Disinvestment and Decline in Infrastructure Studies

My favorite webpage in the history of the Internet is a Wikipedia entry titled, "Lists of Ghost Towns in the United States" For years, I've used it as a launching point for web surfing in the old, true, beautiful, and good sense, for Lewis Carroll-esque Lee in Wonderland rabbit hole dives. While the lists include the kinds of stereotypical abandoned places you'd expect - like mining towns in the American West - I have always been more interested in less known, more surprising examples, such as farming communities that arose in 18th century New England only to be left behind by the early 19th. (In some cases, the explanation is as simple as "the land sucked for farming.") Taken together, the lists show our capitalist culture's long history of cycles of development and decline. It's a good reminder that economist Joseph Schumpeter's description of the capitalist process as "the gale of creative destruction" is not only a metaphor.

In a recent interview for the Peoples & Things podcast, historian Mary Bridges threw me for a bit of a loop when she argued that one limit of the vast literature on infrastructure today is that it tends to overly focus us on stability, leaving us unable to explain discontinuity and decline. Mary is Ernest May Fellow in History and Policy at the Belfer Center for Science and International Affairs at Harvard University’s Kennedy School of Government. Her book, Dollars and Dominion: US Bankers and the Making of a Superpower, which takes an infrastructural view of banking institutions, is a fascinating examination of how US banks, almost by accident, became a durable part of the global financial system in the first half of the 20th century.

Mary's point about the limits of the infrastructure concept led me to reflect. I believe there are works on infrastructure and maintenance that address decline, senescence, and death, including my book with Andy Russell, The Innovation Delusion. For sure, decline is not a central theme in the literature, however - in part, I think, because many works in it are morally addressed to the need for better maintenance and repair practices.

We'd better have a picture general enough that it can handle all phases of a technological or social system's lifecycle, though. I've been thinking a lot about issues of decline as I work on my current book project, A Good History of Shit Jobs, which includes reflection on what the complicated term "deindustrialization" has and has not meant for workforces, regions, and economies. It has absolutely meant the destruction of communities in places where production evaporated. Thinking through this set of issues, after being spurred by Mary, led me to pull together some themes that I've been chewing on for some years.

A good part of what I write below was inspired by conversations I had at a workshop on the theme of degradation that danah boyd, who recently became a professor of Communication at Cornell University, held at Microsoft Research in May 2024. (You can listen to my conversation with danah on the podcast here. I'm also hoping to have her on to talk about her forthcoming book on the US census soon.) Participants in the workshop wrestled with questions about what causes degradation in technological system - inspired by recent discussions such as Cory Doctorow's concept of enshittification. I was particularly inspired by Lauren Bridges - Assistant Professor of Media Studies at University of Virginia, who was also recently a guest on the podcast - who started painting a picture of cycles of investment and disinvestment in technological systems.

I think this is a good starting place. The trick then becomes describing what causes investment in some times and places, and disinvestment in others. What Mary's challenge draws our attention to is that most of the existing literature so far has focused on two forms of investment: 1. The initial investment that leads to the building of an infrastructural system. These are investments that lead to expansion and growth. We could even read classic books like Robert Caro's book on Robert Moses, The Power Broker, as being about this. 2. The kinds of ongoing RE-investments of maintenance and repair work that are necessary for keeping infrastructures going. Which is not to say human groups are always very good about making such re-investments; often they are lousy at it, a major theme of the maintenance literature, including my own work.

There is a funny aspect of our terms, interest and investment: Both terms have both psychological and financial meanings. So, we can talk about becoming interested in something, or finding someone interesting (they attract our interest), but we also talk about financial interests, interest payments, conflicts of interest, and the rise of interest groups made up of individuals and organizations who share interests or stakes. Similarly, we refer to individuals becoming cognitively, psychically, or libidinally invested in something, but we also speak about investments in the sense of putting money into something with the expectation of returns. Often this kind of polysemy, a profusion of meanings, can get us into trouble, if we are trying to think clearly. But I think, in this case, the dual psychological and financial senses of our everyday language is onto something, and it can help us think about infrastructural growth, stability, and decline.

When groups become invested in something, it can become a part of their routines, practices, and habits, and in doing so, gains a kind of heft and obduracy. What Mary found in her own work is that the bankers she was studying were surprised that the local banks they'd built in foreign countries continued to do so well during the Great Depression. For sure, some of them closed down, but a surprising number survived. There's a kind of beauty in this finding - the kind of beauty that goes like this: initially, it surprises us, but then after more reflection, it seems perfectly natural. The banks had become a part of the way people in these places - especially elites, no doubt - got things done.

So Mary's research uncovered a surprising form of continuity. But we have now reached the point in the story where she - rightly, I think - argued we have a problem. We have lots of stories of how technological systems and infrastructures come into being, and we have increasing numbers of accounts of their maintenance and repair. But what about how these things fall into decline? But what about disinvestment? What kind of changes in the world lead people to abandon ways of doing things?

For sure, there are the kinds of cataclysmic events that lead to the abandonment of places - wars, disasters, pandemics, famine, volcanic eruptions, and so on. But what is striking is how few stories of decline fit this mode. As a theoretical starting point, we might start by turning to economist Joseph Schumpeter, the so-called "father of innovation studies." There are many things to criticize about Schumpeter's evolutionary picture of economic change, including the way it overemphasizes the role of entrepreneurs. But his picture of the five types of innovation can be helpful for thinking about declined. The five types being these: 1. A new good, 2. A new method of production, 3. A new market, 4. A new supply of raw materials, and 5. A new organizational form. Because the flip side of these forms of novelty are all the ways of doing things that are left behind.

To make things more concrete, let's turn back to ghost towns and what causes them. For this, we are lucky to have a very helpful resource in the book, Lost Communities of Virginia by Terri Fisher and Kirsten Sparenborg. The book divides the 30 declining communities it examines into these seven categories: Gathering Places, Farming Communities, Cultural Enclaves, Resort Communities, Transportation Hubs, Resource Extraction Towns, and Company Towns.

In these categories, we see the shifting sands of American industries and business. For example, in the 19th century, Virginia was full of mineral spring resorts, promising healing waters, including two long abandoned examples just down the road from where I live. But by the early 20th century, most of them were gone, as consumer tastes and health beliefs changed.

With mining communities, sometimes the story was as simple as, for all intents, purposes, and technical constraints, the mineral ran out. But if we read between the lines in the communities covered in the Farming Communities, Transportation Hubs, Resource Extraction Towns, and Company Towns sections of the books, we see that, more often, other ways of doing things became more attractive, that other companies in other parts of the nation or globe had developed process innovations that undercut the viability of the activities that had given rise to the local community.

The community of Pocahontas in Tazewell County, Virginia, for example, grew up around coal mining. Eventually it developed an entertainment district around this industry, including an opera house and forty-five saloons. Its population peaked at 2789 people around 1900. But already by the 1920s, its way of doing things was unraveling; its population, declining. As Fisher and Sparenborg write,

The population began a slow decline as coal demand decreased after [the First World War] and people became more mobile. Automobiles were cheaper, roads were better, and miners were able to move in search of the best work situations . . . The coke ovens cooled as it became easier and more profitable to produce the coke at iron furnaces. Little additional construction occurred in the town after 1920. (210)

By the time Lost Communities of Virginia was published in 2011, the population of Pocahontas was 419. In 2024, it was 262.

Technological change plays other, perhaps more surprising, roles in how other communities declined, too. An interesting case is the town of Boydton in Mecklenberg County, Virginia. Boydton was the county seat, and people from around the county would come on court days to handle business, and also apparently to watch court cases as a form of entertainment. As one old timer told Fisher and Sparenborg, "We used to go to court to hear the trials. We all had our favorite lawyers, and we'd go and root for them." (24)

When people traveled to court on foot or by horse, they often stayed the night and needed to be fed and entertained. As a result, hotels, restaurants, bars, and other such venues flourished. But as the authors note, "The town's decline began with a slow loss of businesses around the courthouse. Improved transportation meant that people no longer needed to spend the night when they came for court and could travel farther for goods and services."

Indeed, automobility plays a significant role in stories of decline in communities where I live. These little towns almost all feature small downtowns that are now completely shuttered. Often the story is as simple as residents began using cars to do their shopping at larger towns nearby that had bigger, better supplied stores. To give another example, the Floyd County Historical Museum in Floyd, Virginia, near here, sits in a former hospital, and its director told me the hospital declined and eventually closed when town residents began using cars to access better, more modern health services in the nearby, larger city of Roanoke, Virginia.

To conclude, I think Mary Bridges is right that infrastructure studies and allied fields haven't thought enough about decline, and I thank her for pushing me on this topic. No doubt, we could use a lot more case studies of degradation and things coming to an end. My hunch is that studies would find that, more than existing things becoming undesirable or even simply disasters bringing places to an end, the stories would feature new desires, new ways of doing things, that led people to change their interests and invest in new ways of life, upending older, once vital practices.