What Technology Has Accidentally Killed the Most People?

Illustration: Angelica Alzona, Gizmodo
Illustration: Angelica Alzona, Gizmodo

Show me a museum of important historical inventors and I will show you a gallery of deluded mass murderers. I’m not talking about machine gun manufacturers or nuclear scientists — those people, at least, have some sense of what they’re up to. I’m talking about the folks behind the printing press, the automobile, various kinds of boat technology. These people tried to improve the world, and succeeded, but also indirectly killed millions of people. That, at least, is the lesson of this week’s Giz Asks, in which a number of historians wrestle with the question of which technological innovation has accidentally killed the most people.

Peter Norton

Associate Professor, Science, Technology and Society, University of Virginia

In 1963, Tiny Helwig of the Winchester Repeating Arms Company said “Guns don’t kill people. People kill people.” But with the help of guns and other technology, people kill in far greater numbers than they otherwise could.

If we measure lethality by the fraction of the world’s population killed, the deadliest technology was the ocean-going vessel, made possible by innovations in ship design and by navigational instruments such as the compass and the cross staff. Like guns, ships didn’t kill by themselves; it took irrational ambitions, such as extravagant wealth accumulation, to give them deadly effect. In the century after 1492, the indigenous population of America declined by about 50 million as Europeans brought diseases such as smallpox and measles to the Americas.

But for total numbers killed, the combustion engine — any technology that turns fire into work — surpasses even this the ocean-going vessel’s horrific record. Toys that turned heat into action date to ancient times, but the transformative moment came in 1712, when England’s Thomas Newcomen designed an engine that used fire to create a vacuum, which in turn moved a piston in a cylinder big enough for a grown person to climb into. The whole machine occupied a freestanding building. It burned tremendous quantities of coal in return for a little work. But at the pithead of a coal mine, fuel was cheap, and by draining the mine of floodwater, it earned its keep.

Like the sorcerer’s apprentice, Newcomen released powers greater than he could imagine. His machine’s hardly recognisable descendent is the gasoline engine in a car. The first is an external combustion steam engine burning coal; the other is an internal combustion engine burning gasoline. But essentially they are the same technology: both apply combustion to drive a piston and thereby convert the stored chemical energy in a fossil fuel into useful work.

Until the eighteenth century all work came from muscle (whether from humans or from other animals), wind, or falling water. Since Newcomen’s 1712 invention of the first practical fuel-burning engine, a growing share has come from the combustion of fossil fuels. This machine relieved humans of toil and saved countless lives, for example, through better food and water distribution, better sewage disposal, and better access to medical care — all prodigious life savers. But in combination with human greed, the combustion engine also caused mass death. Applied to English textile mills, it drove insatiable demands for fibre, for colonies to supply the fibre in the form of cotton, and for slave labour to produce the cotton. The Atlantic slave trade preceded the combustion engine, but the engine made slavery far more lucrative.

The engine also enriched a new industrial aristocracy. Lacking hereditary titles, they proved their status through displays of wealth. One way was to serve tea sweetened with sugar — from slave labour. The slave trade killed many millions, and slave labour killed many millions more. Sugar plantations were the deadliest forced labour camps of the Americas. In combination with greed, racism, and indifference, a machine devised as a labour saver worked millions of people to death.

Pandemics precede the combustion engine, but steamships helped propagate the first global pandemic. Cholera, once confined to South Asia, spread worldwide in the 1820s and 1830s. Steamships accelerated the disease’s spread and extended its reach. Many of the cities that cholera reached were far denser than they could have been without combustion engines. Engines concentrated work, powering large factories; as locomotives, engines extended foodsheds enough to support cities of millions. Combustion engines made cities crowded; in crowded cities, disease spread quickly. Above all, waterborne diseases, especially cholera and typhoid fever, ravaged dense cities because sewage contaminated drinking water supplies. Combustion engines were responsible for the cure as well as the disease: their power made vast sewer and waterworks possible.

Motor vehicle crashes alone have killed about 70 to 90 million people over the last century; each year another 1.3 million die this way. People who live near busy traffic arteries are exposed to enough vehicle emissions to shorten their lives. In car-dependent areas, particularly in the U.S., sedentary living among people who have no good alternatives to driving contributes substantially to heart disease (the number-one cause of death in the U.S.), and also to life-shortening obesity and type 2 diabetes.

The combustion engine’s worst ravages may be yet to come. Invented to raise water, the engine is now performing this task on a global scale. Like the sorcerer’s apprentice, we don’t know how to stop it. Through their CO2 emissions, combustion engines are raising sea levels. They are changing the climate in ways that threaten livelihoods, food and water supplies, and the inhabitability of areas now crowded with millions of people. We find ourselves dependent on a machine that threatens to kill us. To manage this threat will require all the creativity we applied to develop it.

Jenny Leigh Smith

Associate Professor, History, The Hong Kong University of Science and Technology, whose research focuses on the history of food and food technology, among other things

The top three global killers in the past 40 years have been heart disease, cancer. and respiratory disease. Have there been technological inventions that have unintentionally increased these kinds of deaths? Two culprits leap to mind.The wide variety of machines and chemicals that transform staple foods into more delicious but significantly less nourishing processed goods have contributed to a rise in deadly diseases of affluence, including cancer, diabetes, hypertension, and heart disease. Pinning down a cause of death in people with chronic diseases is not as clear as it is with a fatal car crash, but doctors and epidemiologists are increasingly able to identify modern food choices as major contributors to poor health and increased morbidity. Expanding waistlines and insulin prescriptions are the visible side effects of a diet high in processed foods, but in fact what food processing technologies remove — fibre, micronutrients, “healthy” bacteria — is just as damaging. Polishing, flash frying and extruding the good stuff out of food has increased cancer rates, heightened immune responses and aggravated chronic diseases in ways that we are just beginning to understand.My vote for a second accidentally lethal technology is air conditioning. Climate control makes our modern way of life possible, but who is being controlled by what? Without A/C, America’s booming sunbelt would not exist, to say nothing of global financial hubs like Hong Kong, Singapore, and Dubai. Modern offices are designed around a constant temperate indoor climate, as are shopping centres. And here is where being comfortable seems like it might also be a little bit fatal. In the same way that food processing removes vitamins, roughage and bitterness, leaving only under-nourishing approximations of food, air conditioning removes the need and desire to be outside, to shift locations, to rest in the shade or exercise in the cool of the evening. Scientists have only very recently discovered that sitting is killing us. The technologies that keep us tethered our desks are worth interrogating. Air conditioning is a fair contender for the main culprit. By encouraging certain lifestyles — indoor work, information economies, shopping as a form of leisure, and a dependency on networked gadgets that work best in cool, dry environments — we humans have become conditioned to work year-round and at all hours. We sit, we stare at screens, produce knowledge and add value, pausing only momentarily to put on a sweater to guard against the eternal chill of the office thermostat.

Blair Stein

Assistant Professor, History, Clarkson University

Although there’s a temptation to say something like “the wheel” or “iron,” a lot of how I would answer this question depends on what you mean by “technology” and what you mean by “accident.” Here’s a thought experiment: if a pedestrian “accidentally” gets hit by a bus, what is the “technology” that killed them? The glass windshield that caused the fatal blow? The brakes that didn’t activate quickly enough? The headphones that prevented the pedestrian from hearing the bus? Or do we take a “guns don’t kill people, people with guns kill people” approach, which pre-supposes that technologies aren’t capable of killing anyone at all without human intervention? That is, was it the driver, not the bus, that killed the pedestrian? There are historians who make their living studying these questions about risk, culpability, and the consequences of technological accidents.

And once we start addressing historical technologies, we face the thorny issue of what we mean by “accident.” What is and isn’t an accident can depend on where, when, and who you are. Take railroads, for example. Setting aside people actually killed in railway accidents and while building the railroads, the infrastructural changes caused by the intrusion of rail in colonial contexts caused the deaths of millions of people. In the Indian famines in the last few decades of the nineteenth century, British colonial railways moved grain away from drought-stricken areas and into large stockpiles, and the global system of railways and telegraphs allowed Indian grain to enter larger markets instead of supplying local needs. In Canada, the “numbered treaties” system of the 1870s extinguished Indigenous claims to the land so the state could build the Canadian Pacific Railway, laying the groundwork for 150 years of fatal structural inequalities. To some historical actors at the time, these deaths may have been “accidents,” inadvertent casualties in the name of technological progress and nation-building. They also may not have been seen as directly caused by rail, but rather an oblique consequence of it. But railroads were part of the machine of empire in both of these cases, a technological manifestation of the impulse to remove people and ways of life that did not fit the imperial and settler-colonial dream.

I’m not necessarily saying that railroads have killed more people than any other technology by accident. I am saying that the history of railroads shows us that the more we think about blame, risk, and causality, the harder it is to answer that question.

Jonathan Coopersmith

Professor, History, Texas A&M University, whose research focuses on the history of technology

The automobile has been a major killer for well over a century. Every day in America car crashes kill 100 people and injure thousands more. If aeroplane crashes or terrorists killed 100 people a day, you’d have a much greater outcry. Instead, we take death by automobile as normal. What we have seen worldwide is that as automobiles are introduced on a larger scale, deaths go up significantly.

There was tension in the early 20th century over who owns the streets. What are streets for? Who should be allowed in them? By the 1930s, pedestrians were herded out of the streets — and when the cars became faster, the roads were built to be less accessible to pedestrians. One of the few positives of this coronavirus has been the lack of automobile traffic in cities, many of which have restricted streets to bike-only and pedestrian-only traffic.

That said, if people were dying at the same rate they were in 1970 per mile travelled, given how much more driving is being done now, you’d have nearly 150,000 deaths a year instead of “only” about 35,000. So the technology is certainly getting safer. And it’s striking how much safer the United States and especially Europe are than other parts of the world.

Raja Adal

Assistant Professor, History, University of Pittsburgh, whose research focuses on the history of technology, among other things

How about the printing press? Here’s the reasoning. It has often been noted that nationalism is responsible for most modern wars. Think of the first and second world wars, of Vietnam, etc. Nationalism is also very much linked to genocides from the Holocaust to the Rwandan genocide to the recent expelling of Rohinga from Bhutan. In all of these cases, one group of people who think of themselves as bound by some imaginary link — such as language, race, religion, or history — killed large numbers of people who, in their minds, belonged to another group. In the modern world, this sort of group identity is often nationalism. And…according to Benedict Anderson and others, the printing press was critical to the spread of nationalism in the modern world. It made it possible to print newspapers, novels, and other forms of literature that made people who did not know each other and lived far apart imagine that they all belonged to the same community. Therefore it can be argued that the printing press, which in itself is a completely innocuous machine, can be thought as linked to genocide, accidentally, of course. That’s the critical part. Technology does not have agency.

Peter Shulman

Associate Professor, History, Case Western Reserve University

If we’re looking purely at numbers and not population percentages, the world’s exponential population growth since the 18th century will mean that the number of recent deaths will far exceed earlier ones. The global population today is almost 8 billion; in 1900, it was just over 1.5 billion. A century before that, not even a billion. So, in a sense, this becomes a question about accidental deaths in the era of industrialisation. (If we were to look instead at percentages of total deaths, to the extent we could estimate such things throughout history, we might arrive at somewhat different answers.)

With that caveat, I want to propose that the technology that led to the greatest number of accidental deaths was James Bonsack’s 1881 invention of the mechanical cigarette roller.

Bonsack’s machine weighed a ton but produced as many cigarettes in a minute as an experienced hand-roller could in an hour; within five years, aspiring tobacco industrialist James Buchanan Duke had ten machines in operation. As Duke secretly secured preferential use of the Bonsack rollers on terms much better than any competitors, he initiated the mechanization of an industry. Add to the mechanised roller the introduction of flue-curing in the mid-19th century, which produced a milder dried tobacco that encouraged deep inhalation instead of holding smoke in the mouth, and the growing cigarette industry had the tools to mass produce a global public health disaster.

Tobacco had, of course, been consumed in the pre-Columbian Americas, and by the 17th century, was a major cash crop grown for export in the tidewater English colonies. As its use began to spread across Europe, critics assailed its safety, but it grew in popularity nevertheless. By the later 19th century, most of this tobacco was smoked in pipes or as cigars or chewed (the latter a particularly American habit). But the cheapness, ease of consumption, aggressive advertising, new global markets, alleged cleanliness, and simple ubiquity of the cigarette brought it into far more hands than ever before, just as the global population rocketed ever upward in the 20th century. All these features depended on an ability to mass produce cigarettes, a dubious accomplishment we can attribute to James Bonsack and his most important customer, James Duke.

Even today, when US tobacco consumption has plummeted from over 40% of adults to less than 15% over the past half-century, the country still logs nearly half-a-million deaths a year attributed to smoking, from heart diseases to lung diseases to cancers. Globally, over 7 million people a year die due to tobacco-related causes; as the World Health Organisation estimates, that’s about one in ten of all deaths.

Yet after over half-a-century of public health warnings and the incontrovertible evidence that tobacco companies both privately knew of the health risks of their products while publicly insisting on their safety–all the while working to obfuscate public understanding–should these deaths still be considered “accidental”?

Asif Siddiqi

Professor, History, Fordham University

The slave ship: Although the Atlantic slave trade started fairly early, roughly in the early 16th century, the use of ships specifically designed to transport slaves peaked in the 17th to the early 19th centuries. As Markus Rideker has described in his book The Slave Ship, these ships were essentially mobile prisons. The slave ship was also a factory of sorts, in the sense that was a mechanised system where merchant capitalists assembled large numbers of workers that transformed free men and women from the West Coast of Africa into a commodity, a “slave.” Finally and most obviously, the slave ship was a complex technological system designed to transport human beings over very long distances with minimum care to the wellbeing of those people. Because it was not specifically designed to kill the ‘cargo’ during the Middle Passage (although this was undoubtedly seen by many as a necessary part of the process), I include the fatalities as accidental deaths. Estimates vary on how many died during the Middle Passage, but in a recent study, David Ellis and David Richardson suggest that 15% of roughly 10 million who left the African coast died on the way. That would give us at least 1.5 million deaths. The true number is probably much higher.

Dams! Although they are not necessarily known for high-profile disasters and typically don’t involve large numbers of fatalities, accidents involving dams are not uncommon and in fact, their deleterious effects become even more evident when one considers the frequent long-range damaging effects on the nearby environment. The other reason I would include them is that because dam use goes back to antiquity, it’s an example of a technology that has accidentally caused major fatalities across several millennia. In the modern era, major dam failures include one in Johnstown, Pennsylvania in 1889 when over 2,000 people died, one in Madhya Pradesh in British India in 1917 that killed several thousand, one in Italy in 1963 that killed several thousand, and one in Gujarat, India in 1979 that may have killed 5,000 people. But the most horrific disaster was in 1975 in China when close to 200,000 people perished after very intense rainfall led to the collapse of a network of dams in the province of Henan. From what I understand, over 10 million people were rendered homeless by the disaster. My guess is that with the continued adverse effects of climate change, large-scale human migrations, and lack of infrastructural maintenance, we might see more dam failures in the near future, and thus more fatalities.

Alan Marcus

Professor and Head of History at Mississippi State University, whose research focuses on the history of technology, among other things

I would say the technology of exploration — jumping on ships and coming to the new world. Estimates suggest anywhere from 80-95% of the Native American population perished. The most lethal diseases transmitted were measles and smallpox, but typhus was a factor as well.

We think of measles as a really moderate disease, but it could devastate a population without the proper protections. Imagine what’s happening now with covid-19, but with no hospitals, no ventilators, no medicine. You’re on your own.

No one in the old world anticipated or expected these consequences, and it took a long time — around 150 years — for them to notice some cause-and-effect. To a large extent, disease was looked upon as a local phenomenon: it was either something your body had become imbalanced from, or based upon the idea of what they called bad air, or malair — malaria. Bad airs made you sick.

When the Europeans went back to Europe they carried new world diseases with them. When they went to Africa and inaugurated the slave trade, they brought both new world and old world diseases. The carnage was in the tens or hundreds of millions. So it’s not just new world/old world — it’s triangular.

Erik Loomis

Associate Professor, History, The University of Rhode Island, whose research focuses on the labour and environmental history of the United States, among other things

I don’t know if trains have in fact killed more people by accident than automobiles, but it’s worth noting the shocking level of fatalities trains caused in the nineteenth century. Put bluntly, both workers and passengers placed their life at risk by interacting with trains. For many decades, American trains were far more dangerous than those of Europe. The rail historian Mark Aldrich has written about this in great and horrifying detail. Railroad workers died at appalling rates, often crushed between the cars. Derailments killed passengers and workers alike. An utter indifference to accidents by the rail companies, the courts, and politicians stood in stark contrast to Britain, where only a fraction of people died compared to the U.S. Given that the railroads also became the nation’s first coordinated transportation network, people also died frequently after hopping them to ride the rails from place to place and job to job.

Moreover, trains routinely killed bystanders in cities. Tracks ran through heavily populated city centres and the companies did very little maintenance to ensure that people could cross them safely. Ruts between tracks and roads meant that people with carts often got stuck and the trains ran them down, for instance. On top of this, trains added massive amounts of smoke and noise to the cities, seriously lowering the quality of life for residents. The rail industry’s insatiable greed also led its leading capitalists such as Jay Cooke and Jay Gould to sink the economy due to their irresponsible speculation. The Panic of 1873 and Panic of 1893 were major economic depressions that also caused collateral damage in lives from the conditions of poverty and hopelessness. It is worth noting that the major labour actions against railroads in the late nineteenth century, such as the Great Railroad Strike of 1877 and the Pullman Strike in 1894, were community rebellions against the railroads as much as they were traditional labour strikes, with the majority of the protestors not workers on strike.

Railroads became safer in the twentieth century, but the corporate indifference to Americans’ lives has consistently led to unnecessary deaths, including in the meatpacking plants during the covid-19 crisis.

Mar Hicks

Associate Professor, History, Illinois Institute of Technology, and the author of Programmed Inequality: How Britain Discarded Women Technologists and Lost Its Edge in Computing

When we think about technologies that have killed a lot of people by accident, we have to think about technologies that have been around a long time, and whose utility has been so great for industrial expansion that its negatives have been overlooked — or, worse yet, intentionally hidden.

The cotton gin, patented by Eli Whitney in 1794 and in widespread use throughout the US throughout the 19th century, is one such technology. The cotton gin (short for “enGINe”) was a machine that made cleaning and preparing raw cotton much quicker and more efficient — and therefore made the growing of cotton much more profitable.

What the cotton gin also did was to make slavery far more entrenched, through making cotton picking by enslaved people in United States much more profitable. Slavery had not been expanding as rapidly until the invention of the gin encouraged more and more white cotton growers to expand their production. White southerners “imported” more than 80,000 Africans as slaves between 1790 and the ban on “importing” enslaved Africans in 1808. Between the years of 1790 and 1850 the number of enslaved people in the US rose from 700, 000 to more than 3 million through generational enslavement (chattel slavery). By the start of the civil war one third of all Southerners were enslaved people.

This was all in the service of the booming cotton industry that the cotton gin created: the US supplied the vast majority of all the word’s cotton by the mid 19th c. and the production of cotton doubled every decade after 1800. When people say that the U.S. economy was built on the backs of enslaved Black people, they are talking about industries like cotton and all the personal and national wealth created at the expense of enslaved Black people’s lives.

Had it not been for the invention of the cotton gin, it is likely that slavery would’ve been abolished more quickly instead of massively expanding in the way that it did, in a relatively short period of time. The calculation of deaths that includes enslaved Black people who died en route to the US, and enslaved Black people who died or were killed while in the US, already more than qualifies this technology for a high spot on this list — to say nothing of the widespread misery and pain caused to enslaved people, and the generations of their descendants who have been deprived of their full civil rights as a result.

Right now, we are seeing all too clearly how Black people living in the US today lose their lives as a result of this economic and technological history — how white business owners in the South in the 18th and 19th centuries used technology to amplify and extend racism, misery, and death, much in the same way that we see happening with certain technologies today.

So I think this is an important history of technology to keep in mind. Because it shows how technologies are always constructed for and by the contexts in which they come into being. And if that context is racist, they are likely to uphold racism, if what they do is help make the existing economic and social structures stronger and more efficient without caring about existing inequalities. When technologists try to “fix” things with merely technical solutions they ignore the broader context and how those technologies work in it.

This is one reason why it’s so important for STEM practitioners to learn and know history, and why STEM programs at universities do their students, and all of us, a disservice by not having more humanists and historians. Narrowly technical “advances” that don’t understand the broader context can lead to terrible unintended — but not unforeseen — outcomes. And that isn’t real progress at all.