The Kauffman Foundation's Samuel Arbesman on his new book, The Half-Life of Facts.
If you ask most people, they’ll tell you facts are facts. But the reality of the matter, as Samuel Arbesman points out in his brilliant new book, The Half-Life of Facts, is "[f]acts change all the time." To cite just a few of Arbesman’s most compelling examples: We used to think that the earth was the center of the universe, that Pluto was a planet, and that brontosaurus was a real dinosaur. Arbesman is an applied mathematician and network scientist by training, as well as a senior scholar with the Kauffman Foundation and fellow at Harvard University’s Institute for Quantitative Social Science. I talked with Arbesman, who writes a great deal about cities and is a frequent contributor to this site, about the role of place, community, and cities in the way we know, discover, and innovate.
Does place factor into the development of new knowledge and new facts? Why do some places develop an advantage in developing new science? How and why has this changed over time?
Knowledge and ideas have to start somewhere. When a scientist makes a new discovery, or debunks a long-held belief, it often occurs in a particular place. This new information then takes time to diffuse outward, winging its way around the globe according to certain rules and regularities, and in some cases, overcoming (or not) the factual inertia in each of our own heads which prevents us from accepting new knowledge.
But just as knowledge arises somewhere, it also is more likely to occur in certain places. New scientific knowledge is more likely to develop in Boston than Wichita, for example. And, new ideas arise in certain times and places, such as Athens during its Golden Age or Edinburgh during the Scottish Enlightenment.
While we by no means understand everything about why certain places are poised to create new science and knowledge, there seem to be certain factors, such as population size, density and connectivity between individuals, and even the presence of universities. The makeup of the population is also an important factor for why certain places are integral to knowledge. Robert Merton, a sociologist of science, has argued that one reason that Great Britain oriented towards science in the seventeenth century was because a disproportionate number of individuals chose to become scientists, relative to other professions, such as becoming a priest or joining the military. When cities and regions value the types of individuals that catalyze new knowledge creation, it is unsurprising if these places excel in the development of new ideas.
You write about the relationship between knowledge and population growth. How do the two go together?
Simply put, the more people, the more potential for innovation. If you have more people, more good ideas will arise. But it turns out it is much more complicated than that. It does not appear to be simply linear (you add one more person and you have an additional unit of creativity). Adding another person actually seems to allow for additional connections between individuals, which in turn creates the possibility for the diffusion of knowledge and the cross-pollination of ideas.
This is seen most clearly in cities. While we know that the productive cities — those with universities developing scientific innovations, places patenting new ideas, and even those that have the largest number of creative individuals — are those that are in general more populous, the relationship is actually superlinear. Work by a team of scientists led by Luís Bettencourt and Geoffrey West has found that adding an additional person does not increase productivity of a city by a constant amount, it actually increases the productivity of city per person. In other words, it is as if everyone else in the city have themselves become more productive through the addition of each new person. And hopefully, in all this productivity, there will be an increasing churn and growth of knowledge and facts. So, ultimately, larger cities allow for new knowledge.
You have a great example of the diffusion of the printing press being driven largely by social contacts. What does this process mean for cities and the spread of knowledge today?
The printing press technologies developed by Johannes Gutenberg did not diffuse rapidly like a piece of gossip or a virus. Rather the printing press took its time to spread across Europe. But this is somewhat odd. Information should theoretically be able to diffuse rapidly. For example, the Black Death spread at about the speed with which humans could travel in the Middle Ages, allowing it to sweep across Europe and into England in less than a half a decade. On the other hand, it took decades until the technologies developed by Gutenberg first appeared in England after their invention in Mainz.
Clearly, it is not enough to be able to spread information, whether on foot, or nowadays via the Internet. Information is spread via social connections, and by individuals who have the information (especially tacit knowledge) that is necessary for the technology to spread. And in the case of the printing press, it was through a network of Gutenberg’s personal and cultural contacts and apprenticeship. And which places even got it first? The bigger cities, because they contain more social contacts to other locations, allowing for the flow of new ideas.
Cities are incubators of information and allow for the processing of information. And this interaction occurs at a personal level, in a way that is not possible if someone outside this pressure-cooker culture were to read the staid knowledge in textbooks. Cities are where new ideas occur, where they are developed and tested, and then the source from which they can spread.
Figure 6 (appears on page 72 of the book): "Diffusion of the movable-type printing press over time. From Dittmar. "Information Technology and Economic Change: The Impact of the Printing Press." The Quarterly Journal of Economics 126, no. 3 (August 1, 2011): 1133-72, by permission of Oxford University Press."
You argue that “a long tail of expertise—everyday people in large numbers—has a greater chance of solving a problem than do the experts.” How do cities factor into this process?
When it comes to the “long tail of expertise” — a term first used by Alpheus Bingham of InnoCentive — the Internet is great in that it allows for large numbers of individuals, who would otherwise be too diffuse to manage, to come together and use their diverse skills and insights in order to solve a problem. Cities can also act as aggregators of that long tail of expertise. When it comes to problems that can’t be solved by a virtual interaction at a single time-point, but require a sustained effort, and one that must be done face-to-face, you need to find a way to bring lots of people together to allow for a problem to be solved. And that’s what cities can do. Cities aggregate a wide variety of individuals and allow for their productive interaction.
You write: “...the innovative pace of cities is ever quickening. In order to continue growing, cities seem to actually require periodic drastic innovative reboots—from advanced sewage systems to building methods that allow for skyscrapers—that are happening more quickly. For the first time in human history these rapid changes are occurring multiple times in a single generation, and they don’t seem to be slowing down.” I know Atlantic Cities’ readers will want to know more about this.
Unlike biological systems, where growth slows until a certain size is reached (such as adulthood), in the case of cities growth increases its pace. This pace can be so rapid that in order for an urban system to avoid being overwhelmed, it must innovate out of this overwhelming growth, thereby “resetting” the system.
In the early days of cities, growth happened slowly, with these resets happening every few centuries, or even longer. I spoke with medieval scholars to see if one’s intuition that you could be dropped down into large stretches of time in villages in the early Middle Ages, without seeing few differences is correct. And it turns out, despite the advances that did indeed occur during this time period, a villager’s life would be little different when placed anytime within certain long stretches of time.
But as cities have themselves become more populous, growth happens more quickly, and requires more rapid urban developments. Unlike in the cities of centuries ago, if we were to be placed in a city mere decades back or even several years ago, we would notice the difference. What this means is that big cities not only generate fast-changing facts, but almost actually require these shifts in knowledge in order to continue to exist! Which can of course create a virtuous cycle of ever-increasing knowledge.
At the Kauffman Foundation, you've been studying the clustering and concentration of innovation and entrepreneurship in particular locations. What have you been learning about this?
Entrepreneurship and innovation do not appear in a vacuum; they appear in certain places. Geography and clustering play a clear role. We certainly all know this intuitively: whatever the reason, Silicon Valley’s primacy in entrepreneurship and innovation is world-renowned, while we don’t imagine many people go to Anchorage to start a company. But what we’ve been finding is that the traditional narrative is far more complex. Using data from the Inc. 500, we’ve found that many start-ups are created in the South, defying the normal story of the importance of the coasts on the United States. In addition, specific cities that one might not expect end up topping these rankings, such as Salt Lake City or my hometown of Buffalo.
We have also been exploring the mobility of individuals, and how people choose where they locate their company. While there is a great deal of movement, many people end up forming a company in the same region where they went to school, implying that the irrelevance of distance is far from complete. We are now working to better understand how certain places can become hubs for entrepreneurship and innovation, and what factors are correlated with this.
This interview was edited for clarity and length.