The power of the written word is hard to overestimate. The printing press is usually blamed for the European wars of religion, with the resulting multiplication in ways of life ranging from Calvin’s Geneva to Puritan Massachusetts and the Dutch experiment in toleration. Later, the Republic of Letters would foster a community of scholarship that transcended national borders and contributed to the rise of modern science. With the mass adoption of the Internet in the 1990s, a whole universe of possibilities seemed to open up. Why force employees to commute to offices in expensive city-centers, when they can do their work online and communicate with one another instantaneously?
But the Internet has not resulted in the “death of distance,” to use Frances Cairncross’ elegant phrase. As urban economist Ed Glaeser has put it, paradoxically, “metropolis—proximity has become ever more valuable as the cost of connecting across long distances has fallen.” He adds that “Americans who live in metropolitan areas with more than a million residents are, on average, more than 50 percent more productive than Americans who live in smaller metropolitan areas.” The productive power of density has, if anything, increased. What could explain this?
Glaeser believes that physical proximity is conducive to generating new ideas and fresh innovations. Using New York as an example of a general phenomena, he writes:
Just as globalization killed off New York’s advantages as a manufacturing hub, it increased the city’s edge in producing ideas. While there isn’t much sewing left in New York, there are still plenty of Calvin Kleins and Donna Karans, producing designs that will often be made on the other side of the planet. Honda may have brought heartache to Detroit’s Big Three, but managing the international flow of finance has earned vast sums for New York’s bankers. A more connected world has brought huge returns to the idea-producing entrepreneurs who can now scour the earth in search of profits.
New York reinvented itself during the bleak years of the 1970s when a cluster of financial innovators learned from each other and produced a chain of interconnected ideas. Academic knowledge about trading off risk and return made it easier to evaluate and sell riskier assets, like Michael Milken’s high-yield (junk) bonds, which made it possible for Henry Kravis to use those bonds to get value out of underperforming companies through leveraged buyouts. Many of the biggest innovators acquired their knowledge not through formal training but by being close to the action, like mortgage-backed security magnate Lewis Ranieri of Liar’s Poker fame, who started in the Salomon Brothers mailroom. Today, 40 percent of Manhattan’s payroll is in the financial services industry, the bulwark of a dense and still-thriving city.
A crucial question is why proximity would have anything to do with idea creation. Wither the Republic of Letters? Economist columnist Ryan Avent takes a stab at an explanation:
Density also facilitates idea creation. Better communication and the development of social networks strengthen ties between people located near to each other. Researchers can continue conversations begun at the lab, swap data more easily than before, edit each other’s spreadsheets, plan meetings more quickly and easily, and generally face fewer of the frustrations that used to make cooperation slower and more onerous. Across professional networks, new technologies make it easier for people to sort communications – to filter out the noise of the crowd and focus on the idea-rich signal. If intellectual conversations are the fuel for innovation, then better IT has supported the process of idea generation.
But there’s little here that could not in principle be done online. Social networks develop online all the time. Serendipitous conversation happens in the many online contexts that facilitate it, from public social media like Twitter or a Facebook Page, to private social media, to forums and Slack channels and text messages. Where swapping data, editing spreadsheets, and planning meetings are concerned, it’s simply not clear what about being there in person speeds any of it up.
I don’t mean to be excessively critical of Avent or Glaeser; the premise that “idea generation” and innovation occur more readily when people are physically in the same room than apart seems to be true. Taking this for granted is sufficient to develop their lines of argument, which are focused elsewhere.
Let’s pick up where they left off at this point. Why would the mass adoption of Internet technologies increase the value of population density rather than decreasing it? To answer this question, we will leave the realm of economic analysis and look instead to language.
The resources to do things with languages
The philosopher Alasdair MacIntyre rejected the notion of a universal reason which could stand above all frameworks and judge their merits. Instead, he argues that each framework carries its own “resources” which are available to resolve the problems internal to it. He writes:
[W]hen, in spite of systematic enquiry, issues and problems that are of crucial importance to some tradition remain unresolved and unsolved that a question arises about it, namely, just why it is that progress in this area is no longer being made. Is it perhaps because that tradition lacks the resources to address those issues and solve those problems and is unable to acquire them so long as it remains faithful to its own standard and presuppositions?
Is it perhaps that constraints imposed by those standards and deriving from those presuppositions themselves prevent the formulation or reformulation of those issues and problems so that they can be adequately addressed and solved? And, if the answer to those two questions is ‘Yes’, is it perhaps the case that it is only from the standpoint of some rival tradition that this predicament can be understood and from the resources of that same rival tradition that the means of overcoming this predicament can be found?
He went on to develop this position in greater detail in a later work.
This notion of resources can be misleading if we think in terms of money. Think instead of human resources—a company can only solve the problems its employees have the skills to address. A law firm is unlikely to create the next Facebook or iPhone competitor. The number of employees at the firm who have a given set of skills, along with what the range of skills is at the company overall, also constrains the sorts of problems the company can tackle.
Similarly, the difference between a car factory and a computer factory comes down largely to the equipment used. It may be that the robot arms used on car manufacturing could be repurposed in some way to take part in manufacturing computers instead. However, the computer factory gains greater resources for tackling the specific task of manufacturing computers if it uses extremely specialized equipment for that task. Of course, this comes at the cost of having fewer resources to accomplish other tasks—but in the case of the computer factory, that isn’t much of a price to pay.
For MacIntyre, traditions of thought provide resources for refining existing theories and addressing the problems with those theories. Consider the retrograde motion of planets from the vantage point of the Earth; heliocentric theories can explain this in terms of the Earth passing another planet, such as Mars. This motion bedeviled the geocentric Ptolemaic model for centuries, however, as a framework which presumes a constant motion around the Earth lacks the resources to explain these occasional about-faces. And as long as cosmologists “remained faithful” to the “presuppositions” of the geocentric model, this movement simply could not be explained.
MacIntyre’s notion of a tradition of thought is itself limited by its focus on theories and philosophical traditions specifically. It is further limited by his presupposition that traditions are radically incommensurable; that is, a judgment rendered from within one tradition will appear irrelevant or even incoherent from within another one. Analyzing the world through this lens provides the resources to deliver some valuable insights, as MacIntyre’s body of work demonstrates. However, it provides a paucity of resources to address all but a narrow range of highly intellectual questions.
The focus of our analysis should not be on traditions but on languages. Theories are relatively regimented languages for talking about a specific subject matter. Traditions of thought are relatively less regimented, thus relatively less specialized in their resources, but provide more resources to address a broader array problems.
Now, consider a modern corporation. “Corporate language” is practically a cliché; press releases and corporate memos everywhere are full of this stiff hybrid of legal language and buzzwords. The headline of a typical P&G presser exemplifies this all-too-familiar flavor: “P&G Reiterates Significant Transformation is Delivering Shareholder Value.” Or “New Coca-Cola Sustainability Report Uses Infographics to Highlight Continued Progress Toward Company Goals.” Sustainability and infographics are invoked to seem relevant, but the goal is the dry and uninspired “continued progress towards company goals.” Anyone who has worked in a mid-sized company or larger has likely encountered this sort of language.
So the proposition that corporations develop their own languages should not seem radical. What may come as a surprise is that this language, like all language, rests on a crucial physical component. As explained in a previous piece, Charles Taylor calls this physical component the “enacted” dimension, in a three level hierarchy which puts this dimension at the bottom, includes a naming dimension in the middle, and highly abstract descriptions at the top. Language in the world is a complex mix of components across this spectrum. “Body language” is close to Taylor’s enacted dimension, but underspecified. For Taylor, such physical expression has strong, normative evaluations baked in, even before we attempt to articulate them at the higher levels of language.
Developing a shared language at the level of enacted practice turns out to be very important for a corporation’s productivity and creativity. This is not surprising if your starting point is Taylor’s theory, which takes the enacted level as foundational to all of language. It is slightly surprising if your starting point is the Republic of Letters, which demonstrated the power of language-games played at a distance and at the highest level of abstraction.
But perhaps not. Science, after all, is a practice. Perhaps the Republic of Letters was less a lot of disparate individuals collaborating primarily through letter writing, and more a set of local hubs of scholarship networked by letter writing.
Regardless, I am here proposing that corporations obtain the resources for engaging in their business activities in no small part through the development of languages in general and enacted practices in particular. The enacted level is precisely the one that can only develop through face to face, in-person interaction. By observing and participating in existing practices and language conventions, newcomers can make these their own. Rather than pure assimilation, however, newcomers bring something to the table. Hans-Georg Gadamer described understanding as an event, something that happens rather than something we choose to do, and an event which involves mutual transformation. The newcomer brings resources into the language they learn, but also gains the resources of the language by learning it.
When we spoke of “resources” in discussing theoretical frameworks, we were talking primarily about “standards and presuppositions,” as MacIntyre put it. In the case of, say, a corporate CEO, those enter into it as well, but so does her entire orientation. As Daniel Klein has elegantly put it, entrepreneurship is not a matter of having information or not having it, as if knowledge were like a physical object with set units that one can obtain. Instead, entrepreneurship is dictated by interpretation, by a novel way of looking at the same situation as everyone else and seeing an opportunity where no one else does. Our orientation, the way we approach the world, provides us with the resources for pursuing some opportunities, while also biasing us against even seeing some opportunities as opportunities. In short, all the resources for innovative action that we have individually, and the constraints on those resources, are embodied in our orientation.
And that orientation is developed through participation in the various languages that we encounter on a daily basis. The language developed within a corporation, therefore, orients the employees in how they are towards one another and towards clients, at least while they are on the job. And in guiding employees’ orientation, the corporation’s language structures the resources that are available to the company as a whole.
From corporate culture to city culture
If the physically enacted level of language is central to developing the resources for innovation, and this level of language is highly dependent on physical proximity, this goes a long way towards answering our question about population density. If languages can develop in corporations, it’s no stretch to imagine they can do so in cities. We are already used to speaking of “a Boston accent” at the level of talking, and people commonly describe a certain feel to different cities (and communities of all sorts).
Let us posit that some cities develop languages which provide greater resources for innovation than others. What is more, a virtuous cycle can kick in once this gets off the ground: cities that become known for innovation will draw innovative people, who will contribute resources to the language when they participate in it for their own sakes. And to bring economics back into it, density creates service-sector opportunities that would not otherwise exist. As Ryan Avent puts it:
Suppose that 1 person in 100 has developed a taste for Vietnamese cuisine, and suppose that a Vietnamese restaurant needs a customer base of 1,000 people to operate profitably. In a city of 10,000 people, there wouldn’t be enough people to support a Vietnamese restaurant. The only restaurants that could operate profitably would be those appealing to considerably more than 1 in 100 people — those offering less daring culinary fare. In a city of 10,000 people, there is little room for specialization, and less for experimentation.
A city of 1,000,000 people, by contrast, can support multiple Vietnamese restaurants. Not only will this larger city enjoy a cuisine that might not be available in less populous places, but its ability to support multiple producers of this specialty cuisine improves the price and quality of the Vietnamese food that’s served in them. The presence of multiple Vietnamese restaurants arouses the interest of sellers of the fresh ingredients used in Vietnamese cooking, who then invest in distribution of those products in the larger city. (…)The larger market facilitates competition, which boosts quality and reduces prices. This is good for consumers. But competition also means better service from suppliers and growth in the consumer market, which is good for the restaurants. This is a stronger, more productive, and higher quality microeconomy than one might find in the city of 100,000, where only one Vietnamese restaurant stays open, or the city of 10,000, where there is none at all.
Avent further argues that thick local labor markets reduce the risks faced by entrepreneurs, by allowing them to reasonably expect employment should their ventures fail. Thus once highly profitable and innovative industries make their home in a dense city environment, other economic effects will kick in to make that city even more desirable a place to work and start a business.
Finally, it should be noticed that a large percentage of the populations of innovative cities in America are foreign born. Perhaps innovative cities have developed languages that are relatively stronger at drawing in resources from highly diverse backgrounds. There are probably trade-offs involved here, of the sort discussed in comparing the car manufacturing equipment to the computer manufacturing equipment. Perhaps the ability to draw in and from a highly diverse population results in a language that isn’t as good at facilitating the sort of thick, small community interactions that cities are often accused of having less of.
Speculation aside, it is clear that densely populated cities are more innovative and more inclusive than the alternative. There is good reason to believe that language, so central to the human experience in general, may provide us with an explanation for this.
Featured image is View of Shinjuku skyscrapers and Mount Fuji as seen from the Bunkyo Civic Center in Tokyo, by Morio