It was probably inevitable that last week’s discussion of the way that contemporary science is offering itself up as a sacrifice on the altar of corporate greed and institutional arrogance would field me a flurry of responses that insisted that I must hate science. This is all the more ironic in that the shoddy logic involved in that claim also undergirded George W. Bush’s famous and fatuous insistence that the Muslim world is riled at the United States because “they hate our freedom.”
In point of fact, the animosity felt by many Muslims toward the United States is based on specific grievances concerning specific acts of US foreign policy. Whether or not those grievances are justified is a matter I don’t propose to get into here; the point that’s relevant to the current discussion is that the grievances exist, they relate to identifiable actions on the part of the US government, and insisting that the animosity in question is aimed at an abstraction instead is simply one of the ways that Bush, or for that matter his equally feckless successor, have tried to sidestep any discussion of the means, ends, and cascading failures of US policy toward the Middle East and the rest of the Muslim world.
In the same way, it’s very convenient to insist that people who ask hard questions about the way that contemporary science has whored itself out to economic and political interests, or who have noticed gaps between the claims about reality made by the voices of the scientific mainstream and their own lived experience of the world, just hate science. That evasive strategy makes it easy to brush aside questions about the more problematic dimensions of science as currently practiced. This isn’t a strategy with a long shelf life; responding to a rising spiral of problems by insisting that the problems don’t exist and denouncing those who demur is one of history’s all-time bad choices, but intellectuals in falling civilizations all too often try to shore up the crumbling foundations of their social prestige and privilege via that foredoomed approach.
Central to the entire strategy is a bit of obfuscation that treats “science” as a monolithic unity, rather than the complex and rather ramshackle grab-bag of fields of study, methods of inquiry, and theories about how different departments of nature appear to work. There’s no particular correlation between, let’s say, the claims made for the latest heavily marketed and dubiously researched pharmaceutical, on the one hand, and the facts of astronomy, evolutionary biology, or agronomy on the other; and someone can quite readily find it impossible to place blind faith in the pharmaceutical and the doctor who’s pushing it on her, while enjoying long nights observing the heavens through a telescope, delighting in the elegant prose and even more elegant logic of Darwin’s The Origin of Species, or running controlled experiments in her backyard on the effectiveness of compost as a soil amendment. To say that such a person “hates science” is to descend from meaningful discourse to thoughtstopping noise.
The habit of insisting that science is a single package, take it or leave it, is paralleled by the equivalent and equally specious insistence that there is this single thing called “technology,” that objecting to any single component of that alleged unity amounts to rejecting all of it, and that you’re not allowed to pick and choose among technologies—you have to take all of it or reject it all. I field this sort of nonsense all the time. It so happens, for example, that I have no interest in owning a cell phone, never got around to playing video games, and have a sufficiently intense fondness for books printed on actual paper that I’ve never given more than a passing thought to the current fad for e-books.
I rarely mention these facts to those who don’t already know them, because it’s a foregone conclusion that if I do so, someone will ask me whether I hate technology. Au contraire, I’m fond of slide rules, love rail travel, cherish an as yet unfulfilled ambition to get deep into letterpress printing, and have an Extra class amateur radio license; all these things entail enthusiastic involvement with specific technologies, and indeed affection for them; but if I mention these points in response to the claim that I must hate technology, the responses I get range from baffled incomprehension to angry dismissal.
“Technology,” in the mind of those who make such claims, clearly doesn’t mean what the dictionary says it means. To some extent, of course, it amounts to whatever an assortment of corporate and political marketing firms want you to buy this week, but there’s more to it than that. Like the word “science,” “technology” has become a buzzword freighted with a vast cargo of emotional, cultural, and (whisper this) political meanings. It’s so densely entangled with passionately felt emotions, vast and vague abstractions, and frankly mythic imagery that many of those who use the word can’t explain what they mean by it, and get angry if you ask them to try.
The flattening out of the vast diversity of technologies, in the plural, into a single monolithic shape guarded by unreasoning emotions would be problematic under any conditions. When a civilization that depends on the breakneck exploitation of nonrenewable resources is running up against the unyielding limits of a finite planet, with resource depletion and pollution in a neck-and-neck race to see which one gets to bring the industrial project to an end first, it’s a recipe for disaster. A sane response to the predicament of our time would have to start by identifying the technological suites that will still be viable in a resource-constrained and pollution-damaged environment, and then shift as much vital infrastructure to those as possible with the sharply limited resources we have left. Our collective thinking about technology is so muddled by unexamined emotions, though, that it doesn’t matter now obviously necessary such a project might be: it remains unthinkable.
Willy-nilly, though, the imaginary monolith of “technology” is going to crumble, because different technologies have wildly varying resource requirements, and they vary just as drastically in terms of their importance to the existing order of society. As resource depletion and economic contraction tighten their grip on the industrial world, the stock of existing and proposed technologies face triage in a continuum defined by two axes—the utility of the technology, on the one hand, and its cost in real (i.e., nonfinancial) terms on the other. A chart may help show how this works.
This is a very simplified representation of the frame in which decisions about technology are made. Every kind of utility from the demands of bare survival to the whims of fashion is lumped in together and measured on the vertical axis, and every kind of nonfinancial cost from energy and materials straight through to such intangibles as opportunity cost is lumped in together and measured on the horizontal axis. In an actual analysis, of course, these variables would be broken out and considered separately; the point of a more schematic view of the frame, like this one, is that it allows the basic concepts to be grasped more easily.
The vertical and horizontal lines that intersect in the middle of the graph are similarly abstractions from a complex reality. The horizontal line represents the boundary between those technologies which have enough utility to be worth building and maintaining, which are above the line, and those which have too little utility to be worth the trouble, which are below it. The vertical line represents the boundary between those technologies which are affordable and those that are not. In the real world, those aren’t sharp boundaries but zones of transition, with complex feedback loops weaving back and forth among them, but again, this is a broad conceptual model.
The intersection of the lines divides the whole range of technology into four categories, which I’ve somewhat unoriginally marked with the first four letters of the alphabet. Category A consists of things that are both affordable and useful, such as indoor plumbing. Category B consists of things that are affordable but useless, such as electrically heated underwear for chickens. Category C consists of things that are useful but unaffordable, such as worldwide 30-minute pizza delivery from low earth orbit. Category D, rounding out the set, consists of things that are neither useful nor affordable, such as—well, I’ll let my readers come up with their own nominees here.
Now of course the horizontal and vertical lines aren’t fixed; they change position from one society to another, from one historical period to another, and indeed from one community, family, or individual to another. (To me, for example, cell phones belong in category B, right next to the electrically heated chicken underwear; other people would doubtless put them in somewhere else on the chart.) Every society, though, has a broad general consensus about what goes in which category, which is heavily influenced by but by no means entirely controlled by the society’s political class. That consensus is what guides its collective decisions about funding or defunding technologies.
With the coming of the industrial revolution, both of the lines shifted substantially from their previous position, as shown in the second chart. Obviously, the torrent of cheap abundant energy gave the world’s industrial nations access to an unparalleled wealth of resources, and this pushed the dividing line between what was affordable and what was unaffordable quite a ways over toward the right hand side of the chart. A great many things that had been desirable but unaffordable to previous civilizations swung over from category C into category A as fossil fuels came on line. This has been discussed at great length here and elsewhere in the peak oil blogosphere.
Less obviously, the dividing line between what was useful and what was useless also shifted quite a bit toward the bottom of the chart, moving a great many things from category B into category A. To follow this, it’s necessary to grasp the concept of technological suites. A technological suite is a set of interdependent technologies that work together to achieve a common purpose. Think of the relationship between cars and petroleum drilling, computer chips and the clean-room filtration systems required for their manufacture, or commercial airliners and ground control radar. What connects each pair of technologies is that they belong to the same technological suite. If you want to have the suite, you must either have all the elements of the suite in place, or be ready to replace any absent element with something else that can serve the same purpose.
For the purpose of our present analysis, we can sort out the component technologies of a technological suite into three very rough categories. There are interface technologies, which are the things with which the end user interacts—in the three examples just listed, those would be private cars, personal computers, and commercial flights to wherever you happen to be going. There are support technologies, which are needed to produce, maintain, and operate the output technologies; they make up far and away the majority of technologies in a technological suite—consider the extraordinary range of technologies it takes to manufacture a car from raw materials, maintain it, fuel it, provide it with roads on which to drive, and so on. Some interface technologies and most support technologies can be replaced with other technologies as needed, but some of both categories can’t; we can put those that can’t be replaced in the category of bottleneck technologies, for reasons that will become clear shortly.
What makes this relevant to the charts we’ve been examining is that most support technologies have no value aside from the technological suites to which they belong and the output technologies they serve. Without commercial air travel, for example, most of the specialized technologies found at airports are unnecessary. Thus a great many things that once belonged in category B—say, automated baggage carousels—shifted into category A with the emergence of the technological suite that gave them utility. Thus category A balloons with the coming of industrialization, and it kept getting bigger as long as energy and resource use per capita in the industrial nations kept on increasing.
Once energy and resource use per capita peak and begin their decline, though, a different reality comes into play, leading over time to the situation shown in the third chart.
As cheap abundant energy runs short, and it and all its products become expensive, scarce, or both, the vertical line slides inexorably toward the left. That’s obvious enough. Less obviously, the horizontal line also slides upwards. The reason, here again, is the interrelationship of individual technologies into technological suites. If commercial air travel stops being economically viable, the support technologies that belong to that suite are no longer needed. Even if they’re affordable enough to stay on the left hand side of the vertical line, the technologies needed to run automated baggage carousels thus no longer have enough utility to keep them above the horizontal line, and down they drop into category B.
That’s one way that a technology can drop out of use. It’s just as possible, of course, for something that would still have ample utility to cost too much in terms of real wealth to be an option in a contracting society, and slide across the border into category C. Finally, it’s possible for something to do both at once—to become useless and unaffordable at something like the same time, as economic contraction takes away the ability to pay for the technology and the ability to make use of it at the same time.
It’s also possible for a technology that remains affordable, and participates in a technological suite that’s still capable of meeting genuine needs, to tumble out of category A into one of the others. This can happen because the cost of different technologies differ qualitatively, and not just quantitatively. If you need small amounts of niobium for the manufacture of blivets, and the handful of niobium mines around the world stop production—whether this happens because the ore has run out, or for some other reason, environmental, political, economic, cultural, or what have you—you aren’t going to be able to make blivets any more. That’s one kind of difficulty if it’s possible to replace blivets with something else, or substitute some other rare element for the niobium; it’s quite another, and much more challenging, if blivets made with niobium are the only thing that will work for certain purposes, or the only thing that makes those purposes economically viable.
It’s habitual in modern economics to insist that such bottlenecks don’t exist, because there’s always a viable alternative. That sort of thinking made a certain degree of sense back when energy per capita was still rising, because the standard way to get around material shortages for a century now has been to throw more energy, more technology, and more complexity into the mix. That’s how low-grade taconite ores with scarcely a trace of iron in them have become the mainstay of today’s iron and steel industry; all you have to do is add fantastic amounts of cheap energy, soaring technological complexity, and an assortment of supply and resource chains reaching around the world and then some, and diminishing ore quality is no problem at all.
It’s when you don’t have access to as much cheap energy, technological complexity, and baroque supply chains as you want that this sort of logic becomes impossible to sustain. Once this point is reached, bottlenecks become an inescapable feature of life. The bottlenecks, as already suggested, don’t have to be technological in nature—a bottleneck technology essential to a given technological suite can be perfectly feasible, and still out of reach for other reasons—but whatever generates them, they throw a wild card into the process of technological decline that shapes the last years of a civilization on its way out, and the first few centuries of the dark age that follows.
The crucial point to keep in mind here is that one bottleneck technology, if it becomes inaccessible for any reason, can render an entire technological suite useless, and compromise other technological suites that depend on the one directly affected. Consider the twilight of ceramics in the late Roman empire. Rome’s ceramic industry operated on as close to an industrial scale as you can get without torrents of cheap abundant energy; regional factories in various places, where high-quality clay existed, produced ceramic goods in vast amounts and distributed them over Roman roads and sea lanes to the far corners of the empire and beyond it. The technological suite that supported Roman dishes and roof tiles thus included transport technologies, and those turned out to be the bottleneck: as long-distance transport went away, the huge ceramic factories could no longer market their products and shut down, taking with them every element of their technological suite that couldn’t be repurposed in a hurry.
The same process affected many other technologies that played a significant role in the Roman world, and for that matter in the decline and fall of every other civilization in history. The end result can best be described as technological fragmentation: what had been a more or less integrated whole system of technology, composed of many technological suites working together more or less smoothly, becomes a jumble of disconnected technological suites, nearly all of them drastically simplified compared to their pre-decline state, and many of them jerry-rigged to make use of still-viable fragments of technological suites whose other parts didn’t survive their encounter with one bottleneck or another. In places where circumstances permit, relatively advanced technological suites can remain in working order long after the civilization that created them has perished—consider the medieval cities that got their water from carefully maintained Roman aqueducts a millennium after Rome’s fall—while other systems operate at far simpler levels, and other regions and communities get by with much simpler technological suites.
All this has immediate practical importance for those who happen to live in a civilization that’s skidding down the curve of its decline and fall—ours, for example. In such a time, as noted above, one critical task is to identify the technological suites that will still be viable in the aftermath of the decline, and shift as much vital infrastructure as possible over to depend on those suites rather than on those that won’t survive the decline. In terms of the charts above, that involves identifying those technological suites that will still be in category A when the lines stop shifting up and to the left, figuring out how to work around any bottleneck technologies that might otherwise cripple them, and get the necessary knowledge into circulation among those who might be able to use it, so that access to information doesn’t become a bottleneck of its own.
That sort of analysis, triage, and salvage is among the most necessary tasks of our time, especially for those who want to see viable technologies survive the end of our civilization, and it’s being actively hindered by the insistence that the only possible positive attitude toward technology is sheer blind faith. For connoisseurs of irony, it’s hard to think of a more intriguing spectacle. The impacts of that irony on the future, though, are complex, and will be the subject of several upcoming posts here.