Technological Stagnation Is a Choice
REVIEW ESSAY
Invention and Innovation:
A Brief History of Hype and Failure
by Vaclav Smil
MIT Press, 2023, 225 pages
Do we live in a world of “ever-increasing change” characterized by “disruptive innovation”? Is “technology moving faster than ever before”? Are these, in fact, “unprecedented times”?
Contra the bromides of TED-talkers and Davos men, a growing chorus of contrarian scientists, scholars, and investors hold that the pace of innovation has slowed, not increased. They argue that the explosive growth and spread of digital technology has misled the public (and many policymakers) about the state of affairs in every other area of scientific progress. If we look up from the extraordinary virtual worlds depicted on our screens, we see that the future is receding before us.
But among the contrarians, there is a critical division. One camp, championed by Peter Thiel, holds that there are no physical or scientific reasons why we could not increase the pace of innovation, that technological stagnation is ultimately a social problem. The bureaucratization of science, the explosion of the regulatory apparatus, the demise of meritocracy, and any number of other social factors have slowed the pace of innovation, replacing actual technological magic with a digital simulation thereof. As Thiel famously summarized: “We wanted flying cars, instead we got 140 characters.” Call this the social theory of stagnation.
The other camp is best exemplified by the work of Vaclav Smil. Smil, a Czech-Canadian professor of environmental studies, has won acclaim for his prolific output of books on energy and society, characterized by telling sweeping stories while maintaining extraordinary attention to detail. Smil and others hold that, while the rate of innovation is slowing, this is a return to historical norms and not the result of some kind of social decadence. It was the extraordinary transformation of the nineteenth and twentieth centuries that was the aberration, as the scientific method unlocked the secrets of physics, chemistry, and biology in a onetime boost. Smil and others believe in the Singularity (an exponential transformation of the human species); they just believe we are already on the other side of it.
As outlined in his 2017 Energy and Civilization and in the final chapter of the book under review here, Smil draws our attention to the problem of scale in particular. As its name implies, the digital revolution was a revolution in the representation of numbers. Moore’s law relies on the ever-smaller physical means of representing numbers (since 1975, the number of transistors in an integrated circuit has doubled about every two years). In principle, a bit of information can be encoded into a single photon (in fact, in a subatomic particle). If computation is simply moving bits around, the physical limits of miniaturized computation have no analogy in other fields of human endeavor.
Contrast computation’s sustained geometrical increase in scale with advances in crop yields, energy density, transportation speeds, energy efficiency, infrastructure costs, and more. Where computational density has increased approximately 35 percent per year for the past fifty years, these other factors have improved around 1–2 percent per year. Thus, over the whole time span, they have improved by 1.65 to 2.7 times, while microprocessor performance has improved 10,000,000,000 times.
Moreover, as Smil is acutely aware, the world of macroscopic change (in contrast to microscopic digital improvements) besets us with trade-offs. Increase transportation speed and drag comes for you. Move away from energy-dense fossil fuels, and much of the carbon emissions savings are reversed as a result of building the many more solar panels, wind turbines, and hydroelectric plants required in order to power all of these electric vehicles, including manufacturing significant amounts of steel, aluminum, and other metals for which there are no reduced-carbon smelting methods. Almost everything about modern civilization is downstream from unlocking the energy density of fossil fuels. Moving toward decarbonization will demand incredible ingenuity and innovation just to tread water. Short of a radical new energy source, like fusion, growth will remain incremental at best. Call this the physical theory of stagnation.
In Invention and Innovation: A Brief History of Hype and Failure, Smil sets out to deflate our expectations of a futuristic utopia just around the corner, by focusing on the immense difficulty of innovation and the possibility of failure. Real innovation, understood as the social transformation wrought by mastering new inventions with the ideas, processes, materials, and organizations needed to make them useful, is extremely difficult. Smil wants to persuade us that it can go awry or can stubbornly fail to arrive altogether. His biggest bugbears are hucksters like Yuval Noah Harari and Ray Kurzweil, who seem to promise an audacious future that will inevitably land in our laps.
And yet, Smil’s book admits of radically different interpretations. Smil begins from the notion that we live in a world beset by innovation hype and that a healthy dose of realism is just what the doctor ordered. But is innovation hype really the dominant tendency today? What is the meaning of the stories Smil recounts if you presume instead that a lack of concern for innovation and growth characterizes our society? Smil’s book was highly persuasive for me, though not in the way he might expect; I began sympathizing with his position and finished the book vehemently rejecting it.
Smil’s distinctive approach is a phenomenology of innovation: put aside the blustery concepts and paradigms, and rush, like philosopher Edmund Husserl, “to the things themselves.” Smil wants us to understand that a technology as foundational as the diesel engine or nuclear fission reactor does not enter into the world all at once: it is the product of thousands of incremental improvements, path dependencies, and false starts. While one never wants to go toe-to-toe with Smil on the facts—the intellectual equivalent of facing George Foreman’s right hook—one can reexamine Smil’s careful narratives and find that they tell the story, not of failed innovation and of hype, but of human ingenuity and failure of imagination.
Smil’s prescription for innovation hype is exposure therapy: three case studies each for innovations that brought with them catastrophic side effects, that failed to meet their potential, and that stubbornly failed to arrive. But to the reader concerned that we don’t have enough innovation, these cases instead seem to tell the story of innovations which did tremendous good and whose unexpected side effects were readily dealt with, of innovations that were strangled in the cradle, and of innovations that demand we pursue them, even if they fail to arrive in the end.
In Defense of Leaded Gasoline
I was surprised to find myself, while reading Smil’s first case study on innovations that “turned from welcome to undesirable,” defending the utility of leaded gasoline. This initial section is intended to fully disabuse the reader of his naïve belief in the goodness of innovation as such. But each case admits of a subtler reading.
At first blush, it’s hard to defend leaded gasoline. Lead is one of the most damaging and durable neurotoxins known to man, particularly damaging to children’s neurological development. Due to the ever-increasing regulation of lead in the environment, the number of children with damaging levels of lead in their bloodstream has precipitously declined. Case closed, right?
Popular scientific narratives would certainly have you believe so. Viral articles on the internet proclaim Thomas Midgley “one of history’s most dangerous inventors” for his role in the invention of leaded gasoline as well as Freon-12, the first commercial chlorofluorocarbon refrigerant (Smil irenically defends Midgley only on the basis that many others were to blame too). But it is not merely a popular narrative. The Proceedings of the National Academy of Science, one of the most prestigious academic journals in science, published and later gave a prize to an article proclaiming that “half of US population exposed to adverse lead levels in early childhood” and calculating all of the millions of IQ points lost due to its use.
And yet, a closer examination of the study immediately reveals a troubling methodology. It establishes a correlation (actually, cherry-picks among several correlations) between leaded gasoline consumption and levels of lead in blood in the period for which it has actual data, and then attributes levels of lead in blood retroactively for the period (1940–75) about which it has no actual data, using an estimate based on leaded gas consumption! This is a pure quantitative instance of begging the question.
What actually is the relationship between leaded gasoline, lead in the bloodstream, and negative health outcomes? The sources Smil cites are curiously quiet about this question. The paper above (which Smil does not quote) justifies itself by positing that leaded gasoline was “the dominant source of lead exposure from the 1940s until the late 1980s.” This is not remotely in evidence.
The issue is one of correlation and causation. There is no doubt that the decline in environmental lead is correlated with the elimination of leaded gasoline (really a fuel additive called tetraethyl lead or TEL). But the abolition of leaded gasoline coincided with a revolution in environmental awareness. The immediate cause of bans on leaded gasoline was the rise of catalytic convertors to fight a variety of exhaust-related public maladies, including smog. Moreover, better methods for measuring lead in the blood and early studies on its neurological impacts on children led to a broad-based movement to remove lead everywhere, including other common sources of lead in paint, canned foods, industrial pollution, and more.
So to understand the actual connection between leaded gasoline and lead in the bloodstream (and presumed detrimental side effects) we need to go beyond correlation to causation. One way to examine this would be to carefully measure changes in lead in the blood at times and places when the only major environmental policy change was the elimination of leaded gasoline (the sharper the better). While every study focused on this transformative period (including those that did and did not coincide with other anti-lead measures) has found a strong decline in environmental lead with the elimination of leaded gasoline, post-ban lead levels depended heavily on other industrial activity. In an industrial city in China, one study showed that eliminating leaded gasoline only marginally decreased lead in the air. Another study in Shanghai estimated that leaded gasoline accounted for 30 percent of the environmental lead.
The best kind of study would prove that a specific lead molecule found in the bloodstream came from leaded gasoline. While difficult to design and implement, one study in northern Italy in the late 1990s managed to examine this, by substituting the TEL added to all of the gasoline in an entire region for one made with a distinctive lead isotope, and then isolating this isotope in the resulting blood samples. The study found that, for the most industrialized urban areas, leaded gasoline accounted for 25 percent of environmental lead consumption, dropping to 12 percent in suburban areas and 8 percent in rural areas.
And why did we add TEL to gasoline in the first place? Early engines faced a problem of fuel ignition called “knocking,” which resulted in inefficient combustion and damaged engines. A tiny fraction of TEL added to each gallon dramatically improved fuel efficiency and eliminated knocking; it also greatly lowered environmental pollution (a by-product of incomplete combustion).
Certainly, the elimination of leaded gasoline (alongside the deployment of catalytic convertors) has been an important victory for public health. But the vast majority of the hazards of leaded gasoline have been associated with the massive uptake of automobiles globally, and in the United States after the 1950s. What would have happened if leaded gasoline had been banned in those critical early years, 1921–50? Banning leaded gasoline at the outset of the automotive revolution would have lead to higher costs, worse efficiency, and worse pollution from the start. Worse, it likely would have set the advent of aviation back by decades: achieving the octane required for airplane engines without TEL is enormously expensive, and was initially impossible. The most common aviation gasoline used in propeller planes contains TEL to this day. In other words, even a simple counterfactual reveals that preemptively banning leaded gasoline would have substantially retarded innovation. It was precisely the massive success of automobiles and aviation that created the surplus and the incentives to solve the technical problem of fuel additives like tetraethyl lead!
I relay all of this detail not to defend leaded gasoline in particular but to call into question the replacement of one fantasy by another. What has been presented as a slam-dunk case turns out to be a nuanced and complicated story with, at best, an unsettling counterfactual.
Smil’s other two cases—DDT and chlorofluorocarbons—turn out to follow the exact same pattern, to an even greater degree. At least leaded gasoline has incontrovertible health effects. DDT is perhaps the most human-safe pesticide ever devised (to the point where one of the original methods of application was to pour it down one’s shirt), and an enormously powerful tool against disease-carrying insects. Its use saved an estimated five hundred million lives and entirely eliminated malaria (one of the most pernicious diseases known to man) in the United States, Europe, and many other parts of the world. CFCs, for their part, were a significant breakthrough, providing a safe, nontoxic refrigerant that made indoor usages possible (the previously used refrigerant, propane, had a tendency to leak and explode, so people put their refrigerators on their porches).
Each, of course, had substantial unanticipated downsides. DDT turns out to accumulate in soil and fat cells, disrupting, in particular, the reproduction of birds of prey. CFCs are so inert that they don’t react with anything until they get pushed eventually into the upper atmosphere where, catalyzed by solar radiation, they break apart, and the released chlorine atoms react with and destroy ozone molecules, leading eventually to a so-called hole in the ozone layer. Neither DDT nor CFCs are known (or even suspected) of killing anyone.
I would submit that the proper analytical structure for assessing the value of an innovation is the counterfactual: What would likely have occurred in the absence of the underlying technology? What were the actual trade-offs in play? In these counterfactuals, the case for the presumptive banning of each chemical is complicated. The bans solved significant problems but substantially weighed down a new invention that promised increased growth and prosperity. Moreover, because of the massive social utility each invention had created, including fostering entirely new industries, firms in the pesticide and refrigeration sectors had the incentives and the capability to devise replacements which provided most of the same benefits without all of the harms (although, it must be said, not without cost; DDT replacements are less effective, while CFC replacements turn out to be worse greenhouse gases). Needless to say, scientific evidence is used shoddily by advocates throughout. Smil set out to make the sobering case against certain kinds of innovation, but instead makes a strong case against public health officials using statistical analysis without a license.
Smil purports to be a fair observer, but his carefulness is largely unidirectional. Overhyping alarmists like Rachel Carson (author of Silent Spring, almost single-handedly responsible for the banning of DDT) are discussed with great sympathy (even if Smil points to some of their errors), while Smil heaps scorn on Elon Musk for the etymology of the word “hyperloop” (amounting to saying, effectively, your so-called hyperloop, good sir, is neither hyper nor a loop!). In no case do Smil’s innovation-skeptics or alarm-raisers have the science firmly on their side. The head of the newly created EPA admitted to overruling his own experts to ban DDT because of the public outcry not doing so would create.
Smil thus takes a curiously unscientific approach to the problem of iatrogenesis. By now, we know that interventions which aim to ameliorate a serious social ill may paradoxically create a worse one somewhere else. One study on the contraceptive effects of car seats in the United States found that, while the devices saved fifty-seven children’s lives in the study year, the additional costs they create for having a third child (and thus squeezing out of a regular sedan) prevented over eight thousand children from being born. Far more serious are, for instance, environmental regulations that incentivize electric utilities to preserve existing facilities instead of replacing them with newer, cleaner ones, or “historical preservation” laws that protect parking lots and laundromats.
The Chernobyl Effect
Smil’s next set of case studies, on innovations of great promise which never managed to meet their full potential, begins to inadvertently suggest the root of the problem. At first glance, the three cases addressed in this section—airships, nuclear fission reactors, and supersonic flight—seem like classic Smil stories, delving into the unit economics of cargo and passenger air transport and of grid-scale energy production. Certainly, in each case there is little doubt that promising technological pathways were closed off at least in part by fundamental, nitty-gritty factors: for instance, what drag at supersonic speeds implies for cabin size (very skinny design), and what this does to cost structure.
But this section raises an additional objection to Smil’s overall argument: that the headwinds these innovations faced were never merely economic, but political and social, and intimately tied to the rise of electronic media. You can see this at work in the first case study, on the stalled innovation of the airship.
Prior to roughly 1935, airships, like dirigibles and blimps, had enjoyed a far steadier career than the newer airplanes, which had significant limitations of range and speed: a transcontinental airplane trip like the British Imperial Airways London-Singapore service might require eight days and twenty-two layovers, in a cramped and noisy fuselage. In contrast, the airship offered luxurious public lounges and comfortable cabins, with gentle, steady, and safe service (in its 1.7 million kilometers of service, the Graf Zeppelin had zero injuries to its crew or passengers).
Over the long term, advances in airplane production doomed the airship, at least as the main aerial transport system for cargo and passengers. Better engines meant more thrust, which meant more lift, which meant bigger planes with extended range and speed. There’s no doubt that the airship’s days were numbered.
But the airship did not succumb to a doomed rearguard battle of unit economics. Its future went up in flames alongside the Hindenburg, which exploded while landing in New Jersey in one of the most spectacular and infamous disasters of the twentieth century.
Of course, what matters here is the context. Measured by loss of life, the Hindenberg explosion scarcely ranks among the top hundred or even thousand transportation disasters. The majority of passengers survived: only thirty-six were killed. Existing designs for airships did not require the use of hydrogen. In fact, the Hindenberg itself had been designed to use inert helium, except that America’s Helium Control Act of 1927 meant only flammable hydrogen was available.
What made a definitive difference in this case was the media environment and the political context. The great airship’s New York tour had been something of a PR coup for the Nazi Party, the swastikas boldly juxtaposed over the Manhattan skyline. But what really doomed the airship, even more than newspaper photographs, was the advent of the newsreel. In an era before special effects, the horrific inferno displayed in movie theaters throughout the country must have inspired dread and awe. An audio recording of the incident was broadcast nationally on radio—with it, the exclamation “Oh, the humanity!” entered into American parlance, so ubiquitous were the reports. It was, Smil records, “the first media event of the twentieth century,” yet he seems to view this as a coincidence, instead of a central thread throughout the book.
An alternative hypothesis hangs over the book, largely unexplored by Smil: that the story of the decline of actual innovation and the rise of a digital simulation of it is largely a story of the end of the print era and the advent of instantaneous electronic communication and the environment it generates. Print, a classic “cold” medium, could tell its readers of new scientific wonders, but could not grab their sensorium with two hands and shake them about. The radio, the newsreel, the television, the Twitter bring to one’s immediate presence not just the immediate terror of some technologized violence, but also the equally disequilibrating chit-chattering of the terrorized. Perhaps the “Global Village,” like the primitive village, is inherently suspicious of new technology and conservative about its transformative potential.
This certainly appears compelling for the two additional cases Smil examines in this section—nuclear fission and supersonic flight. Each technology has practical disadvantages (some obvious, some quite subtle) that dampened their transformative potential. But each was also the target of sustained and alarmist political opposition and sensationalist media reporting. More than any of the rather minor technical defects of nuclear power, the shuttering of the “Atomic Energy Commission” (job: build atomic energy) and the creation of the “Nuclear Regulatory Commission” (job: make nuclear power safe) stands in for a far broader shift in our mindset about innovation than Smil lets on.
This shift in the media and political environment is especially deadly when connected with the bureaucratic administration of science. As Edward Dougherty previously wrote in these pages (“Science without Validation in a World without Meaning”), modern science puts the policymaker, the bureaucrat, the administrator (he might have added, the journalist) in an impossible position. Their authority in part depends on clearly communicating to their organizations and to the public about their decisions. But modern science rarely (never?) offers readily-grasped rules of thumb, simple models, or clear reasons. Instead, the latest advances are shrouded in a probabilistic cloud, especially in the scientific modeling of complex systems like the climate, the immune system, and the spread of infectious diseases. The scientific method does not allow us to prejudice ourselves ahead of time as to what the “right” answer is going to be.
But this uncertainty is antithetical to bureaucratic rationality, with its emphasis on procedure, control, and the “one best way.” Combined with the immediacy of electronic communications, the temptation is to grab hold of a kabuki theater version of a scientific model, declare it “the official consensus,” and begin the construction of a regulatory and funding apparatus—regardless of whether the basic scientific questions in play have actually been answered (we saw this phenomenon in ludicrous relief during the pandemic).
The result is the appearance of scientific motion robbed of the thing that gives science its traction: the relentless commitment to careful experimentation and theory development, even if (especially if!) it makes no sense within the existing model. Since the advent of modern scientific government funding, has a single major scientific paradigm been overturned? How could it be?
Hyper Loopy
Smil’s real concern in this book—his disdain for “innovation hype” ungrounded in difficult realities—comes to the fore in his final section, on long-promised innovations that failed to arrive (the vacuum railway, nitrogen-fixing cereal crops, and nuclear fusion). Smil finally meets the object of his ire face to face in his case study on vacuum railways. Here, in Elon Musk’s 2013 release of the plans for a “fifth mode of transportation” dubbed the Hyperloop, you have all of Smil’s bêtes noires: a self-promoting huckster, duping all of the naïfs and “new-tech enthusiasts” who are “unaware of the long history” of similar failed projects and instead believe Musk’s proposal to be “amazingly original and stunningly transformative.” Meanwhile, Smil, grasping in exquisite historical and technical detail all of the challenges the hyperloop faces, knows better. “Transportation experts have rejected the Hyperloop plan. . . . All in all, not even a half-baked idea.”
By way of illustration of how long the fantasy of a hyperloop has beset inventors and innovators, Smil references an 1829 satirical cartoon (“The March of Intellect”) by William Heath (1794–1840), mocking the fantasies of inventors and the delusions of the investors they sucker into paying for their schemes, featuring a “Grand Vacuum Tube Direct to Bengal.” Smil liked the cartoon so much that it adorns the cover of the book.
And yet, the cartoon may not convey the meaning Smil intends. For almost every other fantastical invention mocked by Heath in the same cartoon came to pass within one hundred years, including the railroad, the airship, the airplane, the automobile, the powerboat, and the industrial factory. By what light should nineteenth-century inventors and investors have known that the vacuum tube would fail but all of these others would succeed?
Indeed, Isambard Brunel, perhaps the greatest engineering mind in history, thought the idea held great promise, especially for the power of pneumatic pressure to propel a train up steeper grades than locomotives could climb. He believed in the technology so much that he built a section of the South Devon Railway with it, at a time when the vacuum in the tube was maintained via “tallow-treated leather flaps.” (These ultimately proved the doom of the project, as rats found them delicious.)
While the Hyperloop solves no central problem for mankind, the other cases Smil addresses (nitrogen-fixing cereal crops that do not require fertilizer, and nuclear fusion) would transform human civilization. And yet, Smil seems curiously indifferent to their success, chronicling instead the hurdles they face (including, of course, regulatory, bureaucratic, and media-driven ones). At the end of the chapter, Smil tallies up the total cost of all fusion R&D between 1950 and 2020: $200 billion. In his conclusion, he suggests we would be better off focusing on extending and diffusing inventions we already have instead of pushing the boundaries of innovation.
And this conveys the real problem with the book. Smil, such a careful historian of innovation, has missed the forest for the trees. Whether any particular innovation will succeed, what determines whether a particular inventor is a huckster or the genuine article, may largely rise and fall based on the classic Smilian details. But what drives innovation for all of society forward are not the details, but the concrete belief that, with the application of curiosity, drive, and capital, it is possible to build transformative technology.
In 1967, the great social theorist Albert Hirschmann hypothesized the “hiding hand” principle. In fields of human endeavor which require creativity and invention, no purely rational calculus of costs and benefits based on a realistic assessment of the starting conditions could ever justify taking action. But, in the social aggregate, this would result in a severe underestimation of the transformative power of human ingenuity. What saves the day is the heedless and unmerited optimism (one might even say hype) of committed founders and doers, which leads them to oversell the project to others and to themselves, and to undertake a challenge which a fully “rational” accounting would never permit:
Creativity always comes as a surprise to us; therefore we can never count on it and we dare not believe in it until it has happened. In other words, we would not consciously engage upon tasks whose success clearly requires that creativity be forthcoming. Hence, the only way in which we can bring our creative resources fully into play is by misjudging the nature of the task, by presenting it to ourselves as more routine, simple, undemanding of genuine creativity than it will turn out to be.
To be clear, when it comes to technological innovation, Hirschmann’s hiding hand is better applied at the level of a particular technology, or perhaps even social transformation as a whole, not to a particular founder, firm, or invention. But the history of technology is clear that it was irrational enthusiasm, moonshot bets, and risky ventures that, in the aggregate, have driven technological transformation, not rational planning or careful calculation. Railroads were overbuilt and many investors lost their shirts, but social welfare as a whole increased. Google got off the ground buying servers for pennies on the dollar from bankrupt dot-com companies, and accomplished what none of them could. Airplanes and AI alike were forever a decade away—until they weren’t.
Smil’s crotchetiness predisposes him to dismiss major transformations in many of the technologies he chronicles. Venture capitalists are, for the first time, pouring funds into both fission and fusion start-ups, working with new reactor designs and scientific breakthroughs. Using new wave propagation models produced with supercomputers, both private companies like Boom Aerospace and NASA are actively developing quiet supersonic jets.
And this is perhaps where Smil’s own method damns him. The targets of Smil’s ire are those hypesters, hucksters, and fabulists who dream of a brighter tomorrow just around the corner (despite all contrary evidence) and sell this vision to the hapless public. He has attempted to apply a more or less realistic and scientific approach to the failure of invention and innovation. But he has entirely failed to apply the same approach to the alarmists and doomers and the iatrogenic effects that they might have, or even to the empirical effects of hypesters themselves. Whether or not innovation hype is a larger or smaller feature of our public discourse today than in the past is a relatively empirical question. Why not investigate it as such? Similarly, the effects of environmental activism and regulation seem like something that could be also investigated in a Smilian manner. The question emerges: why has Smil himself not done so?
Perhaps he intuits that a dispassionate accounting would undermine his whole thesis. For his two central claims stand in complete tension with each other: on one hand, that innovation hype is a damaging social phenomenon and, on the other, that the natural rate of innovation for mature technologies besides computing is 1–2 percent per annum. For if the latter claim is true, as the physical school of stagnation suggests, it implies the extraordinary danger of unwarranted caution and safetyism toward new technologies. Building new things is already so difficult, so unlikely, that we must be highly circumspect about what we let get in the way. Perhaps in a world where innovation was easy and science was generating new solutions every single day, we could afford the loose claims of a Rachel Carson or a Greta Thunberg. But in the real world, the Smil world, we cannot. In the end, the physical theory of stagnation seems to complement, not contradict, the social theory. Stagnation is a choice.
Smil did choose an apt subtitle for the book. In the stories he provides of the failure of nuclear power, nitrogen cereal crops, and DDT, we do have a “brief history of hype and failure.” It is the history of the hype of alarmists, degrowthers, doomers, safetyists, skeptics, and naysayers who demand we bow low under the diktats of their preferred regulatory bodies, “just in case.” The failure he recounts is a failure of imagination, and of will.