We Declared 1,157 Sites Sacred. Then We Watched Them Burn.

It takes a particular kind of optimism to believe that declaring something sacred will prevent people from destroying it. Yet in 1972, UNESCO adopted the World Heritage Convention with precisely this faith: that inscription on a special list would protect humanity's greatest cultural and natural treasures from the depredations of war, development, and neglect. Fifty-four years later, we have 1,157 sites on that list. We also have Palmyra in rubble, Timbuktu's manuscripts burned, Aleppo's Old City shelled into dust, and Sana'a's medieval quarter bombed by Saudi Arabia with American-made weapons. One is tempted to ask what exactly the list protects, besides our ability to say we made one.

The mechanism is elegant in its simplicity. A country nominates a site of "outstanding universal value." Technical advisory bodies—ICOMOS for cultural sites, IUCN for natural ones—conduct evaluations. The World Heritage Committee, composed of 21 member states elected on rotating terms, votes. If inscribed, the site joins the list. UNESCO sends a certificate. There are no peacekeepers, no enforcement provisions, no binding obligations to stop construction, prevent looting, or halt military operations within buffer zones. The protection is reputational. The assumption is that governments, armed groups, and developers will respect the designation because the international community has spoken.

This is not, of course, without precedent. The Geneva Conventions also rely on the goodwill of combatants. The difference is that when the Conventions are violated, we call it a war crime and theoretically prosecute. When a World Heritage site is destroyed, UNESCO issues a statement expressing "grave concern" and sometimes places the site on the List of World Heritage in Danger, which is rather like putting a fire on a watch list after it has already burned down the building.

The Record Since 2011

Let us review what has happened to World Heritage sites in conflict zones over the past fifteen years, a period that conveniently overlaps with the Arab Spring, the Syrian civil war, the rise of the Islamic State, the war in Yemen, and assorted other catastrophes that apparently did not receive the memo about outstanding universal value.

In May 2015, Islamic State forces captured Palmyra, the Roman-era city in central Syria that had been inscribed as a World Heritage site in 1980. Over the following months, they dynamited the Temple of Bel, built in 32 CE; the Temple of Baalshamin, from the second century; and the Arch of Triumph. They beheaded Khaled al-Asaad, the 82-year-old archaeologist who had spent five decades as director of antiquities at Palmyra, in the site's Roman amphitheater. His crime was refusing to reveal the location of artifacts he had hidden to prevent their sale on the black market. UNESCO Director-General Irina Bokova called it "cultural cleansing" and an "intolerable crime against civilization." Islamic State sold the artifacts anyway. The international response involved airstrikes against IS positions, but not to protect the ruins—those were incidental to the broader military campaign. By the time Syrian and Russian forces retook Palmyra in March 2016, approximately 20 percent of the ancient city had been destroyed.

In Mali, jihadist groups linked to Al-Qaeda destroyed fourteen of sixteen mausoleums in Timbuktu in 2012, along with thousands of ancient manuscripts housed in the Ahmed Baba Institute. The mausoleums were Sufi shrines, which the militants considered idolatrous. Ahmad Al-Faqi Al-Mahdi, a member of Ansar Dine, was later convicted by the International Criminal Court for directing attacks against historic monuments—the first ICC conviction solely for cultural destruction. He received nine years. The mausoleums have since been rebuilt with international funding, which is to say they are now replicas of the fifteenth-century structures that were destroyed, which raises interesting questions about what exactly we are protecting. The manuscripts, many of them smuggled out in footlockers and metal trunks by librarians fleeing the city, are authentic. They are also scattered across Mali and neighboring countries, unprotected and undocumented.

The Argument They Haven't Made

UNESCO defends the Convention on the grounds that inscription raises awareness, mobilizes international funding for conservation, and creates legal frameworks for national protection. All of this is true. The World Heritage Fund, financed by mandatory and voluntary contributions from member states, disburses approximately $4 million annually in emergency assistance and technical cooperation. Italy has spent an estimated €3.2 million restoring sites in Syria, including preliminary work to stabilize the Umayyad Mosque in Aleppo. France has trained Malian conservators. Poland has documented damage at Palmyra using photogrammetry. These are not trivial efforts.

But they are efforts that commence after destruction has already occurred. The Convention contains no provisions for preventive deployment, no authority to demand access for emergency stabilization, no sanctions for member states whose militaries target heritage sites. Article 6 of the Convention states that each country "recognizes that such heritage constitutes a world heritage for whose protection it is the duty of the international community as a whole to co-operate." The operative word is "co-operate," which in diplomatic language means "if you feel like it."

There is, in fact, a separate instrument designed for conflict: the 1954 Hague Convention for the Protection of Cultural Property in the Event of Armed Conflict, along with its two protocols. The Second Protocol, adopted in 1999, establishes "enhanced protection" for cultural property of the greatest importance to humanity and criminalizes intentional attacks. As of April 2026, 133 countries are party to the 1954 Convention, but only 85 have ratified the Second Protocol. The United States ratified the Convention in 2009—fifty-five years after it was drafted—but has not ratified the Second Protocol. Russia ratified both, then shelled the Kyiv-Pechersk Lavra and damaged the historic center of Odessa, both on Ukraine's UNESCO tentative list, during its 2022 invasion. Syria is a party to the Convention; this did not prevent its air force from barrel-bombing Aleppo's covered souks.

What Actually Protects Sites

The uncomfortable truth is that heritage sites are protected by the same forces that protect anything else in a conflict zone: the presence or absence of armed groups, the strategic value of the location, and the degree to which attacking the site imposes a cost—military, political, or reputational—on the attacker. Palmyra survived decades of Syrian governance, benign neglect, and mass tourism. It did not survive Islamic State because IS explicitly sought to destroy pre-Islamic heritage as part of its ideological project and faced no immediate military deterrent. When Russian and Syrian forces retook the city, they held a concert in the Roman theater featuring the Mariinsky Orchestra conducted by Valery Gergiev, which was meant as a cultural counter-statement but which also demonstrated that the site's protection was a function of military control, not its presence on a UNESCO list.

The Hague Convention's provisions are somewhat more realistic. Article 4 requires parties to "respect cultural property" by refraining from targeting it and from using it for military purposes. Article 8 allows for the appointment of "Commissioners-General for Cultural Property" to work with occupying powers, though in practice none have been appointed in recent conflicts. The Second Protocol creates a Committee for the Protection of Cultural Property in the Event of Armed Conflict, which can grant enhanced protection and monitor compliance. As of 2026, nineteen sites in twelve countries have received enhanced protection. None are in Syria, Yemen, or any other active war zone, because enhanced protection requires a management plan, a legal framework, and a declaration by the state party—documentation unlikely to be forthcoming from governments currently bombing their own cities.

The Repatriation Question Nobody Wants to Answer

There is a separate, parallel irony in the heritage protection discourse, which is that many of the institutions most vocally committed to preserving cultural property are also the ones refusing to return objects removed during colonial occupation. The British Museum holds approximately 73,000 objects from sub-Saharan Africa, including the Benin Bronzes looted by British forces during the 1897 punitive expedition against the Kingdom of Benin. Greece has been requesting the return of the Parthenon Marbles since 1832. Egypt has been asking for the Rosetta Stone since the 1940s. France holds an estimated 90,000 artifacts from sub-Saharan Africa; in 2021, President Emmanuel Macron signed a law enabling the return of 26 objects to Benin and a ceremonial saber to Senegal, while emphasizing that this did not set a broader precedent. Germany returned 23 Benin Bronzes to Nigeria in December 2022. The British Museum's position remains that it is prohibited by the British Museum Act of 1963 from deaccessioning objects in its permanent collection, a convenient legislative obstacle that Parliament could repeal at any time.

The argument against repatriation typically proceeds along two lines: first, that the objects are safer in European museums than in their countries of origin, which may lack climate-controlled storage or trained conservators; second, that the museums hold these objects in trust for all humanity and that returning them would diminish universal access. Both arguments assume that the populations from whom the objects were taken are incapable of caring for their own heritage and that "universal access" is better served by displaying artifacts in London and Paris than in Lagos and Cairo. One might observe that this is precisely the logic that justified the original removals.

The safety argument is particularly rich coming from institutions that watched Notre-Dame burn in 2019 and the National Museum of Brazil lose 20 million items in a 2018 fire caused by chronic underfunding. The Smithsonian suffered a roof collapse in 2011 that damaged collections. The British Museum admitted in August 2023 that approximately 2,000 objects from its collection had been stolen over several years by a staff member, who sold them on eBay. If "safety" is the standard, the case for keeping artifacts in Western museums is less compelling than generally advertised.

Languages We Are Letting Die

If physical heritage faces destruction by war and looting, intangible heritage faces a slower, quieter obliteration through language death. UNESCO estimates that 40 percent of the world's approximately 7,000 languages are endangered, with one language disappearing every two weeks. When a language dies, it takes with it oral histories, ecological knowledge, music traditions, and entire ways of conceptualizing the world. This is not a metaphor. The Kallawaya healers of Bolivia possess botanical and medical knowledge encoded in their language, which has fewer than 100 fluent speakers remaining. The Ainu language of northern Japan, once spoken by tens of thousands, had ten native speakers as of 2020; it contains specialized vocabulary for salmon runs, snow conditions, and seasonal cycles that have no equivalent in Japanese. The Siletz Dee-ni language of Oregon, spoken by precisely one fluent elder as of 2024, preserves place names and ecological knowledge spanning 1.2 million acres of the Oregon coast. When that speaker dies, the language will survive only in recordings and academic papers.

The causes are structural: economic migration to cities where dominant languages are required for employment; education systems that teach only in national languages; the association of indigenous languages with poverty and backwardness; and the sheer convenience of linguistic assimilation in a globalized economy. There are also specific policies. Indonesia's transmigration program, which relocated Javanese speakers to outer islands, reduced the speaker populations of dozens of minority languages. China's standard Mandarin education policy has accelerated the decline of minority languages including Manchu, which has fewer than twenty native speakers despite being the language of the Qing dynasty. The Russian Federation recognizes 37 official minority languages but provides education funding almost exclusively for Russian-medium schools, which has contributed to the endangerment of languages including Votic (four speakers), Ter Sami (two speakers), and Kerek (extinct as of 2012).

The scale of what is being lost is difficult to grasp. Linguists estimate that 3,000 languages will disappear by 2100 under current trends. That is nearly half of all languages currently spoken. The majority are indigenous languages in the Americas, Australia, the Pacific, and northern Asia—precisely the regions that underwent the most intensive colonization and now face the most aggressive economic integration. The irony is that we have the tools to document and revitalize languages—digital archives, language learning apps, community immersion programs—but almost no political will to fund them at scale. UNESCO's Atlas of the World's Languages in Danger is a website. The Endangered Languages Project, run by a coalition of universities and foundations, operates on a budget that would not fund a single fighter jet.

What We Choose to Protect

There is a pattern in the heritage we mobilize to protect and the heritage we let disappear. Monumental architecture receives emergency funding; oral traditions do not. Museums in capital cities receive climate control systems; rural sites receive brochures. European colonial artifacts remain in European museums while the institutions that could house them in their countries of origin go without funding. Languages spoken by stateless populations disappear while we fund translation software for commercially valuable languages.

This is not an accident of resource allocation. It is a set of choices about whose heritage counts as universal and whose counts as particular, whose past is worth preserving in international law and whose can be allowed to fade. The World Heritage Convention inscribes sites. It does not deploy peacekeepers to defend them, sanction states that bomb them, or fund the legal battles to repatriate the objects removed from them during conquest. The Hague Convention criminalizes the destruction of cultural property but relies on domestic prosecutions that rarely occur and international tribunals that lack enforcement power. Language documentation projects exist, but on budgets measured in thousands of dollars while heritage tourism generates billions.

One might argue that this is simply realism—that international law cannot prevent wars, that economic forces drive language shift, that museums have legitimate conservation concerns. All true. But then we should stop pretending that a designation confers protection, that a convention prevents destruction, or that a list represents anything more than our capacity to catalog what we are willing to let burn. The World Heritage List is not a shield. It is an epitaph we write in advance.

What Enforcement Would Actually Require

If we were serious about protecting heritage in conflict zones, we would need binding protocols with actual consequences. The Hague Convention would need enforcement mechanisms beyond domestic prosecutions—perhaps ICC jurisdiction triggered automatically when heritage sites are attacked, or sanctions against states whose forces target protected property. The World Heritage Convention would need a rapid-response fund in the tens of millions, not the low single-digit millions, and authority to demand access for emergency stabilization. UNESCO would need the power to compel state parties to halt construction or military operations near inscribed sites, with consequences for refusal.

We would also need to address the museums. Repatriation cannot be voluntary and piecemeal; it requires legislation in countries holding colonial-era collections that presumes return unless the holding institution can demonstrate clear legal acquisition and ongoing consent from the country of origin. We would need funding for museum infrastructure, conservation training, and security in countries receiving returned objects—not as charity, but as restitution for the institutional capacity that was destroyed or never built because the objects were removed in the first place.

For languages, we would need funding at the scale of the crisis: billions, not thousands, directed to community-controlled immersion programs, teacher training, curriculum development, and media production in endangered languages. We would need to recognize indigenous land rights and resource access, because language survival is tied to the survival of the communities and ecosystems in which those languages are embedded. None of this will happen, because it would require wealthy countries to cede control, spend serious money, and acknowledge that heritage protection is inseparable from questions of sovereignty, restitution, and power.

So instead we will continue to inscribe sites, issue statements of concern, fund modest restoration projects after the damage is done, and congratulate ourselves on our commitment to humanity's shared heritage. We will keep the Benin Bronzes in London and the Rosetta Stone in the British Museum, arguing that we are holding them in trust for the world. We will document dying languages and upload the recordings to digital archives that the communities themselves may not have reliable internet to access. We will do all of this, and then we will watch the next heritage site burn, the next museum get looted, the next language fall silent. And we will call it tragedy, as if it were weather—something that happens to us, rather than something we chose.