Picturesque but contentious
: Google Maps made this village Chinese, temporarily. India wasn’t pleased.
Photo: Annabelle Breakey
ne fateful day in early August, Google Maps turned Arunachal Pradesh Chinese. It happened without warning. One minute, the mountainous border state adjacent to Tibet was labeled with its usual complement of Indian place-names; the next it was sprinkled with Mandarin characters, like a virtual annex of the People’s Republic.
The error could hardly have been more awkward. Governed by India but claimed by China, Arunachal Pradesh has been a source of rankling dispute between the two nations for decades. Google’s sudden relabeling of the province gave the appearance of a special tip of the hat toward Beijing. Its timing, moreover, was freakishly bad: the press noticed that Google’s servers had started splaying Mandarin place-names all over the state only a few hours before Indian and Chinese negotiating teams sat down for talks in New Delhi to work toward resolving the delicate border issue.
Google rushed to admit its mistake, but not before a round of angry Indian blog posts and news articles had flourished online. Some commentators posited outright conspiracy between Beijing and the search engine. “Google Maps has always been more biased towards China over the Arunachal Pradesh border dispute,” surmised an Indian blogger. Even more ominously, one former member of Parliament told the Times of India, “The Chinese know how to time their statements ahead of a bilateral meeting.”
Google responded in a manner that radiated chilly omnipresence—by posting a statement in the comments section of what appeared to be every single Web site that had discussed the mix-up. “The change was a result of a mistake in our processing of new map data,” Google announced. “We are in the process of reverting the data to its previous state, and expect the change to be visible in the product shortly.”
One mystery remained, however: how did such an error happen in “the product” in the first place? Why did Google have that perfect set of Chinese names lying around, ready to swap in for the Indian ones?
Google remained silent on this point, but a Belgian blogger named Stefan Geens pieced together a compelling answer. Within China, Geens pointed out, the law commands that all maps represent “South Tibet” (aka Arunachal Pradesh) as fully Chinese. And Google Maps maintains servers in China that fall under Chinese law. In fact, Google runs an entirely separate maps site, ditu.google.cn, for Chinese users, which operates within the great Chinese firewall. This isn’t just a one-off concession to the party leaders in Beijing: Google maintains thirty-two different region-specific versions of its Maps tool for different countries around the world that each abide by the respective local laws. Thus on India’s version of Google Maps, for example, all of Kashmir appears as an integral and undisputed part of the country—because Indian law sees it that way. Similarly, “Arunachal Pradesh” is nowhere to be found on ditu.google.cn. What you find instead are all the same Chinese place-names that caused the uproar of Google Maps in August.
“Somehow,” Geens surmised, “data intended for the China map must have ended up in the global map.” Was it all simply the result of an ill-fated drag-and-drop? Whatever exactly may have occurred, the whole episode illustrated the perils of geopolitics in the age of neogeography.
Just five years since the release of Google Maps and Google Earth, the corporation may well be the world’s most important mapmaker. More than 600 million people around the world have downloaded Google Earth. As a testament to ambition, that number alone would be remarkable. But Google is also intent on upending our very notion of what a map is. Rather than produce one definitive map of the world, Google offers multiple interpretations of the earth’s geography. Sometimes, this takes the form of customized maps that cater to the beliefs of one nation or another. More often, though, Google is simply an agnostic cartographer—a peddler of “place browsers” that contain a multitude of views instead of univocal, authoritative, traditional maps. “We work to provide as much discoverable information as possible so that users can make their own judgments about geopolitical disputes,” writes Robert Boorstin, the director of Google’s public policy team.
Ironically, it is that very approach to mapping, one that is indecisive rather than domineering, that has embroiled Google in some of the globe’s hottest geopolitical conflicts. Thanks to the logic of its software and business interests, Google has inadvertently waded into disputes from Israel to Cambodia to Iran. It is said that every map is a political statement. But Google, by trying to subvert that truth, may just be intensifying the politics even more.
he late modern era—the age that brought us industrialization, high colonialism, and mechanized warfare—was a good time for mapmakers. Colonial authorities like the British Royal Engineers sent survey teams to the far reaches of the globe to sustain the European powers’ outward march. Exuding an industrial-age clear-sightedness (and the vanity of the gainfully employed), cartographers consecrated their profession to the ideals of scientific objectivity and rational standardization—even as their work advanced the narrow mercantile and strategic aims of their state employers.
That progressive scientific ethic in mapmaking found its loftiest expression in 1891, when a German geographer by the name of Albrecht Penck proposed to the world’s cartographers that they create a single “International Map of the World,” composed of 2,500 highly standardized individual maps, each representing four degrees of latitude and six degrees of longitude, at a scale of 1 to 1 million. With great international fanfare, the project sailed off the ground in 1913, sputtered through World War I, and then suffered a blow from which it never really recovered when the program’s central office, in Britain, was wrecked in a bombing raid during World War II. The very powers that had conceived of the grand project were now engaged in another enterprise: destroying each other.
The end of the colonial period, hastened by World War II, ushered in a broad crisis in geographical data collection. “The modern era collapsed under its own weight,” says Michael Frank Goodchild, a British American geographer at the University of California, Santa Barbara. “By the 1970s it was apparent that it was no longer going to be sustainable to have a world in which national governments sustained geographic information.”
In some ways, that postcolonial crisis of information is still with us. “The standard U.S. topographical map is now on average thirty-five years out of date,” says Goodchild, who is widely regarded as a founding father of geographic information science. “Modernist government data collection efforts like the census are in decline all over the world.” In certain parts of Africa, old British colonial maps are still the best thing going.
But where government surveyors have ceased to tread, satellites have zoomed in. These have helped to fill the cartographic vacuum, but not everything can be seen from an unblinking eye in space. Satellites, for example, don’t know the names of local landmarks. That’s where another of Google’s capabilities comes in. Over just the past couple of years, Google has been able to build maps of heretofore barely charted cities in Africa, South Asia, and the Pacific Rim—maps that would have brought tears to the eyes of those who struggled for decades to assemble the International Map of the World—by tapping into the same extraordinary resource that has driven so much of the Internet’s expansion: ordinary people. Using a fairly straightforward editing tool called Google Map Maker, volunteers all over the world have been shaping, revising, and detailing maps of their surroundings. “Essentially,” Goodchild says, “they’re replacing the traditional production systems that governments are no longer willing to fund.”
That new user-generated system of production, married to the technology of searchable “virtual globes” like Google Earth, has given rise to what people have begun calling “neogeography.” In the colonial era, the mapmaker’s imperative was to tame the foreign wilderness with names and boundaries—to discipline a profusion of facts and claims into a narrow and authoritative set of data. Now the profusion of facts and claims is a feature, not a bug. With the ability to zoom in on visual fields of higher and higher resolution, a digital map can contain more and more information—various local names for the same landmark, personal annotations, a picture of someone’s dog in a field. “The modern era was an era of the expectation that every feature should have a single name, and a top-down authority would determine that,” says Goodchild. “I think we’re moving past that with digital technology.” With policies that often favor ambiguity, Google maintains centralized control over the most official features on its maps—national borders, bodies of water, and the like—while in the “community layer” of map information, users have an open canvas. Geography has been democratized.
This isn’t to say that Google is a democracy. In its own way, Google may be just as imperious in its approach to knowledge as the nineteenth-century European powers were in their approach to territory. The corporation simply wants to have searchable dominion over as much information as possible—the more plural and local, the better. (More unique search terms mean more revenue streams.) Meanwhile, by filling the information vacuum left behind by the old state powers, Google has also made it inevitable that it will sometimes be confused for them.
t this point, Google’s reach has become so vast that, whenever the corporation becomes involved in a geographical spat, it’s effectively an international incident. Border disputes have become a common vexation. Earlier this year, the government of Cambodia wrote a formal letter of complaint to the corporation—and shared it with the press—because of Google’s depiction of a disputed border with Thailand near an eleventh-century Khmer temple complex in Preah Vihear Province. In the letter, a senior Cambodian official very publicly declared Google’s representation of the border “devoid of truth and reality, and professionally irresponsible, if not pretentious,” not to mention “very wrong and not internationally recognized.” Suddenly the corporation from Mountain View—which introduced its mapping platform in 2005 with the words “We think maps can be useful and fun”—was making headlines as a major party in a remote jungle conflict that has claimed at least seven lives in recent border skirmishes. Google said it would review the matter, but pointed to Tele Atlas, a third-party firm that supplied the data in question, and suggested that Cambodia should direct its complaints there.
Other disputes stem directly from Google’s participatory model of mapmaking. In 2008, the small Israeli city of Kiryat Yam sued the corporation for libel after a Palestinian civilian named Thameen Darby went on Google Earth and tagged the town as the site of an Arab village destroyed by Israelis in 1948. Darby had annotated hundreds of other sites throughout the country, noting the purported locations of former Palestinian settlements that were wiped away during Israel’s formative years. But the officials of Kiryat Yam took issue with Darby’s geography; they claimed the city was built by Holocaust survivors on a barren stretch of dunes, not over the ruins of evicted Palestinians. Darby, they said, had sullied the city’s reputation. Rather than sue Darby, though, they sued Google. The lawsuit, in other words, was an attempt to hold the corporation liable for something an ordinary user had posted to a community layer of Google Earth.
Another major genre of Google-era geopolitical dispute doesn’t involve borders, but simply names. One of the most active online petitions in the world today is an Iranian campaign hosted on the Web site PetitionOnline.com. In light of recent political events, you might guess the campaign would involve some grievance against the Islamic Republic. In fact, the petition’s target is Google, and its goal is to secure the “immediate and unconditional deletion of ‘Arabian Gulf’ from Google Earth.” First posted on February 19, 2008, the online petition against that decision has gathered 1,235,743 signatures—and counting.
Since the 1960s, the disagreement over what to call the body of water that separates Iran from the Arabian Peninsula has been one of geography’s perennial feuds. “Persian Gulf” is the name recognized by the United Nations, while “Arabian Gulf” is a comparatively new moniker whose use came into fashion during the heyday of Arab nationalism. In the eyes of Iranians, the name “Arabian Gulf” signals a frontal assault on their national history. Defending the name “Persian Gulf” has become one of the few truly unifying pastimes in Iranian politics. Factions that bitterly oppose each other will line up together in the name’s defense—or else attack each other for insufficient loyalty to it.
Unfortunately for Iran, sometime in early 2008 Google Earth decided to display the names “Persian Gulf” and “Arabian Gulf” side by side. The ensuing Iranian online petition, harking back to the values of the old nineteenth-century mapmakers, faults Google for being “unscientific” and ignorant of “international standards.” The petition invokes the authority of the United Nations, whose official endorsement of the name “Persian Gulf” is grounded in a small library’s worth of archival maps, literary references, and other historical source material.
Google responded to the debate with a rather novel set of cartographic norms, throwing all historical criteria out the window. Two months after the “Immediate and Unconditional Deletion” petition went live, the corporation posted a statement explaining its protocol for naming bodies of water. The statement, which made no direct mention of the Gulf or the Iranian petition, explained that Google decides what to call various geographical features purely by determining what names are in use today. Google Earth “displays the primary, common, local name(s) given to a body of water by the sovereign nations that border it,” wrote Andrew McLaughlin, Google’s then director of public policy (and now President Obama’s deputy chief technology officer). “If different countries dispute the proper name for a body of water, our policy is to display both names.”
The statement then transitioned into upbeat talk of the democratization of information. “One of the great features of Google Earth is that it enables us to provide significantly greater amounts of information than flat paper maps,” McLaughlin wrote. “It is our fervent hope that different communities will use Google Earth as an open platform to create content that accurately reflects their views.”
Of course, while an open platform for discussion is exactly what Iranians want in a Web site like PetitionOnline, it’s not what they want in an international map. No party to a dispute likes an undecided referee. (And in fact, few parties to these disputes trust Google’s claims of neutrality: much as Indians suspected foul play when the corporation dropped Chinese place-names into Arunachal Pradesh, the sponsors of the Iranian petition simply suspect a conspiracy between Google and the oil-rich Arab states on the other side of the Gulf.)
Unpopular as it may be, such uncertainty has become a central dynamic of life on the Internet. The erosion of traditional authority is followed quickly by anxiety over its absence, from Google to Wikipedia to the lesser-known precincts of PetitionOnline—where millions of people direct their impassioned grievances not to any official arbiter but straight into the ether. What results is an irony. The digital culture that encourages the inclusion of multiple names for a single feature on a map is the same digital culture that has encouraged hundreds of thousands of Iranians to voice their discontent. The very medium incites nationalism, yet also frustrates it.
It all points back to a simple question: What is Google? Is it a repository for all of our mutually exclusive claims, or is it a higher power to which we appeal? It cannot be both, and yet we seem to treat it as both. This tension may only heighten going forward. “In a world where mapmaking is cheap and anyone can do it,” Goodchild says, “you would eventually expect things to become more and more local.” In such a future, either we will reconcile ourselves to the lack of a central arbiter, or the conflicts will be all over the map.