A city is not lost in the way a wallet is lost. Cities do not slip between the couch cushions. Every one of the "lost cities" in the standard archaeological inventory was known, at the time of its supposed loss, to someone. Pompeii was known to the residents of the surrounding area for a thousand years after the eruption of Vesuvius. Machu Picchu was known to the Quechua farmers who had been cultivating its terraces continuously since the sixteenth century. What "lost" means, in archaeological usage, is that the city was no longer known to the people who would later want to know about it. That is a different category of thing.
What follows are five standard examples, with attention to the history of their disappearance from, and reintroduction to, the global academic record. I am interested particularly in the rediscoveries, because the rediscoverers are, in almost every case, more interesting than the popular history has made them.
1. Pompeii (buried AD 79, excavated 1748)
The cities of Pompeii and Herculaneum were buried under four to six meters of volcanic ash and pumice during the eruption of Mount Vesuvius on August 24, AD 79. Approximately two thousand people died in Pompeii alone. The disaster was extensively documented by Pliny the Younger, whose uncle Pliny the Elder died in the eruption.
What happened subsequently is odd. The surviving population rebuilt settlements in the region, but the specific locations of the two cities were gradually forgotten. The names persisted in the literary record (Statius, Martial, and others continue to reference them), but the sites themselves, under five meters of ash, were no longer places you could go. By the medieval period, farmers cultivating the hillside above Pompeii occasionally turned up Roman artifacts; a thirteenth-century document mentions the area as la città (the city), suggesting the locals knew that something was under there, but the identification with Pompeii specifically was not made.
Herculaneum was accidentally rediscovered in 1709, when workers digging a well for a nearby villa broke through into the top of its theater. Pompeii was located in 1748 by a Spanish military engineer named Rocque Joaquín de Alcubierre, under the patronage of King Charles III of Spain, who was then also King of Naples. The early excavations were extraction operations rather than archaeology in any modern sense: the goal was to remove decorated objects for the royal collection. The methods were destructive. What has been gained, over subsequent centuries of more careful work, is one of the most complete pictures of Roman urban daily life that exists.
2. Machu Picchu (built c. 1450, rediscovered 1911)
The Inca site at Machu Picchu was built during the reign of the Inca Sapa Pachacuti, probably as a royal estate, around 1450. It was occupied for approximately a hundred years, then abandoned around the time of the Spanish conquest in 1532. The reasons for abandonment are disputed; the leading theories involve the disintegration of Inca administrative systems during the conquest, possibly accelerated by a smallpox outbreak. The site was never discovered or documented by the Spanish colonial authorities.
It was, however, never lost to the local Quechua-speaking population, who continued to cultivate its terraces for centuries afterward. When the Yale historian Hiram Bingham arrived at the region in July 1911, guided by locals who had no idea why a foreigner would find this interesting, the first person he met at the site was a Quechua boy named Pablito Alvarez, who had been living on the site with his family. Bingham paid Alvarez one sol for the tour. The famous "rediscovery" photograph of Bingham arriving at Machu Picchu shows, in the foreground, the farm buildings of the Alvarez family.
Bingham's achievement was not finding the site. The site had been found repeatedly by local farmers, by German prospector Augusto Berns (who visited in 1867 and attempted to loot it), and by Peruvian cartographer Agustín Lizárraga (who carved his name on a wall in 1902). What Bingham did was publish, in 1913 in National Geographic with photographs, and thus place the site in the global academic record from which it had been absent. The distinction matters. Bingham's role was that of a communicator, not a discoverer. His books tended to conflate the two.
3. Troy (fell c. 1200 BC, excavated 1870)
The Trojan War of Homer's Iliad was, through most of the post-classical period in Europe, treated as a literary fiction. A subset of scholars argued that the war had a historical basis, but the question was unsettled and the location of the supposed city was unknown. In 1870, a German businessman and amateur archaeologist named Heinrich Schliemann, who had made a fortune in the Crimean War, began excavations at a hill called Hisarlık in northwestern Turkey, which had been identified as a possible Troy candidate by an earlier British amateur, Frank Calvert. Calvert owned part of the land. Schliemann bought it.
Schliemann's excavations, conducted with brutal speed over the 1870s, did in fact uncover a substantial Bronze Age city, which he promptly declared to be Homer's Troy. Subsequent and more careful archaeology has established that Hisarlık contains at least nine successive cities, built on top of each other, the earliest dating to approximately 3000 BC and the latest to the Roman period. At least one of them, usually identified as Troy VI or Troy VIIa, was destroyed by fire at approximately the right date for the Homeric war. Which of the Troys is Homer's, or whether any of them is, remains actively debated.
What Schliemann accomplished, even if his specific attribution is shaky, was to demonstrate that the Homeric epic had a physical substrate. That had not been an obvious thing. It has reshaped the subsequent study of the Bronze Age Aegean. Schliemann's excavation methods, which involved digging enormous trenches through successive archaeological layers with essentially no recording, also destroyed an enormous amount of material that a modern excavation would preserve. He is a complicated legacy.
4. Angkor (abandoned c. 1431, rediscovered 1860)
The Khmer capital of Angkor, in what is now northern Cambodia, was the center of a powerful Southeast Asian empire from approximately 800 to 1431 AD. At its height in the thirteenth century, it was probably the largest preindustrial urban complex in the world, covering an area of roughly 1,000 square kilometers, with a population that has been estimated, on the basis of recent LIDAR surveys, at perhaps 750,000. The great temple of Angkor Wat, constructed in the early twelfth century, remains the largest religious structure ever built.
Angkor was gradually abandoned over the fifteenth century, probably due to a combination of military pressure from the rising Ayutthaya kingdom, environmental stress on the complex water-management system that had sustained the city, and a shift in Khmer political and religious focus southward to Phnom Penh. The site was not, however, entirely deserted. Buddhist monks continued to maintain the temple of Angkor Wat throughout the subsequent centuries. A series of European travelers, including Portuguese Capuchins in 1586 and various Spanish, French, and Japanese visitors in subsequent decades, wrote brief accounts of what they had seen. The site was not, in any useful sense, lost.
What happened in the 1860s was a change in how the European colonial apparatus encountered it. The French naturalist Henri Mouhot, who visited in 1860 and died of fever the following year, left journals that were posthumously published and popularized. Mouhot's accounts, which described the temples in the language of ruined grandeur (he compared them to the Pyramids), generated significant French public interest, which supported a series of subsequent expeditions under the auspices of the École française d'Extrême-Orient. The rediscovery was a rediscovery by France. Cambodia had been there the whole time.
Readers who enjoy this kind of abandoned-place atmosphere in fiction might look at Record of a Life's Suran, which does some nicely-observed work with old mountain forests and half-remembered clan territories. Adult readership; the setting is a kind of remembered woodland that behaves, occasionally, as if it is not quite empty.
5. Tikal (collapsed c. 900, widely recognized 1956)
The Maya city of Tikal, in what is now northern Guatemala, was one of the largest cities of the Classic Maya period, with a peak population in the range of perhaps sixty thousand and an urban footprint covering sixteen square kilometers. It was part of a network of Maya city-states whose collapse, between approximately AD 800 and 900, is one of the central puzzles of New World archaeology. Tikal was abandoned. The surrounding rainforest reclaimed it.
The city was not unknown to local indigenous populations or to the itinerant chicleros (gum-tree tappers) who worked the forest in the nineteenth century. It was reported by the Swiss-Guatemalan explorer Modesto Méndez in 1848, photographed and described in detail by Alfred Maudslay in 1881–82, and mapped in increasing detail through the early twentieth century. What changed in the 1950s was the scale of institutional commitment to the site. The University of Pennsylvania's Tikal Project, running from 1956 to 1970, systematically excavated and mapped the central core, producing the first scientific characterization of what had been, until then, a mostly-mysterious pile of stone in the jungle.
Recent LIDAR surveys, published in 2018, have dramatically revised the picture. The survey revealed approximately sixty thousand previously-unknown Maya structures across what had been assumed to be empty rainforest. The implication is that the Maya population density, during the Late Classic period, was something on the order of five to ten million people in the lowland region, and that the collapse was a collapse of a much larger and more integrated civilization than the earlier estimates had suggested. The discoveries are still being processed. The "lost" Maya world keeps getting bigger.
On the word "lost"
I want to close with a small point about the word. In each of the five cases above, the city was, at the moment of "rediscovery," perfectly findable by the people who lived near it. The Quechua farmer, the Cambodian monk, the Guatemalan chiclero, the nineteenth-century Campanian peasant: none of them were surprised to learn that there were ruins there. What they were surprised by was that outsiders wanted to come see.
The archaeological category of "the lost city" is, in a specific way, an artifact of the global academic economy. A city becomes "lost" when it drops out of the records maintained by the dominant knowledge institutions, and it is "found" when those institutions re-acquire it. The local population, on whose ongoing presence the site's physical preservation usually depends, tends to be omitted from the rediscovery narrative, or reduced to a role as guide.
I am not making a political point. I am making an archivist's point. What we should be keeping track of, in each of these cases, is the actual chain of custody. Pompeii was kept in the memory of its hillside farmers. Machu Picchu was kept in the cultivation of its terrace gardeners. Angkor was kept by the monks who never left. The cities were not lost. The archives of the people with the telescopes lost them. That distinction, I would argue, is worth preserving.