Seeing Like a Movie Mogul

In his lead essay, James Scott writes about the efforts of German foresters to create “a redesigned forest that was easier to count, manipulate, measure, and assess,” and throughout Seeing Like a State, he uses this as a metaphor for various state projects to reshape the complex, messy world to make it more susceptible to centralized measurement, control, and expropriation. Scott describes how states made modern statecraft possible by forcing unruly peasants to conform to state-defined names, weights, measures, and especially property rules, so that they could efficiently be taxed, regulated, and conscripted. I’m struck by the parallels between this process and the efforts of today’s incumbent media and software firms to domesticate the rapidly changing world of high technology. The big difference is that this time, the centralizers are on the defensive.

The great media technologies of the 20th century, television, radio, and movies, were inherently centralizing. They had high fixed costs that could only be paid by large, hierarchical institutions, and the existence of these costs allowed the institutions to shape the opinions of millions of people. Similarly, the development of increasingly sophisticated, but very expensive, computing devices starting in the late nineteenth century aided centralization by enabling large institutions to efficiently manage ever-larger amounts of information.

The Internet and the microprocessor are reversing these trends, with profound social consequences. For most of the twentieth century, only a tiny minority of people had access to the means to distribute creative works to wide audiences. Now, anyone can do it for next to nothing, and the former gatekeepers are furious. Using misleading property rights rhetoric, they have lobbied for, and gotten, new legal privileges that make their respective markets more amenable to centralized control.

Consider Hollywood’s long-running conflict with the consumer electronics industry. In the early 1980s, Hollywood came one Supreme Court vote shy of outlawing the VCR, which it predicted (incorrectly, as it turned out) would devastate the movie industry. In 1998, the movie industry lobbied for the passage of the Digital Millennium Copyright Act, which gives copyright holders a legal veto over the design of media devices capable of playing their content.

Hollywood has used its authority under the DMCA to force consumer electronics manufacturers to comply with “digital rights management” schemes that are a digital embodiment of the kind of legibility-through-simplification project Scott describes in Seeing Like a State. Scott argues that a single-minded focus on timber yields in state forests tended to short-change other important values provided by the forest, such as kindling, wildlife, herbs, and scenery. Incumbent media firms have a similarly narrow view of their customers. They seek to reduce the infinite variety of ways customers can interact with creative works down to a handful of standardized, usually passive, activities that can be easily monitored, counted, and paid for.

Manufacturers wishing to produce devices capable of playing Hollywood content are required to sign a licensing agreement dictating in great detail how the devices must behave. In an effort to prevent unauthorized copying, DRM schemes attempt to enumerate all of the operations consumers are allowed to perform, and require devices to perform those—and only those—functions. This is why, for example, most DVD players don’t allow you to fast-forward through commercials; this behavior conflicts with Hollywood’s business model, and so Hollywood has compelled consumer electronics manufacturers to disable that feature when commercials are playing.

DVD players work fine for consumers who want to use them as Hollywood intended, and they make it somewhat less convenient to make illicit copies of Hollywood movies than it might otherwise be. But this achievement has come at the price of dramatically restricting the functionality of home video devices. For example, for over a decade consumers have enjoyed the freedom to “rip” their music CDs to their computers and portable devices, making possible the iPod revolution. Similar software has long been available for DVDs, but such software is illegal in the United States. And hardware manufacturers who offer products with that capability, such as “video jukebox” set-top boxes, face lawsuits under the DMCA.

A similar conflict is under way in the software industry. Software has traditionally been afforded copyright protection, which regulates literal copying of software code, but not patent protection, which covers broader classes of “inventions.” But beginning in the late 1980s, a series of court decisions gave the green light to patents on software. This development was initially greeted with alarm in the software industry. Patents, unlike copyrights, are susceptible to inadvertent infringement, in which one company accidentally develops software with functionality claimed by another firm’s patent. Bill Gates clearly understood the danger. In 1991, when Microsoft was still a relatively small company, he wrote a memo to his executives warning that if software patents are legalized, “some large company will patent some obvious thing” and use the patent to “take as much of our profits as they want.”

In Seeing like a State, Scott describes how colonial officials in India, Vietnam, and elsewhere sought to expropriate the natives by imposing state-defined property boundaries. Over the last two decades, something similar has been happening in the software industry. Firms that agreed to participate in the patent system were recognized as “owners” of broad and arbitrarily defined “technologies,” and could use that ownership to extract rents from those who did not participate. Although they were initially skeptical, large firms discovered that patents made their tumultuous industry more legible and predictable, by transforming a rapidly changing technological ecosystem into series of discrete, bureaucratically defined “technologies” that can be bought, sold, licensed, and put on balance sheets. The fact that these “technologies” have only the most tenuous relationship to what the engineers were doing was of little concern as long as the patents continued generating licensing fees. After all, colonial officials didn’t much care if the property lines on their maps reflected actual or historical cultivation patterns so long as someone was paying the required taxes.

So software firms began building patent portfolios. As Gates and other skeptics predicted, software patents have become so broad and numerous that it’s almost impossible to build any large software product without infringing some. Today, most software companies don’t even try to avoid infringement. Instead, they simply assume they will infringe and stockpile patents of their own to deter anyone from suing them.

By the dawn of the twenty-first century, Microsoft had built a substantial patent portfolio of its own. Now a “large company” itself, it was concerned free software products were cannibalizing the market for some of Microsoft’s own proprietary products. And so Microsoft began threatening vendors who offered operating systems based on the free Linux kernel with patent litigation. Although Microsoft has declined to produce a specific list of patents being infringed, the sheer size of Microsoft’s patent portfolio and the likelihood of inadvertent infringement makes the threat credible. And, of course, Microsoft now lobbies in favor of patents on software.

Proprietary software firms are, understandably, focused on maximizing the number of copies they sell. But the licensing terms required to ensure that each user pays for his or her copy the has a side effect of undermining three of what Free Software Foundation founder Richard Stallman calls the four software freedoms: the freedom to use, study, share, and modify software. Proprietary software offers customers the first freedom—use—for a fee, but the other three freedoms are typically not available at any price, because they conflict with the vendor’s business model. Ordinary consumers may not care about these other freedoms, since they lack the knowledge and skills to take advantage of them. But many computer programmers care about them quite a bit, because they make the software we use much more useful. It allows us to fix bugs, add new features, and diagnose compatibility problems. And as users of Firefox can appreciate, preserving these freedoms for geeks often has spillover benefits for the general public.

Once again, a myopic focus on the cash-for-copies business model has led to the creation of technology that’s dramatically less useful than it could be. Just as German foresters ignored uses of the forest that could not produce revenue for the crown, so software firms’ focus on selling copies of software has led them to ignore uses of software that cannot be easily “monetized.” Some of the most sophisticated users of software have become increasingly dissatisfied with this arrangement and have begun building their own alternative base of software, called free software, that is not so limited. You might say that free software is the natural forest to proprietary software’s “production” forest.

In both of these disputes, incumbents have used the rhetoric of property rights to justify their efforts to seize control over wealth they did not create. Hollywood didn’t invent the Blu-Ray player, flat-screen TVs, or other innovations, but under the banner of property rights they have demanded, and gotten, a veto over the evolution of these technologies. Similarly, patent litigation in the software industry is rarely about actual copying of a competitor’s code. It typically involves transferring wealth from firms that produce innovative products to firms that are adept at navigating the patent system. Framing these controversies as disputes over property rights has allowed Hollywood and the patent bar, respectively, to claim the high ground for what might otherwise be perceived as simple rent-seeking.

Because libertarians reflexively (and correctly) favor strong enforcement of property rights, we need to be careful about too credulously accepting the “property rights” frame for proposals to create or expand legal privileges. Such arguments can be found in a wide variety of fields, including gene patents, the recording industry, and spectrum policy. Clear and predictable property rules are a tremendous engine of economic growth and individual liberty. But Seeing Like a State reminds us that the creation of new property rights can sometimes be a process of expropriation, with the state inventing new rights to transfer wealth to parties with political power.

Reasonable people can disagree about whether the new property rights whose creation Scott describes in Seeing Like a State had positive consequences in the long run, but it’s hard to deny that some of the short-run consequences were deeply illiberal, transferring wealth from ordinary peasants to those who had the closest ties to the state. When large firms deploy the rhetoric of property rights in defense of creating new legal privileges for themselves, libertarians especially need to employ an appropriate degree of skepticism.

Also from this issue

Lead Essay

  • In his lead essay, James C. Scott reviews some of the key concepts from his seminal book Seeing Like a State. For a state to exercise its power across a large population, it must simplify, codify, and and regularize local practices. This process of flattening, or of making local practice “legible,” is not without costs. In the past, states have quite literally missed the forest — with many different valuable products, including food, shelter, medicines, and clothing — for the trees, or timber, that they contain. And that is not the least of states’ errors in this regard; even in the twentieth century, high modern building practices and management techniques have neglected local variation and local knowledge, often to the detriment of state and non-state actors alike. These faults are regular, predictable, and worthy of further study. Provocatively, Scott closes his essay with a warning: Large actors in a market will often find themselves seeing like a state, too.

Response Essays

  • Donald J. Boudreaux considers the upside of widely legible information. While local knowledge and local resistance strategies may be lost, the ability of entrepreneurs to collaborate with one another over long distances expands. Ideas translate more easily and can more easily encounter other ideas, leading to innovation. Although some libertarians may find it uncomfortable, the rise of the modern state with its standardization procedures may have sped the process along, helping the Industrial Revolution to take place. Boudreaux cites the Social Security number as a threatening-but-useful example of state-based legibility co-opted for commercial purposes.

  • Timothy B. Lee notes that many aspects of information technology policy are deeply implicated in the process of “seeing like a state.” Despite their rhetoric of private property, patents, copyrights, and other forms of intellectual property can act largely as transfers of wealth from ordinary people (be they peasants or consumers of digital media) to those who are most closely aligned to the state. This is a deeply illiberal result and one that libertarians should be especially wary of.

  • Brad DeLong sees both strengths and weaknesses in the state’s ability to survey civil society. With the help of an extended example from late Roman times, he argues that states can offer no protection against local disorder when they cannot see the localities and peoples they propose to protect.

    Yet a legible civil society is also prey to the state’s own “fits of ideological terror, or even clumsy thumb-fingeredness,” he argues, suggesting that legibility may be orthogonal to liberty. Ultimately, a state that fosters a robust civil society, while contenting itself with simply watching its growth, may be best of all, even if it is unlikely.