On a beach south of Haifa, a graduate student recently climbed onto a bundle of dried reeds and paddled into the Mediterranean.
Just below him, beneath a few meters of water, lay the remains of an ancient village where people lived some 8,000 years ago. Similar settlements are scattered along the coast of northern Israel, dating from roughly 10,000 to 7,000 years ago. Their inhabitants fished, farmed, herded goats, and buried their dead in communities that now lie submerged underwater.
These were early farmers in the Neolithic Period, when people in this region moved away from hunting and gathering and settled into permanent villages, where they grew crops and kept animals. Eventually, the sea rose and swallowed the land, and it has been underwater ever since.
Researchers are confident that Neolithic communities in the eastern Mediterranean ventured out into deep water. Some of them even crossed to Cyprus and settled the island with plants and animals carried from the mainland. What nobody knows is how any of this was done. Fishhooks, weights for fishing nets, and fish bones have been found at the submerged sites, but not a single boat from that period has turned up with them, or anywhere else across the eastern Mediterranean.
That missing link is why Shai Niv, a master’s student in the Department of Maritime Civilizations at the University of Haifa’s School of Archaeology and Maritime Cultures, spent months harvesting and drying about 1,600 reeds, binding them into a four-meter vessel, and taking it out to sea.
A stretch of coastline that went under
Around 20,000 years ago, at the peak of the last Ice Age, the sea level was roughly 120 meters lower than it is today, and the Mediterranean shoreline lay about 10 kilometers west of its current location. Numerous villages stood on what was then dry land.
Then the ice sheets melted, and the sea level rose — fast by geological standards, at roughly 12 to 14 millimeters per year during one stretch. Over the centuries, the villages closest to the water were flooded and abandoned.
Israeli archaeologists have located about 19 of these drowned settlements along a 20-kilometer stretch of the Carmel coast between Haifa and Ma’agan Michael. They sit between the present shoreline and about 12 meters deep.
Underwater archaeology in Israel began around 1960, and traces of submerged settlements turned up almost immediately. The first excavation took place in 1969, after a winter storm stripped the sand from a site at Neve Yam. A systematic search began in the early 1980s, when Ehud Galili of the University of Haifa established a project to locate, excavate, and study the settlements.
The sites survived unusually well. Clay and sand sealed them quickly, preserving organic material that would normally have rotted away, such as wooden bowls, woven baskets, straw mats, and crushed olive pits.
How we know they went out on the water
The evidence that these villages were seafaring is indirect, but it stacks up.
Archaeologists recovered fishhooks and stone weights used to sink fishing nets, while the sizes of the fish bones at the sites suggest nets were in use. More tellingly, some of the bones belong to open-water species that don’t come close to shore. Researchers interpret that as evidence that people fished offshore, which would have required a boat.
At one site, Atlit-Yam, large concentrations of gray triggerfish bones were found within specific structures. An analysis of the bone fractures suggests that the fish were gutted and processed, possibly for storage or trade rather than for a single meal.
Then there are the human skeletons. More than 60 skeletons have been recovered from Atlit-Yam. Roughly a quarter of the male skeletons show abnormal bone growth in the ear canal, a condition found in modern surfers after repeated exposure to cold water. Researchers read it as a sign of habitual diving, perhaps for spearfishing or gathering shellfish.
The skeleton of one young woman shows markers of repeated strain in her upper body. The researchers who described it suggested that the strain may have come from rowing, though they were careful to present that as a possibility rather than a conclusion.
The missing link
While evidence that these villages were seafaring has existed for decades, the boats they may have used have been nowhere to be found.
The problem is that boats made of plant material, like wood or reeds, rot, unlike stone tools, pottery, and bone. While some plant material, like baskets, survived, the boats did not. With no physical remains to build from, Niv — working under Prof. David Friesem and Prof. Deborah Cvikel — turned to experimental archaeology. In other words, the team reasoned their way to a plausible version of what existed, built it with period-appropriate materials, and tested whether it worked.
Niv started with a survey of early watercraft from across the ancient world, drawing on carvings, paintings, written descriptions, and small clay boat models found in excavations. He then worked in the opposite direction, reconstructing what the Carmel coastline looked like 8,000 years ago and identifying what raw materials would have been available nearby.
Finally, he compared those findings against records of traditional societies in similar settings — including wetlands, river deltas, and coastlines — and how they built boats. Niv and his supervisors concluded that a reed vessel was the most plausible option for the Carmel coast.
A boat made of dried plant stalks might sound flimsy, but these kinds of vessels were actually widespread and durable.
Reed vessels appear in ancient Egyptian and Mesopotamian art, and versions of them are still built and used today on Lake Titicaca in South America, on Lake Tana in Ethiopia, and in the marshes of southern Iraq. Bundled reeds are lightweight, inexpensive, widely available in wetlands, and surprisingly buoyant.
Before committing to a full-size build, the team made a scale model in a workshop led by Cvikel, head of the Department of Maritime Civilizations, who studies shipwrecks and ancient sailing technology. “Before investing the resources required to build a full-size vessel, it was important for us to first build a scaled-down model with the help of the students, in order to test the boat’s buoyancy and sailing ability,” Cvikel explained to Ynet. “This was especially important in light of the uncertainty regarding the existence of sailing technology using sails during this period. After the scaled-down model proved the feasibility of the idea, we moved on to building a 4-meter-long reed boat.”
After the model held up, the team built the full-size version. About 1,600 reeds were harvested, dried, and cleaned. Then, with help from students in the Department of Maritime Civilizations, the reeds were tied into bundles, lashed together in a crosswise pattern, and tensioned diagonally to give the vessel rigidity.
Once the boat was ready, Niv launched it off Neve Yam beach, directly above one of the submerged villages. The boat performed well, carrying him and his equipment as he paddled offshore.
“It’s like making a dream come true: starting from a theoretical idea and ending with a full-size, tangible and real boat,” Niv said. “The boat remained stable, sailed, and was able to carry me and the equipment, demonstrating that such a structure is capable of functioning as a seagoing vessel.”
Friesem has been direct about the limits.
“The success of the experiment does not prove that this is exactly what the vessels of the Neolithic period looked like, but it shows that a reed boat is an optimal utilization of the natural resources in the coastal area. In this way, the study adds a new layer to the understanding of how the residents of the Carmel Coast were able to live in the area and adapt themselves to the conditions of the coastal environment thousands of years ago.”
Why a bundle of reeds is worth this much trouble
The reed boat could fill a big hole in a much larger story about how humanity developed farming, settled in permanent villages, and migrated across continents.
The Carmel coast falls inside the Fertile Crescent, the arc of land running through modern Israel, Lebanon, Syria, Iraq, and southeastern Turkey. This is where researchers place the origins of farming and settled village life; from there, these practices spread into Europe, North Africa, and beyond. Much of the physical evidence of how that happened is still buried, waiting to be discovered.
Some of it is on a scale nobody expected. In 2019, ahead of highway construction about five kilometers from Jerusalem, the Israel Antiquities Authority excavated a settlement near Motza that its directors described as the largest Neolithic site ever found in Israel or the wider Levant. With an estimated population of 2,000 to 3,000 residents, the ancient site had public buildings, plaster floors, planned alleys between houses, storerooms holding lentils, and thousands of arrowheads.
That inland site dates to roughly 9,000 years ago, making it a relative contemporary of the drowned village at Atlit-Yam.
The coastal sites have also been home to surprising discoveries of their own, producing what researchers describe as the earliest known stone-lined water well, the oldest known coastal defense wall anywhere in the world, and the earliest known evidence for olive oil production. While these “earliest known” claims could be overturned by a new find, they still show how ancient communities kept inventing solutions for a changing world.
Friesem and his colleague Gilad Shtienberg have argued that the coastal communities that were at the center of the recent experiment are missing from the standard story of early farming because they are underwater and hard to reach. Their absence reflects what archaeologists can see, not how much they mattered. The coast was where ancient communities had to innovate because the ground kept disappearing beneath them.
That coast was also the ancients’ road west. People crossed open water to Cyprus, an island never connected to the mainland, and carried plants and animals with them. Boats were the vehicle for one of the largest transformations in human history, and the ones used along this stretch of coastline have left no trace.
Reeds grew in many marshes along this coast. They cost little, but they rotted fast. That may be exactly why people used them, and exactly why nothing is left.
