An Ancient Discovery in Australia Is Leaving Archaeologists Without Answers

[00:00:00]The Nullore plane offers almost nothing to the eye. From the road that crosses its southern margin, the landscape is the same in every direction. Flat, pale, treeless, interrupted only by low salt bush and the occasional fragment of limestone pushed to the surface by what lies beneath. The word itself comes from the Latin nullis arbor, no tree. The description is accurate. What it does not communicate is underneath this surface running more than 200,000 km across southern Australia lies the world's largest semi-arid car system. A honeycomb of passages, chambers, and subterranean lakes carved by water through limestone over millions of years. Thousands of caves sit within that system. Most are invisible from above. The openings where they exist at all appear without warning in the scrubland as collapsed dolines, circular depressions dropping suddenly into darkness. Kunalda is one of these. Its entrance located approximately 99 km west of the Nullore Roadhouse and 30 km north of the cliffs that drop into the great Australian bite is a sinkhole roughly 85 m in diameter. The drop from the rim is approximately 30 m nearly vertical. Below that, a steep slope continues into the main chamber, which sits approximately 70 m below the surface of the plane. The chamber is the size of a football pitch. Beyond it, a series of passages extends for a total of 1,200 m. Near the far end of one of those passages, in a section of the cave that receives no natural light at any time of day or any season of the year, the walls are covered in markings that are at least 22,000 years old. The question of who first reported those markings depends on what is meant by the word reported. Europeans had been reached Kunalda since at least 1860. The cave held a reliable source of fresh water, a series of subterranean lakes, the furthest of which lies some 600 m from the entrance and in the arid conditions of theore water drew visitors as reliably as anything else on the continent. In 1897, the explorer Richard Felwell Maurice was led to the cave by a mering man named Mingo. Maurice documented the lakes. He noted the dimensions of the entrance. He descended into the cave and recorded what he found there. What he did not record, what appears not to have registered at all, was the section of cave wall some 300 m from the entrance in complete darkness where tens of thousands of linear marks had been drawn into the soft limestone surface by human fingers. Maurice was not alone in this omission. For the next 60 years, a succession of government surveyors, private expedition leaders, and military officers visited Kunalda for the water. Captain Maitelland Thompson organized multiple expeditions in the early 20th century. At one point having his team lower a boat into the cave and sail it along the North Passage. No one reported art. The prevailing assumption reinforced through widely circulated articles published between 1920 and 1940 by Irishborn ethnographer Daisy Bates was that indigenous people did not venture onto the nullbore at all. that the plane was avoided territory, left empty because of the mourning belief in Gamba, a serpent living below the surface whose movements were said to cause dust storms to drain the water from clay pans and to take those who wandered into the plane and did not return. Bates's characterization was not contested at the time. It was simply repeated until the physical evidence made it untenable. The evidence in question was not the art, it was the flint. In 1937, museum officers in South Australia acquired a kangaroo skin package containing approximately 40 flint blade blanks from an Aboriginal man living roughly 100 km west of the cave. The material was consistent with the highquality silicon nodules, technically a form of ch or calcedin that occur naturally in the walls of Canalda's main chamber and passageways.

[00:04:09]Those nodules embedded in the limestone when it formed are subject to crystal wedging. Pressure within the rock periodically causes them to fracture, sending large pieces to the cave floor.

[00:04:21]The fallen fragments were available without extraction. Someone had been collecting them, shaping them into tool blanks and moving them across the landscape. At the time, Australian archaeology had not yet established itself as a formal discipline, and the possibility that Aboriginal people had been traveling the Nullerbore, not avoiding it, and using Kunalda as a flint source had no framework in which to be understood. That framework would take another 20 years to appear, and it would require a Hungarian archaeologist without an institutional appointment to construct it. The photographs that changed everything reached Alexander Gallas sometime in 1956. They had been taken during expeditions by the Cave Exploration Group of South Australia, whose members had reported unusual wall markings and indications of Flynn extraction during surveys of the Nullore plane. It was Adrienne Hunt, a member of a University of Sydney contingent exploring the same region, who in 1957 first recognized the archaeological significance of what the photographs showed, markings that he identified as potential prehistoric art forms. Hunt did not know with certainty what he was looking at, but he recognized that someone with European Paleolithic experience might. The photographs reached Gallas. Gallas had directed the prehistory department of the Magiar Nimziti Museum. the Hungarian National Museum from 1931 until 1945 when he fled the advancing communist consolidation of power spent four years in Austria and arrived in Australia in 1949. He was at the time the photographs arrived the most formally trained and experienced archaeologist living in Australia. He had spent years working in the Paleolithic cave sites of central Europe before immigrating as part of the post-war movement of displaced professionals who found that their credentials did not transfer cleanly into the institutional structures of their adopted country. In Australia, Gallas could not secure a university appointment. He tutored in European languages to earn a living. He volunteered his expertise wherever it was accepted. When the photographs from Kunalder reached him, he looked at them and understood immediately what Hunt had found. The linear marks drawn across the soft limestone walls, parallel grooves, lattice shapes, herring bone patterns, fan-like arrangements were directly comparable to markings he had seen documented from Paleolithic cave sites in southern France and northern Spain.

[00:06:49]The technique was identical. fingers drawn across a surface soft enough to retain the impression, leaving traces of the movement in geometric configurations. In Europe, those markings were considered among the oldest forms of deliberate human mark making on record. If the Kunalda markings were of comparable antiquity, the implications for Australian prehistory were significant to the point of being at that moment professionally dangerous to propose. The prevailing consensus in Australian archaeology in the late 1950s held that Aboriginal occupation of the continent extended back no more than a few thousand years.

[00:07:27]The oldest confirmed archaeological sites had been dated to approximately 8,700 years before the present. Gallas examining the photographs and drawing on his European experience estimated that the Kunalda markings were likely to be in the range of 30,000 years old. He organized an expedition. The logistics were formidable. Kunalda lay approximately 1,200 km from Adelaide, the nearest capital city, and the route crossed an unsurfaced highway that was in the late 1950s notorious among drivers for potholes large enough to swallow a vehicle's front axle and corrugations that shook mirrors from their mouths. Gallas assembled a team of volunteers, mostly university students, who raised their own funds by passing a hat and persuaded manufacturers in Adelaide to donate food and equipment.

[00:08:19]In late 1958, the team reached the cave and descended. What Gallas verified on that first visit was the physical reality of the markings, their extent, their depth into the cave system, their position in areas of complete and permanent darkness. The art passage, as it became known, sat beyond the main chamber in a section of the cave accessible only after navigating the steep descent from the entrance and traversing the floor of the chamber itself. The markings covered approximately 350 m of wall surface.

[00:08:52]They were not incidental. They were dense, deliberate, and distributed across the walls in configurations that showed repeated engagement with the same surfaces over what must have been many separate visits. On the floor of the art passage, Gallas's team found charcoal, the residue of torches. There was no other way to explain the presence of people in that section of the cave.

[00:09:15]Without artificial light, it was not possible to reach the art passage at all, navigate within it, or see the walls. Whoever had drawn those lines had carried fire into the cave, descended 70 m below the surface, walked 300 m or more from the entrance into total darkness, and worked on the walls with their hands. The charcoal on the floor was the physical record of that activity. Gallas opened three excavation trenches in the main chamber between 1958 and 1968, the deepest of which reached 4 meters below the cave floor.

[00:09:48]The excavations were conducted with volunteer labor, minimal funding, and almost no institutional support. From the trenches, the team removed tons of flint material for analysis, debitage, waste fragments, and worked pieces that together constituted the archaeological record of systematic extraction. The flint nodules in the cave walls, when they fractured naturally through crystal wedging, produced material that was hard, fine grained, and predictable in its fracture patterns. qualities that made it useful for tool production. The evidence from the trenches showed that this material had been collected, worked into tool blanks inside and outside the cave and moved outward into the landscape. How far outward became clearer later in the 1970s and after analysis of flint artifacts from sites across southern Australia would confirm that Nullerboard shirt had been transported through exchange networks extending hundreds of kilometers from the cave. Pearl shell from Broom in northwest Western Australia had reached the Nullore through the same intersecting routes, a distance of more than 3,000 km. The cave was not a remote anomaly. It was a node in a continental network. The radiocarbon results from Gallas's charcoal samples produced dates that the Australian Archaeological Establishment of the late 1960s found almost impossible to accept. His earliest dates extended to approximately 31,000 years before the present. His proposed range for human activity in the cave from around 31,000 to 8,000 years ago stood in stark contrast to everything that had been established about the antiquity of Aboriginal occupation. When Gallas published his findings formally in 1968 and 1971, the response from the discipline was skeptical. His dating methodology was questioned. His sample selection was criticized. The possibility of carbon contamination was raised. Gallas, without an institutional position and without the access to laboratory resources that an academic appointment would have provided, was not well placed to defend his results against a disciplinary apparatus that had reasons, structural reasons, not personal ones, to resist what he was proposing. In 1967, Richard V s Wright arrived at Kunaldo with resources that Gallas had never had. Wright held a position at the University of Sydney and had trained in prehistoric archaeology at the University of Cambridge. He brought a team that included specialists in rock art, sediment analysis, fondal remains, and flora, a breadth of expertise that Gallas's volunteer crews, however committed, could not replicate. Most critically, Wright had direct access to Australia's principal radiocarbon laboratory. The question of when human activity at Kunalda had begun and when it had ended was by 1967. The central unresolved issue in Australian prehistory. The answer depended on which samples were dated, from which layers, and whether the stratographic integrity of those layers could be defended.

[00:12:52]Wright's excavation opened a new trench adjacent to Galys's deepest work and submitted samples through institutional channels, producing a range of dates between approximately 16,000 and 24,000 years before the present. In 1971, he compiled and edited what remains the only booklength publication on the site, Archaeology of the Galis site, Kunal the Cave, published by the Australian Institute of Aboriginal Studies in CRA.

[00:13:20]The title was itself a form of acknowledgement. Whatever the disputes about methodology, it was Gallas who had opened the site and the record was named for him. The gap between Wright's results and Gallas's oldest date, 31,000 years before the present from a sample in the upper white horizon of Trench 3, was not resolved by the 1971 publication. Both men agreed on certain fundamentals. The cave had not been a place of habitation. There was no evidence of cooking, sleeping, or the accumulation of domestic refues across any of the excavated trenches. What the cave contained was the evidence of two distinct and repeated activities. The extraction of flint and the making of marks. The flint extraction had left behind an enormous quantity of debidage, waste material produced in working raw nodules into tool blanks along with worked and unworked material that had not been removed. The marks covered approximately 350 square meters of wall surface in the art passage. In a section of the cave that had never received natural light and could not be reached without artificial illumination.

[00:14:28]Charcoal fragments scattered across the floor of that passage were the physical trace of the torches that had made access possible. Together, the Flint and the Marx described a place that people had chosen to enter repeatedly over thousands of years for purposes that were not domestic and not incidental.

[00:14:46]The interpretation of the marks was and has remained a separate and more difficult problem. Rock art specialists Leslie Maynard and Robert Edwards documented the flutings in detail, recording their distribution, density, and formal configurations. The term finger flutings itself introduced by rock art researcher Robert Bedner in the 1980s to replace the earlier macaroni and meander coined by French prehistorian Henri Bol described marks made by drawing fingers across a surface soft enough to retain the impression at Kunalda. The display was unlike anything recorded in European Paleolithic caves.

[00:15:26]The configurations included not only the parallel and crossing groove patterns common in France and Spain, but also lattice forms and oval shapes that had no direct equivalent in the Francoantabrian record. In Europe, finger flutings frequently appeared alongside figurative depictions of animals, horses, bison, mammoths, a proximity that had led some researchers to interpret the marks as connected to sympathetic hunting practices. At Kunalda, there was no figurative art.

[00:15:55]The marks stood alone without the visual context that had shaped their interpretation elsewhere. Gallas regarded the engravings as deliberate communication and encouraged art historian Christine Sharp to examine the insized cuts alongside the flutings. A separate question who specifically had drawn the lines generated its own research trajectory decades later. In 2015, researcher Lesie Van Gelder published a claim in childhood in the past proposing that some of Canalda's flutings had been made by children as young as 2 years old, applying a methodology derived from finger width measurement, first used in French caves.

[00:16:34]The claim attracted significant attention and then sustained challenge.

[00:16:38]A 2024 critical analysis published in the journal of archaeological method and theory authored by Karen Walsh, April Null, and Clem Lorie examined the underlying data set and concluded that the evidence was insufficient. Sample sizes were too small. The distribution of age and sex within the samples was uneven. The medium used an experimental comparison was not standardized against actual cave surfaces. And the demographic modeling rested on untested assumptions. The question of who had drawn the lines at Kunalda, their age, their relationship to one another, their reason for being there, remained, as a direct result empirically open. What the physical record established without dispute was duration. The cave had been used across a span that encompassed the last glacial maximum. The period beginning approximately 23,000 years ago when sea levels dropped, when the coast, now 30 km from Kunalda lay 180 km away, and when the plane was significantly more aid than today, the subterranean lakes had dried. The reliable water source that had drawn every European visitor since 1860 was gone. People had continued to enter the cave anyway, descending 70 m below the surface, walking 300 m into complete darkness, carrying fire, and drawing on the walls.

[00:18:00]The charcoal on the floor of the art passage confirmed it directly. No explanation for that choice offered by any researcher since Gallas first entered in 1958 has been accepted as sufficient by the full breadth of the discipline. The question of what Kunalda cave means to archaeology has a documented answer built across 60 years of excavation, dating, and analysis. The question of what it means to the Ming people, its traditional custodians whose connection to the site predates any archaeological investigation by tens of thousands of years has a different kind of answer, one that does not fit inside a published report. Kunalda is a sacred site. Access is restricted to a small number of male Ming elders. The cave is closed to the public and the Ming have actively resisted proposals to open it on the grounds that doing so would violate protocols of custodianship that have governed the place for as long as those protocols have existed. When Ming Elder Bunnori spoke publicly about the cave after the events of December 2022, he described it in terms that archaeology has no framework to measure.

[00:19:07]Kunalda, he said, is like an ancestor.

[00:19:10]The ancestor left a spirit in the wall, in the story, in the song line. That description is not a metaphor offered for an outside audience. It is the account that the custodians of the place give of what it actually is. In December 2022, Kunaldo was broken into. The perpetrators dug under a steel gate at the entrance, descended into the cave, navigated their way into the art passage, and scratched graffiti directly over a section of the finger flutings.

[00:19:39]The message they rode into the wall over marks that had been placed there between 15,000 and 30,000 years ago, used the soft limestone surface in exactly the same way the original makers had, pressing into it with their hands. The difference was that the original marks had been made in complete darkness by people who had carried fire 300 m into the earth to reach that wall. The graffiti was made with a torch and a disregard for what it covered.

[00:20:06]Archaeologist Karen Walsh, who has spent decades studying ancient Aboriginal sites across southern Australia, told the Guardian that the art underneath was not recoverable. The limestone surface is too soft. Removing what was placed over it would remove what was already there. What the finger flutings recorded across that section of wall, whatever specific act or intention or sustained practice they represented, is no longer fully accessible to any method of analysis that currently exists. The Ming community had been requesting improved security at the site since 2018. In their statement following the December 2022 incident, the community noted that the entrance had already been damaged by access works carried out without their consultation and without their approval and that repeated requests for effective fencing and on-site signage had not been acted upon. Claire Buswell, chair of the Australian Speliological Federation's Conservation Commission, had submitted written testimony to the Aboriginal Lands Parliamentary Standing Committee in July 2022, 5 months before the vandalism, identifying the absence of adequate security measures as a condition that made an incident of exactly this kind foreseeable. The South Australian Attorney General and Aboriginal Affairs Minister at the time, Caamar, described the incident as shocking, called for severe penalties, and noted that under the state's Aboriginal heritage laws, damaging an Aboriginal site carried a maximum penalty of a $10,000 Australian dollar fine or 6 months in prison. In March 2023, the Australian government committed $400,000 Australian dollars toward improved security, including the installation of remotely monitored cameras operating continuously. The cameras were subsequently installed.

[00:21:52]What that sequence of events demonstrates the community's requests, the warnings from the conservation sector, the gap in which the damage occurred, the response that followed is a pattern that archaeology alone cannot address. The physical record at Kunalda is the result of repeated human choices made in darkness over a span of at least 15,000 years and possibly longer in a location 70 m below a plane that the dominant culture of the 20th century believed Aboriginal people had never occupied. the accumulated evidence from Gallas's volunteer trenches, from Wright's radiocarbon laboratory, from the 2013 resampling of trench 3, and from the 2024 critical analysis of the finger fluting methodology has produced a picture of a place that was used continuously and purposefully across a stretch of time that has no parallel in the archaeological record of any comparable site in Australia. What it has not produced is a complete account of why or by whom in any specific sense or what the marks on the walls were intended to do in the world. Those questions remain open and they remain open not because the evidence is insufficient to answer them but because the questions themselves may not be the right ones or because the people best positioned to address them are the ming and the murning have not been adequately heard. The cave is there, the wall is there. Parts of what was on the wall are no longer there in the form in which they were left. What remains? The sections of finger flutings that were not reached in December 2022. The debage on the floor of the main chamber. The charcoal scattered across the art passage. The standing stones identified but never fully reanalyzed continues to hold more than 60 years of accumulated research questions that no single discipline has yet resolved. Kunaldo was added to Australia's National Heritage List on October 15th, 2014, the 102nd place to receive that designation. The listing cited its role in transforming the understanding of the antiquity of Aboriginal occupation in Australia and its particular and ongoing significance to the Ming people. Both of those things remain