mature and perfect their systems by adding inward complexity instead. If things only continually add outward complexity they're continually adding overhead, yes, and working themselves into oblivion. It's a very "popular" way to end civilizations.

The Romans and lots of other societies that became sophisticated problem solvers failed to discover how to solve that problem. That's what we should be talking about. Why are our solutions the main source of our problems!!?? Humanity could also be the only complex problem solving organism in nature that seems to get repeatedly caught in that trap too. What Joe detailed in his book, quite nicely I thought, is how any direction of development naturally runs into diminishing returns. For problem solvers to then fail to change directions in response necessarily leads them to ruin their environments with profiting from scarcity till they quite exhaust themselves. The problem isn't that we design our societies to grow, it's that we don't recognize the need to switch to the other kind of growth when we run into resistance. The other choice, stabilizing at a sustainable level of development, a comfortable distance from the lines of conflict with your environment,lets you change directions to fit better and better with your environment, like you see demonstrated by virtually every other maturing thing in nature, rather than exhaust yourself.

If we want to be part of nature we should look for our models in nature, and look mainly for our mistakes in our own pitiful history of failures.

Very good comment. Looking to nature, insects, rainforests etc should be the ultimate reality check for the way complexity plays out.

Hard to imagine nature using much exosomatic energy, which accompanied language, culture etc. Avg american uses approximately 240,000 kilocalories of energy per day, while only actually 'eating' about 1% of that. I can't imagine a tree or an ecosystem doing that...i.e. our behavioral constraints are different/looser than found in nature (in short term).

absolutely.

I tend to think of complexity in terms of phase state transitions, connectivity, self-organized criticality and interesting emergent stuff. I'm having a little trouble figuring whether J. Tainter is talking about general principles of problem-solving which any sort of system would have to solve in the same way, or more about a human idiosyncratic tendency to not consider anything other than 'complexity' as an answer, where complexity has a somewhat different definition than I might normally think of.

Then again, having not slept last night, I'm not sharp as a tack today, if ever. Cheers.

I tend to think of complexity in terms of phase state transitions, connectivity, self-organized criticality and interesting emergent stuff.

Greenish, I'd be keen to hear your thoughts on C. S. Holling's Panarchy Theory, which deals with just these matters.

There's a rather large divide between natural system and theoretical system scientists on this. I have it well defined for myself, but others don't seem to get it. It's the difference between exploratory science and representational science, thinking in terms of the natural features of the subject or our conceptual representations of the subject. The theorists all seem to cop out and say everything in your mind is theory so give up on discussing the real world because we can't. The naturalists seem to say, but it's what's not in our minds that is the whole subject of theory...

Anyway, I use a rigorous model of developmental sequences as a kind of "notepad" which is extremely helpful in organizing your explorations of
"phase state transitions, connectivity, self-organized criticality and interesting emergent stuff"

It's based on a rigorous extension of the conservation laws that I developed combined with observation to give the math some names. It's a powerful key that lets you start from data exhibiting the phenomena and then lets you look for what is producing it.

Hard to imagine nature using much exosomatic energy,

Huh? ALL heterotrophs use exosomatic energy, in the form of food that they eat. Cows live off of solar energy, captured for them in the grasses they eat.

Can't think of many examples beyond food, however. Birds exploit winds and thermals, some filter feeders exploit water currents. Can't think of any animals that build steam turbines, however.

Using a lot of energy doesn't have to be complex of course. I seem to recall that some early peoples would burn down forests to concentrate copper or kill game. I idly wonder whether insects ever evolved the use of fire, and had it not work out well for them...

I realize this is wildly OT...

and the toxoplasmosis protozoan causes a rat to sacrifice itself to a cat to continue its own breeding cycle; the amount of calories diverted would be huge compared to the size of the protozoans themselves...

I'm not sure I completely follow this, but are beaver damns an example of people are looking for. They don't exactly use the potential energy they create directly as kinetic energy, though (as far as I know), so maybe this is OT.

A beaver dam, and the pond that forms, etc would be an example of an "extended phenotype" which I've mentioned in one of my other comments to this keypost. (note: credit to Dawkins for that concept, too).

There are some trees (lodgepole pines, I think) that need a forest fire in order to for their seeds to sprout

Nate, the trick is to compare units relative to the system they correspond to. The easiest way to do that, that you can recognize without knowing very much, is whether systems switch from multiplying their use of energy to stabilizing it by refining their designs rather than increasing their scales. That's what we can copy.

Great point re: inward complexity vs. outward. This dovetails very nicely with an interview I was reading this morning with Janine Benyus on her work in biomimicry: The Sincerest Form of Flattery

1. Voluntarily Reduce Resource Consumption. While this may work for a time, its longevity as a strategy is constrained by the fact that societies increase in complexity to solve problems. Resource production must grow to fund the increased complexity. To implement voluntary conservation long term would require that a society be either uniquely lucky in not being challenged by problems, or that it not address the problems that confront it.

I see no reason to believe that problem solving must inevitably lead to increasing resource use. It depends upon what type of problem you are trying to solve. If you are trying to figure out how to increase your productivity in a world which is rich in resources but poor in knowledge of how to exploit them, then it follows trivially that successful solutions lead to increased resource consumption.

If, on the other hand, the problem you are trying to solve is how to maintain human welfare in a resource constrained world, then problem solving success will lead to quite different results. Certainly one can doubt whether the social intelligence required to address such a problem can be brought into existence. But to claim that this problem cannot be solved precisely because we are problem solving creatures is a logical absurdity.

Well, the way to read that is to pick out the true part. It's not that societies *necessarily* increase complexity to solve problems, but that using that strategy has been one of the most reliable human errors in problem solving. The way successful systems develop, whether human or non-human, is to switch from quantitative to qualitative solutions, switching off scalar growth which does indeed add overhead in complications, and switching to organizational refinement.

That's what plants do when they switch from their exponential growth using their seed resource to seasonal growth and becoming a part of their environments. That's also what people do with any project they successfully tackle. They start with expanding it and as the limits of manageable scale become visible they finish by perfecting it. That's also what people do in their own developmental growth, the growth of our bodies switches from exponential growth in the womb to maturing the scale of our bodies an then to maturing and expanding our minds, switching from outward growth to inward growth. That's the same switch in developmental direction in each case, from scale toward perfection. Nature made it up, and we might as well observe.

We make that same step to perfecting things before they become unmanageable with most things we manage personally. With over sized projects like building a whole civilization, though, we rely on following social rules. That keeps us from having an awareness of how the development is proceeding as a whole, so we are naturally unresponsive to the experience of the whole.

Using development curves to trace the stages things in lets you learn to experience the whole, and see when it's time to switch directions. Development curves help you see what's happening and provide a way, with a little effort, to find out what's doing it. They show when these developmental change-of-state events have, need to, or are about to occur.

We went through the whole earth "point of diminishing returns" for our approach to wealth creation (it would seem from many indications) in about 1950. Since then all added investment in scalar growth has been adding to our unsustainable kind of complexity instead of the inner growth we need to mature as a living system to become part of nature... (both physically and poetically it seems)

Well, the way to read that is to pick out the true part. It's not that societies *necessarily* increase complexity to solve problems, but that using that strategy has been one of the most reliable human errors in problem solving.

I understand that there is true part to Tainter's statement and acknowledged as much in my comment. What irritates me is the implication he makes that attempting to understand what is wrong with our strategy and adopt a new one is a waste of time. As you point out individual living organisms have solved this problem. I see no a priori reason to despair that the meta-organism of human society cannot ultimately solve this problem as well. This is not say that I have high hopes of a soft landing for human civilization. I think that a wrenching and painful transition is in front of us. But I do not think that we should simply throw up our hands and say that history proves that nothing can be done about our situation. One million years of human kind's history as tool using hunter gatherers did not define the limits of our social adaptability.

I think Tainter nails the issues.

However in my opinion he has in a way overlooked a solution.

First one has to think hard about how complexity is itself defined.

Carving a bowl out of renewable wood and lacquer or making one out of plastic arguably have similar levels of intrinsic complexity. In one case nature does the complex work in the other its humans. The key is mother nature generates complexity under constraint the complex wood is only produced withing the constraints of the natural system. Thus a better way to look at things may not be complexity but balance. In the case of the wood bowl we can even assume a skilled craftsman produced it thus even from a technical or knowledge perspective specializations are not all that different.

However we and Tainter see that the example of the plastic bowl vs the lacquered wood bowl are different. Even moving to clay and firing pottery does not seem to change this.

I call this difference simplicity in the general sense. In the case of wood or a clay bowl man is making the simplest change needed to create the desired solution. While the plastic bowl has a incredibly deep chain of specialists and god knows what processes and shipping things all over the world before you end up with a bowl.

Its still just a bowl.

Thus I think a world that actually avoids the problems we have today would be one where people would think it insane to produce a bowl out of plastic the way we do for the reasons we do. Its a world where there may well be a few plastic bowls but they would be made for very good reasons where other solutions don't work. I.e they would be the last resort and not discarded in the rubbish.

One obvious result is the wooden and clay bowls regardless of if they are local or imported are tied into the constraints of natural resources thus they are naturally if you will constrained by a known functional complex system. While the plastic bowls have no obvious constraints outside of running out of starting materials or energy. In general running out of people does not seem to be a problem.

Thus you have to work within this constraint of renewability and this sense of simplicity defined as the simplest change needed to a accomplish a goal. And the overall change should somehow probably also be energy neutral in the sense that the embodied energy is less than the expected or normal lifetime of the result.

Its ok to make bowls from trees or clay if the products last as long as it takes to replace the tree or fuel used to fire the clay. Preferably of course a lot longer.

This is not yet a rigid set of rules but you can see if you think this way your developing a concept of simplicity that seems to result in avoiding a Tainter style collapse of complexity.

There seems to be a definition of simplicity that can readily allow complexity but in a stable way. We don't have to revert to living in caves and mud huts.
However it seems we have to constrain ourselves first and foremost by living withing the constraints of nature and second it seems we must create extensions of the natural laws to cover when we get to far from natural products to ensure we work in accord with natural like constraints.

But these seem to be simply extensions of the basic laws of thermodynamcis with constraints on energy. I don't see that we need stray that far simply ensure that whatever we do the net energy balance works such that our own creations last long enough to have a minimal or almost zero impact on the natural order.

And finally of course under this sort of hard constraint it seems that reducing our absolute population would be considered sensible as the total for the entire population would also be considered critical. We would sum across not just individual items but also across the number of people.

If your village specializes in fine wooden bowls you would naturally work to ensure your population did not grow large enough that you cut all the trees down. Thus if we changed to effectively religiously manage our resource usage it seems very natural that population control would become almost a non-problem.
It seem that it falls out.

And of course all of this seems to just be a variation of WT's ELP economize localize produce. ELP embodies and probably better illustrates this intrinsic simplicity that seems absolutely required to develop a true sustainable society.

And last but not least I think you see how this sort of society appeals to the artistic over simply amassing a large quantity of manufactured goods. It would use bowls we would consider works of art as daily use items. Certainly it would care for them but you can see how the wooden bowl/ clay bowl society seems to naturally move to embellish and create beautiful works while the plastic bowl society dies in its own filth.

I think you and Nate both nail what I see as the weakest point of Tainter's work - his inability (and / or disinclination) to rigorously define complexity.

Your example of the bowl carved from wood vs. one made from plastic is a good one. Which is more complex? Sure, the plastic is the end product of a chain of manufacturing and transportation that we think of as complex, but it's nothing remotely approaching the complexity of the molecular and cellular machinery involved in growing a tree. The difference is that the one is something we have to manage ourselves, while the other is taken care of for us in mechanisms evolved by nature over hundreds of millions of years. Evidently, "complexity" is at least partially a subjective notion in which perceptions of risk and reliability play a part. We may worry that our access to the manufactured plastic could be cut off, but we don't worry about a breakdown in the growth of trees. (Unless maybe we live in one of the regions where pine bark beetles have been doing their mischief.)

I'm also bothered by Tainter's assertion that "complexity costs". He states that "In any living system, increased complexity (involving differentiation in structure and increasing organization) carries a metabolic cost." That's not obvious to me, and I don't think it's even true. What's true is that the differentiation in structure and increasing organization will generally enable an organism to acquire and process food more efficiently. And since it lives in competition with other organisms, it's usually adaptive for it to use that efficiency to enable a higher level of metabolic activity. But there are plenty of examples where complex adaptations enable organisms with very low metabolic rates to survive in marginal environments.

A first approximation to social complexity is the number of different full-time occupations in a society.

A hunter-gatherer society has maybe a dozen or twenty different roles. A neolithic agricultural society may have had a hundred or so; iron-age societies maybe two to five hundred. Modern Western society has of the order of a hundred thousand different occupations.

Does that tell you whether at any stage they work well together, or are always in conflict? That's another kind of complexity isn't it? I think one has to consider different stages of development, and the paths that lead to either lasting peace of endless war.. and things, to get at that. The first step is the put the puzzle into a manageable format, the basic "life history" that either successful or failing systems all follow in different ways. ¸¸.•´ ¯ `•.¸¸ “dust to dust with a little playful time between”

I did say "a first approximation". That said, peace, war and conflict have no obvious relationship to complexity. Some very simple societies have been warlike, and others peacful; likewise for much more complex ones.

The analogy is society <-> ecosystem, occupation <-> species.

(But like all analogies, it will collapse if you lean on it too hard. Hence, again, "first approximation".)

"peace, war and conflict have no obvious relationship to complexity. Some very simple societies have been warlike, and others peacful; likewise for much more complex ones."

Perhaps, but that may depend on your definition of war (and, of course, of complexity).

According to Raymond C Kelly in his _Warless Societies and the Origin of War_, the simplest societies, those with no or very little hierarchical and nested structures, essentially can't conduct what could be called wars, because they lack the ability to organize at that level. This does not preclude skirmishes, killings, territorial conflict...but it does seem to rule out anything that could be considered war.

A certain degree of wider and deeper organization is needed to mobilize enough people for war, specifically nested groups--families in clans in tribes...ultimately up to nations, to conduct what could be recognized as war.

Tainter asks the question in "The Collapse of Complex Societies" of why complex societies arise. In reading through his analysis, it seems to me that he misses a key economic point. Namely that complex societies arise where there is assurance of preservation of capital. In other words, complexity cannot arise unless their is assurances that investment will not be appropriated either by government or through external force. If society is organized to protect investment, it therefore bestows on those who invest an evolutionary benefit. For example, investing time and effort in irrigation project is only beneficial if one maintains ownership or access to the irrigated land. If it perceived that investment will be appropriated, one will not willingly spend resources on the effort.

Once society assures ownership of capital and investment, the next step for society is increased specialization. Specialization in turn is a form of investment that offers an evolutionary benefit to those who follow it. The investment though comes at a cost of not developing a broad range of skills though. But to specialize, one must also have assurances that one's skills will have long term and lasting evolutionary benefit.

So I, too, disagree with the statement that "increased complexity (involving differentiation in structure and increasing organization) carries a metabolic cost." From an economic perspective (think Smith and Ricardo), complexity arises from investment and specialization which should result in a lower metabolic cost system.

Yes, you're quite right about the key economic point about where complex societies arise. The key to growth is the "conservation of investment" as I would say it. The "increasing...metabolic cost" part of the problem comes into play if you're beginning more and more unfinished business, with a plan to multiply new investments forever, which is what the standard economic model assures will become a terminal albatross around our necks. Nature is fine with "wretched excess" if you find some way to do it, having no complaint about creating things wholly out of proportion to what was done before. Her problem with it is if you don't finish them! What is the phrase... "those whom the Gods wish to destroy they first make mad"?

Could you recommend any material that I might use to further my understanding on this way of thinking?
Books? Anything?
I think that others that frequent this board might also take interest.

I have various things linked from my Synapse9.com website. Browse the series of pop-up menu's across the top.

The main source in the end is going to be anyone's own original observations about how changes in their own lives begin and end, with different kinds of accumulation. Closely watch either planned or unplanned developmental change processes as they begin and end. How you start and complete a project at work, for example, has many of the same steps as starting and completing dinner at home. Any process of conserved change starts with setting the stage and then ends with completing the stage. I first noticed the phenomenon in design school, watching how design ideas would get picked up and passed around the studio, little 'viral fads' of experimentation, that would explore the whole territory of possibilities and then die out.

The one important thinker who seemed to see much the same actively exploratory world of systems created that way that I do is Jane Jacobs, in her two books on the origin of wealth and cities. There are lots and lots of great observers of different pieces, like Malcolm Gladwell, Joshua Ramo, Robert Rosen, Joe Tainter, and D'arcy Thompson, to name a handful. You might click my user name here and look at my recent comments. The OilDrum conversations have brought out a lot of my recent thinking I think.

The main source in the end is going to be anyone's own original observations about how changes in their own lives begin and end, with different kinds of accumulation.

Exactly!
I am exploring the demand side drivers of human behavior (as you might know), and I keep trying out things on myself (and I try to be honest, knowing our prevalence for self-deception). I KNOW (n=1) that we can be happier with far less throughput and less ambition, given certain environmental cues. Robert Franks constant anecdotes about our focus on relative vs absolute ring true to me in real time.

A collection of examples of individuals and groups living much lower on food-chain while leading healthy, fulfilled lives might be kindling for a larger social shift - but those changes will come about in some sort of adaptive iterative process, not by being told we need change - (that was my original thought on Campfire - to have people share those types of personal changes that worked, and didn't)

The costs are there, it's just that economics ignores them. The cost of specialisation and division of labour is borne by the environment in the form of "The Tragedy of the Commons".

Hi Roger, I'm contacting you on behalf of Bruce, the guy who came up with the idea of using hydrothermal vents to make electricity. He was interested in your idea of bubbling cold compressed gas in the opposite direction to get rid of precipitates in an old post. He would like to discuss this idea with you further. Could you please contact him at info@marshallsystem.com ?

The main question is not really the end product, but who does or does not do the finishing. The two kinds of complexity in nature are the kind used in starting things up and the kind used in finishing things off, branching out and then branching in. It's perfecting things that gives them the "intrinsic simplicity that seems absolutely required to develop a true sustainable society."

Our whole civilization is organized around starting ever more things the more we start. That's what using investment returns to multiply investments clearly defines, as the central pillar of finance and regulation in our economy. Soooo... this is a way to tell whether you're just starting ever more unfinished business or responding to the signals to switch to simplifying and completing them.

Your thinking to me is the perfect first approach to anything.
First ask.
What would Nature do?
The answer is all around us.

Indeed! And in some cases we open up to things around us so easily, and in others seem simply unable to shake the appearance that what we're looking at is in our own heads, and so observe nothing around us at all.

Complexity is more challenging to define singularly...

Frankly, I find this definition to be worthless. In fact, in reading Tainter's book "The Collapse of Human Societies", I found his failure to clearly define "complexity" to be a huge flaw in his argument.

Of course, his basic thesis is that societies collapse due to decreasing returns on increased complexity. But he never defines the term "complexity" with enough rigor to actually prove that thesis.

The Western empire collapsed, the Eastern empire didn't. Was this due to some difference in complexity? You can't say because Tainter doesn't define complexity with enough rigor to make any comparisons. Rome collapsed, but China didn't. Was China less complex than Rome? Who knows? If you can't actually measure or quantify something, how can you state unequivocally that it is increasing or decreasing?

Tainter says: we solve problems by increasing complexity. But that too is just a dogmatic statement because he gives no clear way to measure complexity.

Poorly defined terms are a classic hallmark of pseudo-science:

Indeed, in pseudoscience the more scientific-type language employed, the more ‘plausible’ it appears to be. However, all this really accomplishes is confusion. Poorly defined terms like 'complexity' are used with no useful explicit definitions provided. They are meant to look scientific, to look respectable in order to add weight to an idea which is in reality both implausible and improbable.

Link

Ultimately a growing society reaches a point where continued investment in complexity yields higher returns, but at a declining marginal rate. At a point such as B1, C1 on this chart a society has entered the phase where it starts to become vulnerable to collapse. (Source)

Why is this society vulnerable to collapse? Obviously, it's too much complexity. What caused the collapse? "Complexity." Who to blame? "Complexity."

Look, I understand what Tainter is saying. Moreover, he has done a mountain of really excellent research and hard work in a unique specialty.

I guess what I am trying to say is that he (or someone else) should tie it all back into human biology AND our present economic system. For example, capitalism increases complexity because every social pathology creates an opportunity for someone to profit by it. So instead of understanding why every other American weighs 500 pounds and doing something politically to eliminate the upstream cause, a good American capitalist spots fat boy and sells him some exercise machine, which only increases fat boy's appetite. Everything gets worse...

Lay the blame for OUR complexity on OUR present economic system -- on the way OUR current political decisions are made. Give concrete examples. Then perhaps someone besides we idiot savants in the blogosphere will start thinking about doing something to reduce complexity.

Trev--
Lay the blame for OUR complexity on OUR present economic system
No one, but ideological simpletons, can come to any other conclusion, as you apparently have, the capitalism goes, or we go.
The question is, what do we do on the other side of the wall we are about to crash into?
Inquiring minds want to know----

We're not going to go through that wall. We'er going to bounce off it. To reduce the impact I suggest resetting the value of debt obligations to equal the wealth producing capacity of the earth... and then allowing investors to add their earning to their investments only for scientifically qualified investments in long term sustainabiliy.

resetting the value of debt obligations to equal the wealth producing capacity of the earth.

This is an ecological economics theme. The reason it won't happen (and I assume you mean all money, currency, etc. not just debt), is the political will to shrink the system by 2/3s or more will never exist - i.e. the chaos that would ensue because various factions wouldn't understand that we had to pare down 1,000 trillion + in claims down to 30 some trillion in natural capital, and would fight.

Still, I know people working on just such a plan, though my guess is it would be a plan that might work after a collapse, not before. Still, we're on the same page - figure out our assets, then change the economic system to match them.

Yes, procrastinatin till all options are out of reach is a likelihood. I actually think the proposal I made for a practical choice to reset the economy could be done on some sunny afternoon by presidential order as his fiduciary responsibility to preserve our accumulated capital (in concert with the other leading powers). It would create smaller problems to solve, but would also rapidly right the system, end the continued deflation, and prevent the next approaching wave of collapse. Would you pass it on to others working in that direction and send me a link to theirs? As important as pulling off something like that would be having it argued about, I think, since that might open up other avenues.

I think debt forgiveness is the wrong way to go.
It should be a giant tax rebate.......the numbers are already ginned up and it repays those that paid in.
If you link it to earnings it may very well just reward those that "earned" creating the mess itself.
There really is no perfect way but trying to arrive at the lesser of the evils or the most equitable method needs to be looked at thoroughly.
Since the lower income brackets don't pay much if any tax maybe have a base amount that is significant and scale up the tax brackets from there for the rebate amount.
I will keep thinking about it.

Oh, the way I proposed it it's not "debt forgiveness" for the borrower, but "debt discounting" for all lenders. It would be a uniform discounting of debt obligations for the whole economy, not favors granted to people having hard times. You'd design it only to relieve the excess expansion of debt relative to real earnings, so that the information and the reality come back into line. What would prevent imbalance from reoccurring, and the workability of procedures for doing it are important too...

Debt discounting rewards heavy credit users and punishes savers because of the inflationary implications.
It reinforces the wrong behavior in relation for what the next phase of the economy needs to be..........that is, a zero growth mature system.
The reason I think it should be a rebate across the board is so that it at least attempts to keep everyone in the same relative positions financially.
If I knew that there was going to be debt discounting in the future I would go out and borrow every dime I could find and buy hard assets.
This seems like a dangerous response that might be creating a moral hazard.

My comment to this Campfire post.
Your letter to Bernacke.

It is a defacto way of devaluing the dollar and it simultaniously gets money into the hands of those that have been swindled and need it the most and of course back into the economy.
Of course our foreign debt holders will go berserk but who cares at this point.
Are the people running things really that incompetent not to see this relatively simple solution?

After that I guess we can start the long hard work of designing a stable state based mature economy.

http://campfire.theoildrum.com/node/5586#comments_top

Well, owning shares in a rich enterprise that collapses is not such a bargain compared to owning shares of a healthy one that prospers. Those are the two options we get by choosing direction flipping the "hindsight/foresight" coin. Should we flip a coin to see which one we choose?? ;-)

I'll assume that you're referring to this diagram:

That diagram is a perfect illustration of what I'm talking about. It's based on no empirical data whatsoever. Tainter literally just made up it up. I want to see one -- even just one -- diagram of that sort for an actual society, where the axes are labeled with specified, measurable quantities. You can make up any graph you want if it doesn't refer to any specified quantities in the real world. What is a "benefit of complexity" and a "level of complexity"? What units do you measure them in?

Furthermore, I don't accept his assertions that problem solving requires increased complexity, and that increased complexity requires more resources. Where is the rigorous proof of those statements, based on empirical evidence?

Look, I understand what Tainter is saying.

That's not good enough. Science must be based on something more rigorous than certain people being able to "grok" buzzwords with unspecified meanings.

From the paper I linked, some empirical evidence:


Figure 4.2. Debasement of the Roman silver currency, 0-269 A.D


Figure 4.3. Patent applicatications in respect to research inputs, 1942-1958 (data from Machlup 1962)


Figure 4.4. Productivity of the U.S. health care system, 1930-1982 (data from Worthington 1975; U.S. Bureau of Census 1983). Productivity index = (Life expentancy)/(National health expenditures as percent of GNP).

None of those graphs is even remotely in the ball park of the original graph. For example, all of them have time on the horizontal axis, not "level of complexity". In fact, none of them have anything whatsoever to do with complexity.

Note that Simon Tegg claims downthread that Tainter defines complexity as "vertical differentiation", i.e., levels of social hierarchy. What do debasement of silver coin or patent applications or life expectancy have to do with "vertical differentiation"? Nothing!

You are correct that these are not charts do not show vertical differentiation. However, the chart you posted is a conceptual model of the diminishing returns of complexity (anthropological definition). Falling patent applications per unit of input and less efficient healthcare per societal cost (it has probably worsened since then) do demonstrate (by proxy) the diminishing returns of complexity (anthropological definition). Charts may have different axes but show the same general concept. The grams of silver per denarius demonstrate (by proxy) the repeated attempts to fund current complexity from a declining resource base rather than reduce complexity. You are correct that this does not show diminishing returns to complexity (actually it shows the sunk cost effect).

As I point out downthread there are at least two definitions of complexity. If vertical differentiation, behavioural differentiation and number of unique artifacts are aggregated into "complexity" (anthropological definition) (this is usually but not necessarily true), then patent applications can be considered a proxy for "benefits" while number of r&d scientists a proxy for complexity (anthro def)

Seeing as this is topical, differentiation in health insurance providers is driving healthcare costs in the US, resulting in less net benefit per unit of input. Increased behavioural differentiation (private health insurance providers) + Increased vertical differentiation (Increased remuneration to health insurance providers CEOs vs. Nationalised system) = Increased Complexity (anthro def) >> Proxy = Increased cost.

Proxies are of course problematic and open to criticism

If you want to understand a fully rigorous model for discussing the developmental stages of open systems see my physics proof that shows why it takes a process to change a process... and my "chapters" page defining many of the characteristics. The scales used are not the same, but comparable, and it's far easier to understand what's happening in terms of change over time. The point of diminishing returns at the end of compound growth is when the returns on investment begin to decline, however you define the units of investments or the value of returns. It's a very useful "notepad" for seeing where things are in their natural development. The combined result is:

not

yes.. neither one says that everything in nature follows a formula that looks like these diagrams. Nature follows a sequence of changes that these diagrams can help you see in any particular circumstance.

One of the valid things everyone is objecting to here is the use of the term "complexity" as if it were a causal process of some sort. It's really a statistical measure that I think slips into the theoretical discussions from information theory and considering entropy as the universal theoretical cause of change in nature. That approach is "self-consistent" in terms of the physics, but omits a lot of what observably happens, like that everything that decays has to build up first. Statistical measures also do not point you to the actual physical processes and their interactions involved.

For example, one of the 'complicating' developements in our own economic evolution is our plan to have ever multiplying overhead costs and ever multiplying wealth to pay for them. That's now running into diminishing returns on all our resources, but not prompting a change in plan. The physical complication is conflict arising between every community of interest regarding what had become necessities to them, but are now no longer sustainable. Now we also have the complication of our minds being out of cynch with reality... and general solution for shortages seems to be "why not just let them have more", as in "why not let them eat cake", showing that we just don't get the picture yet.

Complexity is more rigorously defined, and contrasted with complicatedness in Allen, Tainter and Hoekstra, 1999. T

Complexity is the vertical differentiation in a hierarchy, complicatedness is the horizontal diffrentiation.

“…In hierarchical terms, complicatedness refers to the span of an entire hierarchy, the width of the lowest level. The notion of span may also be applied more locally to particular levels other than the lowest level. Complicatedness therefore can refer to a particular level in a hierarchy.
When evolution solves a problem with new infrastructure, it commonly leaves old adaptations in place as neutral characters. Thus evolution tends to increase the span of an organismal hierarchy, making organisms complicated. Span, created by horizontal differentiation, is independent of the depth of the levelar structure, created in vertical differentiation. Vertical differentiation in a hierarchy indicates an increase in the degree of organization in the system so described. Higher levels constrain and are contextual to lower levels in a hierarchy. Organization works through constraining the working at lower levels. Since increased organization comes from elaborating the number of levels, in our usage, complexity increases as there is emergence of new levels of control in a system”

In other words, peak hierarchy http://www.jeffvail.net/2008/02/hierarchy-must-grow-and-is-therefore.html

Tainter elaborates on how the Eastern Empire avoided collapse by reducing vertical differentiation. E.g. Redistributing land to the lower classes. While the Western Empire increased vertical differentiation. E.g. increasing taxes on the lower classes.

Tainter does not argue that complexity is solely reponsible for collapse, merely that beyond a locally determined point it weakens the society. When a crisis or series of crisis emerges it will be more economical to reduce complexity. Thus there will be no correlation between a given level of complexity and collapse as you suggest. Locally determined factors, (such as barbarian invasions in the Western Empires case, or good soil management in the case of China) play their part.

For more reading on the Chinese case and a critique/extension of Tainter's theory (with quantifiable terms) see How Civilizations Fall: A Theory of Catabolic Collapse http://www.xs4all.nl/~wtv/powerdown/greer.htm

Allen, T. F. H., J. Tainter, and T. W. Hoekstra. 1999. Supply-side sustainability. Systems Research and Behavioral Science 16 (5):403-427.

Thanks - I asked for a definition via email right before I posted this - he directed me to the 2006 paper - but I like the definition you provided is probably better (at least more understandable to me). I understand there is a paper following this speech that I assume will break it out in more detail.

And I'd say the inverse of this is really my definition of simplicity.

Simplicity is the elimination of these layers of control. In my bowel example a bowl goes from being its raw material to a finished product with as few as one person being involved i.e someone making a bowl for themselves. Higer levels can exist but they are not needed.

So the critical piece seems to be when higher levels become intrinsic and cannot be removed. I'd argue when this happens the system collapses.

As and example I'd suggest most businesses and people can readily get buy with very simple local banking the vast majority don't need complex banking and outside of security most don't even need a bank. A way to store money safely and make transactions cheaply is all thats needed.

However its not obvious that we have this problem of "to big to fail" thats actually a signal that the system has already failed.

Now given this one has to think that civilizations that avoided collapse did so by successfully removing layers of complexity nothing was to big to fail.
Indeed if you look at Eastern Civilizations and other long lasting ones like Egypt you find that they repeatedly fragmented and reconsolidated over time thus the prevented collapse by shattering. This was done by removing layers.

Its like the trick of pulling a table cloth off a table or a card tower. You can pull out a layer and the rest of the system remain fairly stable or it can collapse. However if you don't start pulling out the layers and refuse to do so your certain to collapse.

A stable society simply ensures that all things are readily compressible if you will to its simplest layer.

So regardless I still come to the same conclusion I have for a while regardless of how you refine the definitions and understand the problem better the answer is clear either you simply the situation on purpose even if it leads to some shattering or it collapses.

Trying to avert collapse remains the only certain way to ensure collapse. Practically any other solution results in a more favorable outcome.

It seems to me at least that the US is doing the one thing that is certain to ensure it will collapse.

One of the things omitted by defining a whole model of civilization in terms of control... is the opportunistic side of learning processes that is what serves to animate them. Yes, physics has not yet accepted that any causation is local, but clearly local development is where everything with a shape gets it's shape, and where every developmental process begins. Nothing in nature is "rubber stamped from afar" except the things we buy in the store. Where the idea of control systems breaks down completely is when describing systems in which every part needs to take care of itself. Then local learning systems are needed, responding to their environments opportunistically and responsively. That's most of nature, including us and our civilization.

It's not that understanding control isn't useful, it's just insufficient for defining living things. Understanding how things take care of themselves, using opportunistic learning processes generally, is clean missing from established scientific theory. It a very large part of why we find ourselves repeatedly unable to take care of ourselves when trying to follow our theories.

Complexity is the vertical differentiation in a hierarchy...

I read The Collapse of Complex Societies very carefully. In the section on the collapse of the Western Roman Empire (p. 128-152) Tainter does no counting at all of vertical levels of hierarchy. Did the Roman army actually increase in complexity (i.e. in number of vertical levels of command)? You wouldn't know from reading Tainter because he devotes zero time to actually examining that question.

Do governments or societies actually increase the number of levels of their hierarchy in response to problems? It seems like an empirical question to me. Where's the data? You won't find it in Tainter.

Does problem solving really demand increased vertical differentiation? I seriously doubt it. Even an ordinary corporation deals with hundreds of problems of week. If we are to proceed according to your interpretation, the company should have thousands of levels after a year in business. Is that really the way things work? I don't think so.

You are correct there is a problem of definitions here. In this article and most of Tainter's work the definition is the anthropological one i.e an aggregate of differentiation in vertical structure and behaviour and number of artifacts. A systems theory definition that I posted above equates complexity with hierarchical structure. Tainter implies that differentiation in behaviour and artifacts entails differentiation in hierarchy, therefore it is appropriate to aggregate them. This may not be necessarily be true. I believe this is what pfhenshaw is getting at upthread.

Tainter's model is a conceptual one only and a precise counting of levelar structure is beside the point and impossible to verify. As I have pointed out the model is not: single factor Complexity causes Collapse. therefore counting divisions and correlating this with...something(?).

I gave the example of increasing taxes on the lower classes. Clearly, this increases the vertical differentiation. The rulers appropriate a greater share of the society's wealth. This is not going to necessarily show up in a formal class structure but there will be fewer individuals with greater power, therefore more hierarchy.
The increasing structural complexity of the roman army is well established in other literature (try wikipedia: structural history of the roman army) and even the lay reader will know that the five levels of roman did not materialise with the founding of Rome (and must have grown).

You seem to be arguing from a positivist epistemology i.e if you can't directly measure it, exclude all other variables and correlate it with the phenomenom, then it is just subjective opinion. This may be appropriate for many areas but not for complex systems theory due to the computational complexity problem. Too many variables to realistically do this. Under a positivist framework what do you propose to test the diminishing returns of complexity as a distal cause of collapse?

There are other epistemologies with different but rigorous standards of validity. If you are arguing that positivism is the only valid one and complex systems theory is a pseudoscience then you are on very shaky ground academically.

Following your link in the original comment I note that you have replaced the examples with "complexity" in the quote without indication. Please don't do this for your credibility's sake.

Edit: posted to early

What about the complexity of rearranging everything ever faster. Does that fit into the discussion, the process complexity, or are we only talking about static complexity?

Good synopsis. Thanks. The theoretical model is missing the practical processes that natural systems use to get there, and some of the choices. The main choice for reducing complexity in an economic system would be spending investment funds, which slows down the continual reinvention of everything. Having growing investment means reorganizing everything ever faster, as we all see has been happening for a few hundred years, and now seems to have come to a head.

Using statistical models as causal models is quite hazardous. What, for example, is the difference between the great complexity involved in perfecting things and the great complexity involved in building more and more. In natural systems there is more than one kind of complexity. It takes a complex process to create a smooth piece of writing, but it works much better in the end. We need a smooth civilization, that works much better in the end.

Re: Poorly defined terms are a classic hallmark of pseudo-science:

TOD posts are often plagued with poorly defined terms. This is another one of them. To my mind complexity, sustainability and indeed energy itself are examples of the reification fallacy.

The reification fallacy is the fallacy of misplaced concreteness. Complexity is an abstract idea that exists mostly in the mind of the beholder IMO. So is sustainability. We enter the world of religious dogma when we can not or do not define terms concretely. Science requires concreteness for peer review and hypothesis testing. Undefined abstractions will not do.

Energy as is EROEI/Net Energy is the most egregious example of the reification fallacy that we often see on TOD. Some think that because they can measure energy in various units, it is concrete. It is not. It is undefined. It only exists in its various concrete examples each of which is unique. A generic energy does not exist.

Nor does a generic grain or metal exist. It is like saying grain is defined if we discuss it in terms of bushels or tons. Or that metal is defined if we discuss it terms of tons or ounces. Likewise generic complexity and sustainability do not exist. There are only unique concrete examples of them in the real world. These are what can be and should be examined and evaluated based on scientific evidence.

Complexity, sustainability and energy are all examples of misplaced concreteness. Their use without further concrete examples that refer to the real world rather abstract intellectual concepts is indeed pseudoscience at best and mental masturbation at worst.

Well, I'd agree with your main point except that if you rely on having well defined terms, since nothing in nature is well defined, you limit yourself to abstractions... it's a basic dilemma. My response to that is that one's terms should be well enough defined to point to something real beyond my definitions, like "apple". Apples may occasionally be "undefinably delicious", but I can at least point to one and someone can take a bite and see for themselves.

In that way I think we can identify clear distinctions between the different kinds of systainability. The absolute physical kind is not much like the metaphorical imagery kind, which while more popular is also rather commonly mistaken. Efficiency improvements most often cause increased impacts as a result of their growth stimulus, for example. Some things do complete their growth, become refined and efficient, and learn to successfully take care of themselves sustainably. I think that's intentionally general but quite clear, and if you look around you'll find examples. As you do more of the common characteristics that define the type come out.

You make an excellent point here, and using what I call "Tainter's razor" against him too! The main reason his study of the collapse literature ended up centering on the natural phenomenon of diminishing returns as being behind all the important collapses of history was his use of a strong principle of reasoning. It's that: "If some explanation in one situation doesn't apply in others, it's not the general explanation".

What led him to the phenomenon of diminishing returns was asking what all the collapses had in common, but it seems that the surviving civilizations also necessarily have that in common too. In this case he has slipped up on his own rule it seems, as ever increasing complications didn't occur for the civilizations that didn't collapse. He found a commonality among the collapses, but not the difference from the survivals. I've been writing about that, from a basic physics of natural systems view, since I first published on it in 85. Because of diminishing returns on any form of development, the growth of natural systems changes form in the middle of the process, switching from expansion to completion, or they fail.

What led him to the phenomenon of diminishing returns was asking what all the collapses had in common,

Yes, this is very perceptive and true. For some reason, Tainter was desperate to find the "factor x" which explained all the diverse species of societal collapse. It reminded me of a physician convinced, beyond all shadow of doubt, that all diseases can and should be unified behind a single cause, i.e. the "death factor".

So what better tool for this purpose than a chimerical, poorly-defined word like "complexity" which can fit into a round hole or a square hole, depending on the task at hand?

Personally, I find the traditional, "nominalist" explanation more convincing -- i.e., societies collapse for a variety of different, unrelated reasons, just like human beings die for various unrelated reasons. There is no "essence" of collapse.

I think the more valuable use of the observation is that is seems when the great societies of human problem solvers get tricked into turning their own solutions into their own worst problem, then the natural process of diminishing returns seems to be involved in what makes the trap inescapable. So, yes, it's a passive cause, but a real hazard for us anyway.

We get caught in doing things like last ditch efforts in the name of "sustainability" for maintaining the growth rate... (instead of our future) attempting to maintain the customs of the past.

I agree - it's pablum.

See my post below... the argument that diminishing returns is a problem is really not that interesting. What is interesting is the reaction to it - Diamond's Collapse is a far, far better examination of that aspect and does a much better job at explaining why some societies failed and other succeeded.

"Success, rather, consists of staying in the game."

And yet we all die, as do all species so far.

So what is success, again? Staying alive for as long as possible?

Keeping the fun meter running for as long as possible, and then ending life when it is too painful and not pleasurable enough?

Who decides for the individual? For the species?

I've been reading W. Catton's book "Bottleneck: Humanity's Impasse" and find Tainter's comments interesting in the light of what I've read from Catton so far.

Given human nature as evolved thus far, we are going to have this problem solved for us by nature, which we do not manage or govern or control.

It is worth noting that we do not control our own species: even while speaking of peace and justice, we have developed terrible weapons which are sold for profit to people who want to kill huge numbers of people.

We are far better at killing than at pro-creation, and so will commit species suicide if Mom Nature and Pop Sky do not do us in first simply by allowing the wreckage we have made of our habitat to do us in -- natural law, natural consequences.

We can have babies so far if eggs ans sperm meet and gestation takes place -- anywhere from one baby to several per pregnancy, depending on the circumstances.

We can and do kill thousands or even millions of people with various common uses of the tactics and weapons of war.

Again, when we consider how fast we can kill people, and how strongly motivated we are to do so, it seems like a logical conclusion that we will destroy our own species through warfare.

The other possibility, again, is that we will go extinct from the violence we have accidentally done -- fatally wounding our habitat.

The odds of our species living long and dying of "natural causes" in our sleep at a ripe old age -- metaphorically speaking -- are very tiny.

Eremozoic Age here we come. Apocatastasis, anyone?

I am waiting for a miracle while trying to stay somewhat comfortable, watching those wheels go round and wondering how it all works out and what it all means.

Also trying to help others -- and be helped by others -- through whatever this is.

Thanks for the song, Mr. Tainter.

Hmm. As the Benadryl kicks in at height of ragweed season, after reading your comment, I am struck by thought that capitalism, horizontal drilling, nitrogen injection, complexity, killing in name of tribe, etc. all concentrate per capita resources further towards top of hierarchy than without them. And something else central was on the periphery of a few neurons, but I just lost it. Nite, nate.

Ahhh -- will we be done in by .....gasp......ragweed?

I found your comments up above in the responses about the possibility of "pulsing" ahead with fewer humans, better technology, and a maybe some wisdom picked up from the Bottleneck we are in now. This is very good description of the kind of continued evolution I still remotely hope will happen. (And why do I hope that it will....? Nevermind that now....)

Catton keeps emphasizing that a "definition of the situation" is crucial to taking meaningful action that will help us survive and thrive.

One of the problems we have is that we have learned how to lie in order to survive -- blending in as hunters or as prey is a kind of lying to survive. But we have also learned to lie in very harmful ways to ourselves and to others.

Catton emphasizes that our complex, specialized and hierarchical cultures have brought so many more ways to defraud at the intersections of the almost infinite number of interfaces between interdependent people and specialties and hierarchical relationships.

This evil kind of fraud, is, I think, our fatal flaw thus far. We lie to ourselves so that we can maintain an illusion of security and an experience of immediate intense pleasure. We lie to others in order to take what they have and use it for ourselves. We lie to others so that they will go and fight and die so that we can feel safe within our own comfort bubble.

We are deeply disturbed by war and by the rapacious nature of our corporate culture, and so we invent really big lies -- meta-narratives, religious dogma, obsessive rituals, ideology, obsessive political rituals -- in order to hide from our own terror at the crimes we commit.

Ultimately our religons and ideologies break down simply because of their non-correspondence with the reality of our situation.

We are caught in our own lies whether we believe them or not.

We destroy our social, cultural, and political habitat through terrible acts of fraud and violence. We believe (or at least espouse) absurdities and so we continue to commit atrocities -- war and the like.

Truth -- accurate definitions of the situation we are in -- includes an accurate and unflinching look at ourselves, individually and collectively.

Perhaps some of these ideas would be helpful to some small remnant of survivors through the Eremozoic Age?

And we lie to our children to shield them from these truths, in the hope that when they grow up they will be be honest and strong enough to cut through the lies and win in the games of prisoners dilemma and tragedy of the commons that we all have to confront.

But they see through our lies and learn to lie in their turn.

I found this article quite illuminating. On one hand it reinforces my belief that the techno-fix with its increasing complexity is the complete opposite direction to where sustainability might be found. But I can't quite get my head around the author's assumption that conservation and simplification can only lead to short term sustainability (an oxymoron). I find this some-what of a contradiction in his argument when combined with his idea that sustainability is an ongoing process. Surely, mindful non-re-complexitivity (a new word I think!) can maintain a level of sustainability.

non-re-complexitivity :)

Thats the real problem you effectively have to fabricate new laws of nature if you will to ensure this does not happen.

Probably to the point of self genetic modification see my other post :)

Do like the dolphins; return to the sea.

I once posted something along the lines that we need to return to a simpler, less complex way of living to survive the various crises we're facing. I was remonstrated by a poster that I cannot remember the name of, but basically the argument was that we cannot return to a less complex way of living. IIRC the argument was based on the 2nd law of Thermodynamics and entropy. The argument being that whatever we do, it will automatically increase complexity and therefore resource use.

Whether the argument put forward is right or wrong in our context, it did make me think. My simpler, less complex way of living is to go back to a agrarian village lifestyle, but strangely enough, it is a more complex way of living than my previous city lifestyle. So maybe Tainter and the forgotten poster had a point.

For example; If someone buys farmland that was previously used for a monocrop of say wheat and turns it into a traditional farm with vegetables, pasture, orchards, animals, bees, etc. Is that an increase in complexity and resource use? Yet, it is what we would normally consider as a simpler way of living.

By "complex" I mean in the everyday common fuzzy usage of the word :)

I know it would be an increase in complexity for me, the individual but it seems as though it would be a reduction in complexity and an increase in redundancy for the group. It is kind of like reverse division of labor isn't it?
It seems to me that any way of life that attempts to get in synch with nature or at least closer to riding along with natural process is more sustainable.
I think you are correct and the phantom poster was wrong.
I don't see how the Entropy argument works or the automatic increase in complexity that is suppose to result from trying something new??
I agree with you and common sense here.

Maybe someone can offer some insight.

1. Voluntarily Reduce Resource Consumption. While this may work for a time, its longevity as a strategy is constrained by the fact that societies increase in complexity to solve problems. Resource production must grow to fund the increased complexity. To implement voluntary conservation long term would require that a society be either uniquely lucky in not being challenged by problems, or that it not address the problems that confront it.

This strategy has been used many times, from Edo Japan to many Indonesian Islands. Jared Diamand documents island societies that have survived for 3 millennium in "Collapse".

The trick is how to deal with the unsustainable societies that choose to exploit resources and energy in a suicidal manner to gain a temporary economic and military advantage?

Wait them out? Capitulate and start schools in the erotic arts so your population becomes highly prized and moves its genes throughout the world? Become paid mercenaries and put army bases everywhere so the populations genes move throughout the world?

Sorry, feeling cynical again. It seems that evolution made a good choice in a steep discount rate. And unless we find a way to address the underlying mechanics of why sustainable societies are going to get steamrollered, it seems suicidal to change how the brain operates. Maybe every sustainability program needs a few long lived nuclear weapon systems to back it up.

Tainter's argument hinges on two assertions:
1) Problem solving increases complexity
2) Increased complexity requires more resources

Now, here's a problem: marijuana smoking.

On the one hand, you could illegalize marijuana, create new laws and departments, beef up the law enforcement apparatus and prisons etc. This would indeed increase "complexity", and require more resources.

On the other hand, you could legalize marijuana. This would eliminate laws and departments, eliminate the need for law enforcement and prisons etc. This would reduce complexity, and free up resources.

It's not at all clear that Tainter's assertions 1 & 2 are true. Certainly they are true for *some* solutions of *some* problems. But for his argument to follow, 1 & 2 need to be true for most solutions of most problems, and he hasn't shown that.

Now, here's a problem: marijuana smoking.

Uh, how is that a problem (for society) in itself? The problems arise from society's attempts to manage this perceived problem.

Who was it who first said "the cause of problems is solutions"?

There is a historical precedent for society wanting to manage the marijuana 'problem': the "Gin Craze" in nineteenth-century Britain. But that was more a manifestation of poverty than of the attractiveness of gin.

Edit: Society's actual response to the marijuana "problem" actually confirms Tainter's assertion that the response to a problem is to increase complexity. Sure, there is another possible solution, but it's a bit further from the "status quo ante" than the chosen solution of trying to eliminate the use of marijuana.

Edit 2: If society ever adopts the alternative solution you propose, it will be because resource constraints - lack of prison space, lack of police time - 'force' the change to the simpler method.

marijuana smoking is a problem because of its negative health effects. It is a trigger for schizophrenia in the genetically susceptible. Long term use causes permanent changes in brain chemistry which cause paranoia and other mental health issues. Short term, it suppresses normal 'rational' thinking sufficiently to make the smoker less personally productive, and a danger to others in our complex society. (Don't smoke and drive!).

Of course, the overall effects of this are probably less than those caused by ethyl alcohol, and less than the cost of attempted (and failed) prohibition. However, the legalisation and management of CH3CH2OH has not exactly been a great success, where the management has been corrupted by the market to the point of making the problem worse.

"(Don't smoke and drive)"

Well, no. The National Highway Traffic Safety Administration sponsored a study that concluded that drivers intoxicated by marijuana were more cautious and seemed to be aware of their state and could compensate for it, than those intoxicated by alcohol. This is confirmed by actual experience, that there are actually very few (if any) cases of intoxication assault and intoxication manslaughter caused by marijuana.

I agree.

Also, even if problem solving only sometimes increases complexity and complexity only sometimes requires more resources complexity will accumulate and resource use will increase due to the sunk-cost effect. Politicians are not likely to risk contradicting the status quo by decriminalising marijuana. Private prisons have an incentive to support legislators who are tough on drugs etc.

You're still making dogmatic statements in the abstract, and not addressing empirical facts. I've shown one example of a problem where some solutions increase complexity and resource use, and some solutions do the opposite. I can think of many similar examples. As I stated above, an ordinary corporation solves hundreds of problems every week. Does the corporation increase in complexity and resource use with each problem? Why, exactly, does solving a problem *require* an increase in complexity?

Here's another example: Gas is getting too expensive, so I quit driving and ride a bike. Problem solved. How did that increase complexity? Where's the increase in resource use?

More generally, where is Tainter's *proof* that problem solving requires increased complexity and resource use? He frequently makes that assertion, but he's presented no evidence to prove it.

More generally, where is Tainter's *proof* that problem solving requires increased complexity and resource use? He frequently makes that assertion, but he's presented no evidence to prove it.

Tainter never said that. In fact he says just the opposite.

He says one course of action that a civilization can take is to simplify. He cites the eastern Roman empire as an example.

Of course a civilization could also increase its complexity. And he's made a very compelling argument in his book that that is a dead end via lesser marginal returns.

But he never said "that problem solving requires increased complexity and resource use"

he never said "that problem solving requires increased complexity and resource use"

His entire argument in this post hinges on that proposition. Read what he wrote again:

Tainter: "The fact that complexity and costliness increase through mundane problem solving suggests a different and startling conclusion: Contrary to what is typically advocated as the route to sustainability, it is usually not possible for a society to reduce its consumption of resources voluntarily over the long term. To the contrary, as problems great and small inevitably arise, addressing these problems requires complexity and resource consumption to increase".

You are misreading what he says. Increasing complexity is not the only problem solving mechanism employed by societies

At least part of the answer is that complexity is a basic problem-solving tool. Confronted with problems, we often respond by developing more complex technologies, establishing new institutions, adding more specialists or bureaucratic levels to an institution, increasing organization or regulation, or gathering and processing more information.

Most of the time, cultural complexity increases from day-to-day efforts to solve problems.

Contrary to what is typically advocated as the route to sustainability, it is usually not possible for a society to reduce its consumption of resources voluntarily over the long term.

But increasing complexity to solve problems (progressiveness) is the focus of this paper so that's what he focuses on.

He has written elswhere about reducing complexity to avoid collapse.
http://www.springerlink.com/content/l26772x2145752q4/
or
http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6V2S-4DPGWK4-3...

Apples and oranges, JD. Your examples are not problems for a society. They may be problems for small parts of a society, but not for the society as a whole.

Do you have relevant examples of problems for us to examine - problems that cannot be solved by individuals or corporations or small groups of either?

Tainter makes no distinction by problem size. In his own words: "as problems great and small inevitably arise, addressing these problems requires complexity and resource consumption to increase."

The examples I gave can easily be construed as social problems. Are marijuana smoking and the Mexican drug cartels/national park destruction which it fuels a social problem? Definitely. Are high gas prices which people respond to by bicycling or carpooling a social problem? Definitely. And yet those problems can be addressed with strategies which reduce complexity and resource consumption, as I showed.

In fact, there are often cases, like the current derivatives debacle or government/management bloat, where people explicitly recognize complexity itself as the problem, and deliberately simplify to address it. Does solving that problem increase complexity?

As Rethin notes upthread, even Tainter himself admits that simplification is a strategy for addressing social problems, and that conflicts with what he says above.

In short, I would say that Tainter is confused. What we need from Tainter and his supporters is an explicit and clear explanation for WHY, in general, problem solving requires increased complexity and resource consumption. Simply dogmatically asserting the proposition is not adequate.

Are marijuana smoking and the Mexican drug cartels/national park destruction which it fuels a social problem? Definitely.

So, it's not the marijuana smoking in itself that is the social problem, but the economic and social behaviour associated with it. Which were caused by making it illegal, creating new laws and departments, imprisoning offenders etc. As you say, society could have chosen to keep it legal and regulate it in the same way as, say, tobacco or alcohol - but it didn't.

Tainter is attempting to answer the question, "why not?"

Are high gas prices which people respond to by bicycling or carpooling a social problem? Definitely.

Are people, in fact, bicycling and carpooling significantly - not just a few percent - more than before? Have Americans' views on transport changed to the extent that bicycling and carpooling are primary options, and running your own car second choice?

Are people demanding more regulation of fuel prices and increased effort going into making fuel available cheaply ("drill, baby, drill", corn ethanol subsidies, etc.)?*

Why do we choose the latter, and not the former? That's what Tainter is exploring.

In fact, there are often cases, like the current derivatives debacle or government/management bloat, where people explicitly recognize complexity itself as the problem, and deliberately simplify to address it. Does solving that problem increase complexity?

You're blurring three kinds of complexity - the opaqueness of the financial instruments and of corporate ethics, and social complexity. The first two are problems, yes.

And what is the proposed solution? To increase regulation, which will no doubt mean more resources allocated to non-productive work - creation of new inspectorates, new laws, etc.

In other words, the proposed solution to the two problems is to increase social complexity and expenditure of resources.

Is this a good solution? Probably not. Will it be implemented? Probably. Are there other solutions that would work better, and more economically? Yes. Will they be implemented? No.

Tainter asks, "why?", and attempts to answer that question.

Finally, don't get hung up on the word "required". As I say below,

"My understanding of Tainter's view is that he says societies use "increase complexity to manage the problem" as a problem-solving strategy precisely because it works at first, and continues to work for quite a while. Eventually, the principle of diminishing marginal returns kicks in, and later still returns on increasing complexity turn negative."

Societies learn a problem-solving strategy that works at first, and continue to apply it beyond the point where it is useful. There is no "required".

----------------------
* I assume you accept that subsidies cause misallocation of resources, and overconsumption of some of them.

Are there other solutions that would work better, and more economically? Yes. Will they be implemented? No.

As a passing curiosity, what might those "other solutions" be? The problem amounts to the complete short-circuiting of human genetic programs which have succeeded for thousands of years, so such a sanguine treatment naturally arouses curiosity.

JD: "As I stated above, an ordinary corporation solves hundreds of problems every week. Does the corporation increase in complexity and resource use with each problem? Why, exactly, does solving a problem *require* an increase in complexity?...Here's another example: Gas is getting too expensive, so I quit driving and ride a bike. Problem solved. How did that increase complexity? Where's the increase in resource use?"

ST: I agree with you.
However, The bike example is a perfect example of "collapse" in action. As Tainter points out collapse is an economising process that reduces complexity. As more people adopt this action the auto industry will (continue) to collapse. And be partically replaced by a simpler bike industry.I prefer to define collapse as "reduction in hirarchical differentiation". As Tainter points out this may or may not be associated with reduced standards of living. Standard of living can improve for some groups.

Elsewhere Tainter points out the Eastern Roman Empire adopted this simplification strategy as its problem solving response avoiding a nasty *collapse* (strong cultural continuity) but still *collapsing* under my definition above. This contradiction and your quote from Tainter that problem solving *requires* complexity is a good example of Tainter's confusion of terms.

I think your criticisms of Tainter's terms are valid.That does not mean we should throw the baby out with the bathwater.

The issue is why do societies *generally* choose an increasing complexity strategy when it weakens the society in the long-term (This is Tainter's general insight). I have suggested the "sunk-cost effect", Charles Hugh Smith points out that elites have "asymmetric stakes in the game" http://www.oftwominds.com/blogjuly09/asymmetry-stakes07-09.html. Other comments in this discussion have added to this. Any ideas?

I believe there will soon be a very serious change in the way we deal with pot here in the states.

My parents generation-nearly all of them dead now-really believed it was the devils own personal smoke.

My generation grew up smoking it and spent the next thirty of forty years denying doing so.

Right now I can get nearly any old right wing redneck(quite a few middle aged right wingers smoke pot,but most don't) I know to agree -in private- that selling it at the liquor store and using a twenty to forty dollar tax per pack (as in a pack of cigarettes) to support the welfare state that supports medicare ,etc, is a very reasonable thing to do.

Nobody really believes any more that pot is as dangerous as pot prohibition except maybe a few little old ladies who are afraid of coffee and drink only tea.

Most of them have a kid or a friend who has a kid or know somebody they work with who smokes and probably every second or third person knows somebody who has been in trouble in one way or another in respect to pot,such as getting arrested, losing a job ,etc.-and very few believe the punishment fits the "crime"any more.

I expect pot to be a major source of tax revenue within ten years.

It would be legal now except for the fact that the law enforcement industry,the penal industry, the legal industry, the alcohol industry,and the pharmaceutical industry without even trying own enough votes and politicians to keep it illegal.But all these factions together can't fight the "gimme " lobby and the gimme lobby is slowly but surely awakening to this new source of revenue.

The educators tripped over themselves in thier eagerness to legalize gambling .They may sign on tp pot too,if they can figure out a way to make it smell good politically.

Any day now some up and coming serious candidate-one who will win- for the senate or a governorship is going to endorse legalization on financial and ethical grounds and once that happens ....

The problem is the bunch of people - authoritarians - who nail their lives' worth on how well they conform to what they think the rules are. And by extension, how well they can force other people into those structures. These are the people that fill the police and political systems. It will take a good chunk of other people to dislodge them and their out-of-proportion grip on power.

The jack-booted ICE butch or the local cop - they are all the same. As is the senate or gubernatorial candidate. Sorry; it has to be worse in any state bigger than Maine than it is in Maine and it's already a lost cause in Maine.

Thought provoking ideas but I think Tainter is wrong. The truth I think is much simpler.

A biological species will expand to use whatever resources are available. If that species happens to be "intelligent" then increased resource use will also result in increased complexity.

We fight a war half way around the world supported by nuclear powered aircraft carriers because we can. We won't when we can't. We will abandon our computers when there isn't enough electricity to run our refrigerators, and we will abandon our refrigerators when we have no excess food to store.

The path to sustainability will be for our culture to enforce a limit on "available" resources lower than that which would be enforced by mother nature. The latter limit of course being much more painful and possibly terminal than the former.

But how to limit available resources given our evolved resource maximization behavior, and the lack of interest and scientific understanding in the majority of our species? And without "good" dictators or waring tribes?

I come back to the same idea I have proposed here in the past. We need a new global religion with god as the biosphere. Religion is the most powerful cultural force for influencing the behavior of (less than optimally informed) people.

But how to get a new global religion going in time to save us? Perhaps we are screwed...

Yes, it's only our pattern in history that points to our being caught in a cognitive trap that other species are not caught in. I think the question is ... What will finally get fiduciaries to act in the true interests of their stockholders? Everyone, or a great many people anyway, sees that we're caught in a global tragedy of the commons. I think our problem comes from our solution, living our lives by rules, and being caught by rules that are out of date. Things that are successful are more responsive and resourceful than that.

This is the same conclusion I always come to. It seem to me that most people are not free thinkers. Most people like to be told what the rules are and wish for those rules to be enforced. The rules save them from doing any thinking for themselves. The rules are a safe haven, free from uncertainty. Just follow the rules and you are "Right".

The herd mentality. Not much advanced from Wildebeest.

I am arriving a little late in this discussion; but I can try to make some comments on this paper by Tainter, which is hugely interesting.

Now the point being discussed is about a definition of "complexity" and, indeed, it is curious that we speak so much about that and we don't have a universally accepted definition. Personally, I have a definition, which I don't claim to be the "right" one, but that you may consider: in my view, a complex system is one that shows at least one negative feedback loop. In other words, it is a system that maintains a homeostatic conditions when some external conditions change. So, a system doesn't have to be complicated to be complex. And it may be complicated without being complex. Said in another way, complexity is dynamic, complication is static. Think of ice in a glass of water, that is a complex system because the ice will melt or form in order to maintain the temperature of the system constant - until there is ice, the temperature of the water will stay around zero degrees C. When it is too hot and all the ice melts, then it is the collapse of the Martini on the rocks. Temperature shoots up; everything changes abruptly.

Social systems are much more complicated, but show homeostasis; they change some of their inner parameters to maintain their overall structure. As long as they can, of course. Reduce the supply of energy, and the social system will try to adapt by using energy more efficiently; up to a certain point. Then, at some point, without sufficient energy, society will collapse.

Now, Tainter is not a system dynamicist, but his work makes sense seen in terms of system dynamics. He has a very clear view of what are the interactions of the main elements that make up the social fabric. Reading Tainter, in a way, is like reading "The Limits to Growth" but without the math and the graphics. It is "mind based system dynamics" which is just as good as the kind made using computers.

This comment addresses just a part of Tainter's view, which is multi-faceted and rich. There is more; there is the question of how complexity arises and Tainter, perhaps, doesn't make it completely clear what is the mechanism that leads to that. He maintains that it is not energy that creates complexity, which I think is correct as long as you take what Tainter calls "the progressivist view" as meaning that there is some sort of automatic transformation of energy into complexity. It is not so, complexity arises because there are advantages to complexity - otherwise it wouldn't make any sense to have it. But, it is also true that without energy to sustain it, complexity could not survive. So, things are never simple and, indeed, we are dealing with complex systems. One characteristic of system dynamics is that there are no simple cause-effect relationship.

All right, then I would have to write a whole essay on this point; just let me point our at my humble essay on the roman empire (http://europe.theoildrum.com/node/5528) where i discuss Tainter's views in some depth. Then, of course, Tainter is a professional historian and I am just an amateur - I only tried to do my best.

Thanks, Ugo, for providing this extra perspective. From my systems dynamicist viewpoint, you've touched the points that were implicit in Dr. Tainter's speech, helping us left-brained types gain further clarity on this matter.

I agree with what others have hinted at already. Expanding populations are a natural outcome of basic thermodynamic law- in particular the law of maximum entropy production.

http://en.wikipedia.org/wiki/Law_of_Maximum_Entropy_Production

Tainter makes a needlessly complex argument about something that can be understood intuitively. He needs to internalize thermodynamic law and recognize what it means in terms of living things.

* Edit: Having said that, I would like to point out that it doesn't diminish the value of Tainter's discussion about what actions could be taken to head off decline. I'm just afraid that the science points to a most definite outcome unless we really are different enough that our consciousness would allow us to break the rules of the physical universe. That level of difference would suggest some kind of divine power or universal duality that many scientists would never admit, so pick your poison.

Thank you, Dr. Tainter, for permitting this speech to be published here. Good choice, Nate.

Surpluses are quickly dissipated by growth in consumption.

And of course we are seeing this via the Export Land Model and cries by some of "Drill, baby, drill".

My first reaction to "complexity increases in response to problems" was to question this in the light of a disappearance of surplus energy. Of course, though, my life has indeed become much more complex through the cultivation of 41 different edible tree/shrub cultivars, a small flock of sheep, a small flock of chickens, and a sizeable garden, on top of my full time job. Identifying those as my hobbies makes all the difference in the world how I perceive them and how I go about enjoying the time I spend doing them.

The complexity of society, however, may be significantly reduced in a post oil society, with a shift in population back to agriculture and away from many specializations.

I have a bit of a problem with this:

The Roman government had a clear sustainability goal: the survival of the empire. In response to the crises, the emperors Diocletian and Constantine, in the late third and early fourth centuries, designed a government that was larger, more complex, and more highly organized. They doubled the size of the army. This was very costly. To pay for this sustainability effort, the government taxed its citizens more heavily, conscripted their labor, and dictated their occupations.

With the rise in taxes, population could not recover from plagues in the second and third centuries. There were chronic shortages of labor. Marginal lands went out of cultivation. Faced with taxes, peasants would abandon their lands and flee to the protection of a wealthy landowner. The Roman Empire survived the 3rd century crisis and achieved two centuries of sustainability, but at the long-term cost of consuming its capital resources: producing lands and peasant population. When crises emerged again in the late 4th century, the empire lacked the resources to respond adequately and in time collapsed.

To me, this is not a problem of complexity per se but of politics. The empire under Augustus and a few later emperors faced many similar challenges but were much better able to handle them. One forgets that civil wars did more to destroy Rome's internal unity than any barbarian invasion - certainly by the crises of the 3rd and 4th centuries more Roman troops had been killed by other Romans than by barbarians. What caused this internal division and endemic civil war? Complexity? If so, why didn't you have the same problem under the earlier emperors? Likewise, the Republic can be seen as an analogy to the late Empire - it transformed itself into an allegedly more efficient system... why couldn't the later empire? Because the Late Republic/Early Empire had a few more places to plunder than the late empire?

Similarly, the economic edicts that are cited above are more of a political choice than a reaction to 'complexity' - the nobility and landed elites could not be taxed because they would bribe troops to rebel against the emperor. The lower classes had no such power and were thus easy targets for predatory elites.

I think Mancur Olson's 'Rise and Decline of Nations' is better at explaining what happened to Rome... or Kennedy's 'Rise and Decline of the Great Powers' - vested interest groups (in this case tied to military expansion) held the state hostage because of the influence they wielded politically. This led to choices being made that satisfied their interests and not what might be considered the 'public good' - over time the state became incapable of dealing with problems and, so, eventually declined/collapsed. This is not complexity or resource use, but politics.

I daresay Diamond makes the same point in Collapse in his comparison of societies facing resource constraints. Inevitably those that failed were unable to overcome issues of societal organization and self-interested behavior fostered by inter-group competition.

The weakness in Tainter's argument is that, yes, diminishing returns exist - but he does not dwell on the reactions to it that are so very important in determining outcomes of the type he is describing. Likewise, if we destroy ourselves over control of diminishing resources today it is not the environmental constraints that will do it but our reaction to it. Oil doesn't kill people, people kill people... (albeit sometimes over oil!)

4. Reduce Population. While this would reduce aggregate resource consumption temporarily, as a long-term strategy it has the same fatal flaw: Problems will emerge that require solutions, and those solutions will compel resource production to grow.

It's hard to take the rest of the speech seriously after reading this.

In plain English: "We can't find solutions to any of our problems, because any solution will have more problems, and those problems can only be fixed with more resource consumption."

Life requires resource consumption. If life is to exist, resources must be consumed. Because life must consume resources, life is unsustainable? Seems like a pretty flimsy argument to me.

I understand that reducing the population isn't something that is easy to accomplish. I understand that the population would have to be greatly reduced to reach a "sustainable" level. But, by dismissing this point without any real argument makes me wonder how carefully the rest of the speech was considered. It seems like Tainter has long since decided that Civilization is fucked, so any proposed solutions are unworkable. If this is the case, then Tainter is useless to me.

Interesting. I didn't read it that way. I read it as if there were an exogenous shock to system - say bird flu knocked out 3 billion people, that for a while there would be less complexity, but that the same drivers that got us here would re-emerge, meaning that the behaviors themselves after the bird flu were still not sustainable. Sustainable to me means steady-state - doable virtually forever. To have resources for 100 years then experience dieoff is not sustainable. Similarly, if due to previous population check, you now have 150 years of resources before hitting the point you would have 50 years earlier, that is just as unsustainable, just longer-lived.

I think the example you raise here - a strain of flu knocking out 3 billion humans - is interesting to look at in terms of complexity (even if I don't have a perfect handle on how JT is using it).

Our fossil-fueled overshoot has been leveraged by built-out complex infrastructure that will not be all that resilient to hiatus. We're now doing stuff with fairly high threshold requirements, like deep-sea drilling. We're leveraged by a globe-spanning extractive and just-in-time distribution network which is already fraying at the edges from rust. Since we're now on the downward slope of net energy, a significant plague, EMP attack, or other such perturbation would prevent a rebound to past levels. Each such rebound will be more modest, even though the behavioral biases in the human mind won't change.

If your coal mine floods, you may not be going back inside. If the factory that makes the computer chips in your mining machine goes out of business, it may halt. And rust never sleeps.

We're going to be learning about hiatus, I think.

It's listed under "some of the approaches commonly advocated:"

If you want to do something forever, I don't see how you can without maintaining a fairly constant population. While simply reducing the population isn't a fix in and of itself, any methods for maintaining stability over long periods of time is doomed to fail unless a certain population level is also maintained.

The problem with...

Problems will emerge that require solutions, and those solutions will compel resource production to grow.

...is the evident assumption that solutions necessarily multiply complexity. That is true for solutions designed as stop-gap measures, but not for solutions that completely resolve things.

That's the difference between unsustainable and sustainable solutions. I think it's mainly the difference between 1)starting solutions and leaving them incomplete, versus 2) starting them and then completing them. The way all our great solutions of the past come back to bite us with ever greater complications (that we push to crisis by trying to ignore them), just says we need to learn how to finish what we start to me.

I agree. The best solutions are simple and elegant and attempt to go with nature. It seems like we try to create Rube Goldberg Machines to force things way too often rather than "steer" things. It probably stems from the Western mindset. Man was made to rule the world and the world was made to be ruled by man......all that nonsense. It is time for a wake-up call. It is time we figure out that we are PART of the world.

Here's the reddit links for this post (we appreciate your helping us spread our work around, both in this post and any of our other work--if you want to submit something yourself to another site, etc., that isn't already here--feel free, just leave it as a reply to this comment, please so folks can find it.):

http://www.reddit.com/r/reddit.com/comments/9iw0e/joseph_tainter_human_r...
http://www.reddit.com/r/energy/comments/9imuf/joseph_tainter_human_resou...
http://www.reddit.com/r/environment/comments/9iw0a/joseph_tainter_human_...
http://www.reddit.com/r/collapse/comments/9iw0b/joseph_tainter_human_res...
http://www.reddit.com/r/Economics/comments/9iw0c/joseph_tainter_human_re...

Find us on twitter:
http://twitter.com/theoildrum
http://friendfeed.com/theoildrum

Find us on facebook and linkedin as well:
http://www.facebook.com/group.php?gid=14778313964
http://www.linkedin.com/groups?gid=138274&trk=hb_side_g

Thanks again. Feel free to submit things yourself using the share this button on our articles as well to places like stumbleupon, metafilter, or other link farms yourself--we appreciate it!

Just a comment: as much as I think Tainter is brilliant, one of the most enjoyable things about this string of response posts is seeing pfhenshaw's adroit suggestions about the way things are. Thanks pf.

I'll 2nd that.

you beat me.

Not that my opinion holds the weight of most here but I second that sentiment.

Replication and existence through time has been possible because our molecular constituents were able to utilize matter and energy that was relatively constant in supply and recyclable.

As we now operate, societal reproduction is impossible, both in terms of matter and energy. Through our ingenuity we have discovered technologies, materials and energies that should have remained undiscovered. We have only become intelligent enough to destroy perhaps billions of years worth of future organic evolution.

Now that the "good life" and the means to acquire and defend it have entered the realm of the non-organic and unsustainable, we have little choice but to continue the technological competition. Our human populations are quasi-speciated, in that we have had enough geographical separation to evolve varying characteristics and languages. This leads to competition for scarce resources and energy and an evolutionary arms race of complexity that will only end when everyone cries uncle and/or runs out of energy and material. It is much like two opossums fighting over road kill, unable to move from the headlights of the car that will soon smash the both of them.

The issue of complexity and failure of nation states is related to the energy and matter acquisition that the complexity allows or protects. Most previously failed nation states were still operating mostly within the organic realm. They pushed complexity until the wood and crops they grew could no longer sustain it. Today we have a different kind of beast, one that uses huge amounts of completely non-renewable energy. It has not collapsed yet, but it will, not because it became too complex, but because it nourished by an unsustainable energy source and created great toxicity in its discovery and use of new materials and processes.

Technological competition and population growth make us do things we probably shouldn't. In the end we will probably go over the cliff in a fatal embrace trying to bludgeon each other with our newest "high-tech" complex weapons.

Perpetual societal reproduction and adequate infrastructure to make lives comfortable is possible but not under circumstances of leadership that rarely rise above primitive tribal impulses.

This discussion about whether or not human societies can "simplify" and thus somehow survive seems to have ignored one huge, glaring fact: it has in fact happened, over and over again.

Just how do you suspect that islands populated by non-boat-building societies happen to get populated in the first place?

Obviously, their ancestors were at a higher level of "complexity". Whatever that is, and however you decide to define that term, it is clearly a more complex matter to build boats and get around in them than to not do so.

Yet, once the populations established themselves on the island, keeping up the complex technology of building boats and navigating with them sometimes proved to be unsustainable. Thus, some of them gave boats up.

This is not just a phenomenon confined to islands. Jane Jacobs devotes a whole chapter to this in her book Cities and the Wealth of Nations. She discusses at length a remote community in the Southern Appalachians, not so very far from where I live, where the population had progressively given up more and more technologies that they had formerly used, but that proved to be too much trouble to keep up.

Are such societies "sustainable". I don't know, but I believe that the point is this: To the extent that they are successfully able to continue to exist, they have done so by giving up those technologies (or "complexity", if you will) that they have found themselves unable to sustain. Sustainability, in other words, is achieved by not sustaining the unsustainable.

Of course, there are practical limits to such an approach. At a minimum, human societies need some way of feeding themselves and finding water and sheltering from intemperate elements, or they die out. There is only so much that societies can give up before the only thing left to give up is life itself. At a certain point, then, the sustainability strategy of giving up unsustainable things needs to shift to a different sustainability strategy, one where a very minimal but essential level of complexity is sustained, for the sake of the sustainability of the entire population.

So is the future to be humanity as the tame pets of a very few governing robot "minds"? The Eloi (H.G.Wells The Time Machine) of the future?

Heliocentricity

Life expectancy was much lower in the past. A lowered life expectancy will contribute to our population decrease from now on.

Islands had an inexhaustable supply of food, they had fish. While they could maintain the capacity to build coastal fishing craft and traps, their existence was seemingly assured. Chopping down trees to move stone heads in an unsustainable way is a sure path to destruction though. Didn't need technology to reveal that.

Don't underestimate our capacity to self destruct without complexity and technology. Island hopping was doomed too for obvious reasons, medium term it allowed the home islands to remain "sustainable" and increase the gene pool, hell knows what would have happened over time if a population was captive to a perceived sustainable existence, especially if methods were not introduced to allow an acceptable gene variation.

Population increase is a normal function of the human condition (and yeast). It can be proportionally managed by birth control, religious restrictions, taboos, human sacrifice, wars, euthanasia or even a combination of them all.
Presently we can afford the illusion that we are much more "civilized", that we don't need radical population control because we have fossil fuels and technology.

Now as the human population is due to crash, it had better crash to a number low enough to allow us to repopulate within the constraints of the remaining (degraded) life sustaining resources.

Wnc,

I intend to read the book you mention.

The Australian natives apparently arrived there with variuos technologies that were lost over the millenia due possibly to lack of the requisite raw materials but they survived and were doing fine up until modern Europeans arrived.

The local people in my immediate area gave up many traditional crafts and technologies apparently because they could easily survive without them-noboby around here has made cheese for instance within living memory- but there has always been plenty of milk even in the winter and plenty of meat and eggs-why should they have bothered?

Yes, that's a great observation. There are other kinds of curious dynamics for social structure development.

What if to prevent that a community like that had a bank, say, and the banker kept taking their profits to build new competition to "stimulate" them into being ever more creative to survive, what effect would it have? They might respond to it as hoped, they might catch on to the trick and throw the banker out. If it worked the banker might or might not see when he should stop that kind of thing on his own, perhaps. Without some level of challenge I think cultures might very well devolve, though.

With top predators in nature, like raptors and lions, a lack of challenge doesn't seem to make them devolve, though. It seems just being top predator is enough to keep them interested and having lots of time on their hands for other things.

I like the direction overall of Dr. Tainter's article, and my immediate tendancy is to agree with the thrust, as well as with most of the comments. One point I would make. While reading the comment about perhaps modifying the human genome to alter humanity into a sustainable species, a thought re-occurred to me. The logic goes as follows. Whenever I hear someone complaining about something they don't like about current society, I often think (though less often say) "All that's needed to fix that to your liking is to convince young women not to like it." It applies to everything from booming car stereos and hotrods to rap / country / pop music to kickboxing fights and rugby football games and yachts with helicopters on the fantail.

What would the future of the species (and the earth) be like if the mating-age female preferred small homes and sidewalking to mansions and sports cars?

Yes! The girls get to choose but they choose expensive. Remember the poor male peacock dragging that big tail around so that he can find some tail!

"They do it only because it works" Len Gould, 2009

I would like to combine two quotes heard often, with my own addition, to form my unified theory of humanity and its problems:

"All of society's problems stem from mans' inability to understand the exponential function, his inability to sit quietly in a room, and female sexual selection."

It appears to me that imbedded in Tainter's world view (plus that of many TODers commenting in this thread) is the presumption that complexity equates with vulnerability and tendency towards collapse. I am not sure that presumption is fully justified.

To begin with, I find that his definition of complexity cited at the beginning of this post is quite opaque, vague, and not very helpful. It really isn't of much help in allowing one to distinguish between what is complex and what is not complex.

As I think I've said before with a comment in a similar thread, to a large extent complexity is in the eye of the beholder. For example, what should we consider to be more complex: i) an elaborate 17th-Century cathedral clock with about a dozen functions showing phases of the moon, tides, plus including automata that go into motion with the striking of each hour, or ii) the tiny chip inside of your PC that keeps track of the time? To me, the answer is not at all obvious, particularly if one goes back into what was required to produce each.

Then there are many examples of how increased complexity can reduce vulnerability and how reduced complexity can increase it. As an example of the former, the Apollo lunar capsules not only had double but also triple redundant systems. This increased complexity considerably but also reduced vulnerability to failure. An example of the latter would be a supertanker with a single hull and a single screw driven by a one large engine. It is less complex than a double-hulled tanker with dual screws and dual engines but is more vulnerable to failure.

I associate vulnerability to failure with i) tight coupling with a high degree of inter-dependence among system components, ii) high precision with little room for error, iii) high efficiency with little slack, iv) lack of alternative modes should one fail, and iv) lack of spare capacity and back-up. Some of these things have to do with complexity, while others do not.

Getting back to what seems to be our favorite example, the decline and fall of the Roman Empire, I am a bit dismayed to see that it has become something upon which everyone seems to force-fit his/her pet theory. Lack of energy, excessive complexity, increasing entropy (a badly misused concept in my opinion), etc, have all been blamed by one person or another.

So I might as well add my own two cents. As I see it, as much energy and human capital were available during Rome's decline as were available during its ascent. I think what changed was that greed, venality, and plain old short-sightedness got the upper hand over civil-mindedness and resulted in a miss-allocation of those resources. The social contract became broken. In other words, a Roman functioning as a legionnaire in some godforsaken corner of the empire consumed just as much food and made just as many demands of production (as it was back then) as that same person would in the role of a productive farmer. The elite increasingly tried to extract more and more wealth from the common people and in doing so hampered their ability to be as productive as they might have. So I don't see how you can blame it on energy.

It all got down to people doing stupid selfish things in a desperate downward spiral of diminishing radius. The Romans basically f*cked up.

The social contract became broken.

Absolutely agree. I think you've nailed not only the fall of the Roman empire, but also the fall (if it was a fall and not simply a transfer of HQ from London to Washington) of the British Empire. What I'm still puzzling about is "what is it which is now vulnerable to falling?" I don't automatically accept that it may be the American Empire, as certainly per capita there are more Swiss at the Apex than Americans, and many other nationalities are included, with no explicit exclusions. So what empire exists today? Perhaps something "like" the Adam Smith / protestant Christain empire?

Regardless of that, "it" is failing now due to a shortsighted breaking of the social contract by excessively greedy "elites" / exploiters.

Joule, Lengould,

I agree heartily with you in regard to the fall off the Romans-the breakdown of the social contract in and of itself is more than an adequate explaination for the fall of Rome,which arose in the first place as a result of that selfsame social contract.

But I do not doubt that many various ecological/economic factors played a serious role in the fall,and the words of some confederate general come to mind too,as when he was asked why he lost a particular battle."The Union Army had a little something to do with it".

And our current impending collapse can be laid in large part at the feet of the elite exploiters you mention without any doubt but I am not ready to say that we would not nevertheless find ourselves in the same situation even if there were a world wide socialist government.

We wouldn't be carrying on this exchange w/o those exploiters and most of us would probably be dead already of natural causes.

I agree with you that (command / communist) socialism is not the solution. Better democratic government would certainly be worth a try.

Excellent post. Many spot-on observations.

It appears to me that imbedded in Tainter's world view (plus that of many TODers commenting in this thread) is the presumption that complexity equates with vulnerability and tendency towards collapse. I am not sure that presumption is fully justified.

Panarchy theory would have it that increasing complexity strengthens a system up to a point, and beyond a later point, increasing complexity weakens the system - more precisely, moves it away from reaching its potential.

This is certainly observed in natural ecosystems, where if interdependencies between species get too strong, failure of one species can cause cascading failures through the ecosystem, causing it to become radically simpler (for a while). For example, the loss of a wasp that is the only pollinator of a fig tree that provides food for several species of bird, lizard and mammal, which in turn provide food for predators, while the fig tree provides the nursery environment for several other tree species which in turn provide the only food and environments for yet other plants and animals...

My understanding of Tainter's view is that he says societies use "increase complexity to manage the problem" as a problem-solving strategy precisely because it works at first, and continues to work for quite a while. Eventually, the principle of diminishing marginal returns kicks in, and later still returns on increasing complexity turn negative. So Tainter's view seems similar to Holling's on this. It appears your own view is similar too.

It appears many commenters believe that our society has moved beyond the point where further increases in complexity are helpful.

As I see it, as much energy and human capital were available during Rome's decline as were available during its ascent.

That doesn't seem to match the evidence. In the late western empire, the minimum height requirement for legionaries was twice reduced, from approx 5' 10" down to approx 5' 6", and administrators still complained of not being able to get enough recruits. The legionaries refused to carry the traditional weight of weapons, armor and supplies on marches, because they were too heavy. Things like that (and other evidence on grain shipments, taxes, and so on) indicate that the late empire had considerably less energy available than the early empire -- for whatever reasons (yours are almost certainly some of them).

gregvp -

That is indeed an interesting fact regarding the reduced size requirements for Roman legions. I am very surprised that there were that many males over 5' 10" tall almost 2,000 years ago. If I recall correctly, that's even well above the average height for US males as recently as the American Civil War.

However, it doesn't necessarily stem from less caloric energy being produced. Were not the ranks of the Roman legions largely filled from the impoverished proletariat? And as that group become more and more exploited through taxes and various predations, perhaps they just couldn't afford as much high-quality food as before. Is it possible that Rome still had the same food-producing capability but that it just wasn't distributed in an equitable manner or that it's agricultural base was underutilized due to political bungling? Maybe more a matter of economics rather than actual physical constraints.

One must also consider that perhaps the people weren't really getting smaller, but that joining the legions just wasn't as attractive as it once was, so they had to cast a wider net. Just like the US Army has been lowering its aptitude standards to get more recruits. It doesn't mean that young people in the US have suddenly gotten dumber (though some might argue otherwise).

Then, demographics might have been a factor, as more and more people from various parts of the empire began to fill the ranks instead of just Romans (though it was reported that some of the 'barbarians' were of larger stature than the average Roman soldier).

When you're dealing with a period that long ago, data is sketchy and not always reliable, so trying to draw firm conclusions can be tricky business.

I too recall being show in grade 6 a typical piece of medieval armor which was too small in height to fit the average grade 6 student in 1960's. I'd vote almost certainly that 5' 10" would have been a giant in the Roman army.

My understanding of Tainter's view is that he says societies use "increase complexity to manage the problem" as a problem-solving strategy precisely because it works at first, and continues to work for quite a while.

The thing is, though, that sometimes a more effective problem-solving strategy can be to simplify rather than to complexify.

Perhaps the most classic example of that is the Japanese solution to the problems which firearms introduced into their society. They just banned firearms, and by thus keeping weapons technology at a more simple level they solved their problems by just avoiding them. This worked until the presence of an armed US fleet reintroduced the problem in a more unavoidable form. Were it not for the ability of armed navies to traverse the seas surrounding Japan, their solution to the problem of firearms could probably have been sustained indefinitely.

At the individual or household level, complexity in the form of more "stuff" creates its own problems - clutter, figuring out where to store it, finding what you need when you need it, cleaning and maintaining the stuff, learning how to use it, the technostress that goes with that, the money involved in acquiring and maintaining and replacing the stuff, etc., etc. Having a good clear out, downsizing, and simplifying one's life can be a very good strategy for solving a whole host of problems.

Another aspect to this: Consider how often the very best and most effective tools and weapons and other artifacts that some cultures produce are also very elegant and beautiful, precisely because they have be perfected though generations of trial and error in an evolutionary process until everthing superfluous has been honed away. The very best things that humankind has made are often quite simple in design. Simple, good design, where form follows and has been shaped by function, solves the problem for which it was created better than do more complex artifacts.

Thinking again of pre-Meiji Japan: The entire ethos and aesthetic of that society was deliberate simplification. At some level, they must have sensed that the continual addition of complexity upon complexity in a severely resource-constrained society was bound to ultimately create more problems than it solved. They must have somehow sensed that the deliberate shunning of complexity and cultivating of simplicity was a way to avoid the build-up of excessive complexity in their society, and was thus an effective sustainability strategy - at least until the modern world intervened.

That last bit, of course, is the big fly in the ointment, as others have noted above. It is easy to shun complexity and cultivate simplicity when you are isolated. When you are competing against other societies that have embraced and are moving up the complexity pathway, that presents a serious problem - maybe THE insurmountable problem. The Romans didn't just collapse of their own accord, they were overrun by the Germanic tribes. The Japanese didn't just give up on simplicity on their own accord, they felt that they were forced to embrace Western industrialism for their own survival. Most societies (with a few rare exceptions like the Easter Islanders and Greenlanders - don't be misled because they feature most prominently in Diamond's book, they are in fact exceptional) that collapse do so most commonly because they fail to survive contact with an enemy. There may very well be internal reasons why this happens, defeat may just be the final straw, but it is the encounter with an enemy force that is the immediate cause of their demise. Societies that sustain themselves can do so only for as long as they can defend themselves against all potential enemies. I would argue that defense is a bigger and more fundamental problem even than obtaining food and water, because if you can't hold your territory, you are not going to be able to get any food or water either - and if you are all dead, then of course nothing matters any longer.

While defense is a problem that requires a higher level of complexity to solve than a society would otherwise need, a common mistake that we have seen made over and over by societies is not knowing when to stop. Thus we see empires - including the current American one - expanded to the point where the empire collapses, and the society then exposes itself to greater danger than if it had never embarked upon imperial expansion in the first place. IMHO, the US would definitely enhance its prospects for long-term sustainability were it to disengage as quickly as it possibly could not just from the Middle East but from all of the Eurasian land mass, and to fall back to a more compact, defensible periphery at the ocean approaches to North America. This is a definitely case where simplification would be a more effective problem solving approach than would adding yet further layers of complexity by intervening in yet more places where we don't really belong.

I think you're onto a thought worth investigation, but find your suggested solution unworkable. The US cannot guarantee to defend itself in the 21st century by withdrawing behind its adjacent oceans, there are simply too many effective means for an attacker to breach any affordable defence structure. When Korea is launching potentially intercontinental missiles, India and China are putting packages onto the moon, and every navy in the world knows how to build practically undetectable submarines (either nuclear or hydrogen fuel cells will work for power), when an attacker simply needs to infect themselves with a biologically altered virus then come for a visit, etc. etc. etc., no defence system could be sufficiently effective to provide citizens what they would consider adequate security. Would you choose to have your military remain behind a defense shield if a foreign power wiped out a large city? Repeatedly?

No, the only possible entity which can choose simplicity and survival over BAU is "the entire world, all persons equally". Untill that is accepted, and whatever barriers to its effective implementation are resolved correctly, humanity remains vulnerable to self-extinction.

All the potential threats you mention are real. Yet, our imperial complexity does nothing to actually "solve" any of those problems. If anything, it just makes them worse. Not being a threat to other countries is actually a pretty good way to avoid being threatened by them. Not foolproof, of course, which is why countries still need militaries. However, most non-empires manage to avoid being threatened by other countries, unless they are unfortunate to live close to an aspiring empire. Even then, there are better solutions to their problem than imperialism. The Swiss did not need to aspire to imperialism in order to defend themselves from the German imperial threat - they just had to invest in a very good army, very well dug in.

Fortunately, the US is surrounded by those very big oceans, and that is quite a substantial advantage over a country like - say - Ukraine.

Wellll. Ok. You've identified the short-term solution, me the long-term.

The elite increasingly tried to extract more and more wealth from the common people

In the modern era, that is known as "maximizing productivity".

_________________________
Query: Is the productivity enhancement curve Bell-shaped or St. Matthew's Island-shaped? (In other words, does productivity crash after extending beyond an unsustainable upper level?)

"We clutter the earth with our inventions, never dreaming that possibly they are unnecessary--or disadvantageous. We devise astounding means of communication, but do we communicate with one another? We move our bodies to and fro at incredible speeds, but do we really leave the spot we started from? Mentally, morally, spiritually, we are fettered. What have we achieved in mowing down mountain ranges, harnessing the energy of mighty rivers, or moving whole populations about like chess pieces, if we ourselves remain the same restless, miserable, frustrated creatures we were before? To call such activity progress is utter delusion. We may succeed in altering the face of the earth until it is unrecognizable even to the Creator, but if we are unaffected wherein lies the meaning?"
--Henry Miller, The World Of Sex (1940)

The question of timing, as usual, is a critical factor!

Were these questions raised 500 years ago, the responses would have been very different.

However, I suggest that these questions are inevitably not raised until 1 minute to midnight (in terms of total species lifespan), when the dilemma all of a sudden becomes apparent.

In the past, during the rule of Rome for example, time wasn't critical because the ends of vital resources and the planets capacity to sustain its population were not critical.

There are also, inevitably, many factors that contribute to events and whilst Complexity is one, there are others.

That said, we now find ourselves at 1 minute to midnight and in need of some solutions!

Why? Because, as Nate's graphic showed, we (humanity) took some 4.5 million years, after the formation of this planet, to reach our first Billion people, it then took another 130 years to reach the second and in the last 80 years we have added another 4.5 Billion.

Equally, the cheap & abundant energy supplies that enabled our population & economic explosion over the last 200 years, is now in permanent decline and at CURRENT RATES OF CONSUMPTION, we will irreversibly decimate known reserves of our current energy capacity, within 20 years.

No, that doesn't mean it will ALL BE GONE, but it will mean that the economic model that drove the Global economy over the last 200 years, will be gone!

So, The question of "The Collapse of Complex Societies" is not just a theoretical one, it is something we are living thru, right now.

Therefore, timing is now critical!

Irrespective of complexity, we either find solutions now or at least buy some time or the operative word in Tainters work, will become the operative word in human society and that word is Collapse!

The only "solution", the only thing that actually will be done, is to give up trying to sustain whatever is no longer sustainable. This will be a matter of millions upon millions of cumulative individual actions rather than any deliberate societal decision.

For example, more and more people will find that traveling by air will be too expensive and too much trouble (added layers of complexity as terrorists probe over and over again for points of vulnerability), and give it up altogether; the increased cost of fuel will just make it even more expensive and thus cause even more people to give up air travel. As more and more people give up flying, more and more airlines will have to cut back flights, downsize, or even go out of business altogether. The airline industry will contract, and eventually it will get to the point where only the elite travel by air at all, and maybe the day will even come when that goes away.

This same pattern will be repeated over and over again across industries and technologies. Nobody will plan it, and indeed governments and corporations will fight against it, but it is inevitable and WILL happen.

I have been saying that the 21st century is going to be one long exercise in giving up things; THIS is what I mean by that.

Population decline is probably inevitable as well, but what I have described above will happen regardless of what happens to population.

Seems people are getting totally hung-up on Tainter's use of the word "complexity" by trying to over analyse it. From an intuitive point of view it makes sense and like pornography, you may not be able to define it, but you'll recognise it when you see it.

To my layman's view the car seems to be a good example of how complexity increases due to problem solving. Take the development that occurred between a model "T" Ford and say a Prius as various problems were solved over time. The Prius is undoubtedly more complex than the model-t and it also has more points of vulnerability as a result.

As complexity increases the framework which holds it together becomes increasingly rigid and aligned/connected/integrated into the environment in which it developed. This makes the system vulnerable to black swan type events or changes in its environment (many civilisations succumbed, at least in part, to climate change). The other vulnerability is due to internal dynamics of the system were complexity becomes too difficult to maintain and control (eg. financial collapse).

In our particular case we face both failure due to inherent internal flaws (financial collapse) and external shocks (resource depletion and climate change) in which our complex system is too rigid to adapt.

Burgundy -

You make some very good points.

I'm glad you brought up the example of the Model T Ford versus the Prius, as I think it nice illustrates several points.

Even though the Prius is orders of magnitude more complex than the Model T, it is also far more reliable and less prone to failure (and the Model T was quite reliable compared to most other cars built at the time). You can drive a Prius for well over 100,000 miles with only routine maintenance. Certain Hondas and Subarus routinely go well over 200,000 miles without major trouble. On the other hand, many of those old cars needed valve and ring jobs well before they even hit 30,000 miles. Tires were only good for a few thousand miles. Those old cars had to be constantly fiddled with.

Now, on the other side of the coin, and this is where complexity rears its ugly head, any person with a reasonable amount of mechanical aptitude and some basic automotive tools could repair and even completely rebuild a Model T. On the other hand, serious work on a Prius (as well as most high-end modern cars) requires highly trained mechanics using the latest diagnostic instruments and specialized tools. There is very little that the average home mechanic can do himself. So, it appears to me that as far as modern cars go, we have bought reliability and durability with increasing complexity. This is not necessarily a good thing, but I think it shows that complexity and vulnerability to failure don't always go hand in hand.

PS: On a personal note, I happen to own a 1968 VW Beetle that I have mainly as a fun car. I plan to keep it because it would make a perfect post peak car, as it is simple and relatively easy to repair. Parts are surprisingly plentiful. No power brakes, no power windows, no ABS, no traction control, in fact the only power items on the whole car are the engine and the windshield wipers.

Refined fuels and oils and due to cheap energy, better roads, tires, brakes, filters and so on.
Emission laws, safety requirements and road rules enforcement drove change.

Cars didn't need to be complex, a Prius on the road in the 1920's would be massively over engineered.
Population increases and the birth and growth of the suburbs mean't cars (and trucks) needed to evolve to cope.

I don't think complexity will be increasing from now on. We'll be looking to the exact opposite as our private worlds shrink and wither.

Many people here in my area (France) still drive the Citroen 2cv. What I would like to see is some of the old cars such as the 2cv, beetle, mini, fiat on the market again but built to modern engineering standards for reliability. Like you say with minimal need of power systems and also with much more fuel efficient engines. As for ease of maintenance, the old East German Trabant had many interesting features, a person could lift the engine out using just their hands :)

I think for any real progress in methods of transportation we'd need to see maximum speed limits reduced to 50mph or even lower. Until people can use the roads safely using cycles, horses, electric vehicles and fuel efficient small cars which will have limited speed abilities then no real change can be made in alternate modes of transportation. The easiest thing that governments could do to radically change the way we live is to prohibit high speeds on all roads accept motorways (freeways?).

As for ease of maintenance, the old East German Trabant had many interesting features, a person could lift the engine out using just their hands :)

I can do the same thing with my lawn mower motor. Not much difference between that and a Trabi, I guess. ;-)

Maybe it is not Complexity itself, but rather the fruits of successful Complexity that bring about a civilizational collapse?

Consider the following sequence of generations:

1. First Generation of Citizens: They work long and hard hours to build up an ever more complex, technological base, say it is computer technology (i.e., microcomputers, the internet, internet driven business models, etc.).

2. Second Generation (their children) observe that Dad (and maybe Mom too) worked exceedingly hard to get to where they are, but why should I when current technology enables a much easier and more luxurious life style? More to the point, Dad struggled hard to get a PhD in computer science and used a slide rule to design the early versions of modern computers. Why should I bother with any of that when I can go to the local Best Buys store and simply buy the latest generation of computer? No need to learn any science. I think I will become a finance person instead and use the complex computer technology to develop ever more complex business transactions involving credit default swaps.

3. Third Generation (the grandchildren) observe that Dad (and maybe Mom too) made millions from trading credit default swaps and got out of the market just before the whole thing collapsed. The grandchildren think to themselves: Why should I bother with any of that (what grandparents did and what parents did) when I can live at home (in the McMansion) and party hardy until the loot runs out?

____________________
Moral of the story: Spare the rod, spoil the child, and by so doing, burn down the empire (e.g. Rome)

That about sums it up.
I have seen your generational theory many times before in formal literature.
Usually there is a 4th generation sometimes called "Heros" that pick up the pieces and create the conditions for renewed advancement.

Your handle says a lot about your practical way of looking at this stuff.................maybe we are over analyzing this topic?

What an interesting discussion.

My reading of Tainter - and remember they're all reading's until he comes here and fully explains his thoughts.

0. Commonly species fill up an amount of resources available to them in due time (ref. plague mammals). I think this is a backing argument/belief, not made explicit by Tainter.

1. Surplus resources/energy above daily means ENABLE growth of complexity (as well as population).

2. Complexity is any activity / process / method / structure that is added on top of getting through the basic daily means (e.g. staying warm/fed/alive/reproducing). There's no hard objective definition of course, but intellectually fairly understandable through our daily experiences: arts, entertainment, bureaucracy, government, businesses, etc.

3. Unsolved problems that threaten resources and/or complexity are OFTEN attempted to be solved through new means, which in turn require added complexity. This speeds up the use of resources available, above the the baseline growth done commonly by species through natural population increase.

4. Added complexity rarely & truly solve problems of too much complexity using up too much resources to carry the population. Moreover, added complexity cannot solve the problem of populations growing in size to use up all resources anyway. As such, the problems remain - mostly unsolved - in which case added complexity (i.e. new ways of solving the problem) are thrown at the issue. This is not to say complexity may not bring benefits (efficiency, innovations, etc).

5. All this results in a natural population growth / complexity growth offshoot where - not the daily basic needs - but added complexity energy use momentarily surpasses the excess energy/resource capacity, as resources can be unevenly allocated and we can diminish future stocks after we have already used up today's allotted quota. We all know what this ends up in (ref: Catton), unless there is a reduction in complexity, possibly in population size and most of all in resource usage (e.g. powerdown).

So, in summary. Resource surplus enables growth/complexity, complexity speeds up resource usage, creates more problems (of complexity), creates more attempts at additional problem solving complexity. Positive feedback until the resources run out in an offshoot.

So why was China different?

Reduction in the overlord system, giving back the land, reduction in overall hierarchical governing complexity? Something else?

Complexity may impact medical care reform. When I was a child there were no antibiotics. My family doctor delivered babies, did major surgery, screwed ortopedic plates, attempted to devise cures for acne and gave vitamin B12 injections to his elderly patients. Amarillo did have one cardiologist who interpreted EKG's. There was a smattering of other specialists. When I first became involved with imaging science there were xrays but no cat scans, MRI's, fMRI's, pet scans, digital radiographs or nuclear medicine. Today there is a shortage of primary care physicians in part due to the explosion of specialists. Medical care has aslo beconme dependent on energy and materials. The coming decades will be interesting.

Elegance.

Accept no substitute.

Look at these guys wasting their time!

http://mousetraptomars.com/main.html