HistoricalChristian.Faith

Did Liangzhu's jades require technology beyond Neolithic capacity to make?

Type: supporting argument (off-timeline) · Assessment: weak — as the claim that the objects require machine tooling or an inherited advanced technology. The puzzle the argument starts from is real and is not a beginner's mistake: nephrite cannot be carved, the incised emblem on the "king of cong" packs four to six non-overlapping lines into a millimetre, and the standard explanation of Liangzhu's polish — quartz sand — turned out to be wrong. But it was corrected from the ground, not from the sky. The abrasives were harder than anyone assumed (corundum, and on the finest surfaces diamond), both occur naturally within reach of the delta, the tool-marks are diagnostic of hand abrasion rather than of a wheel, and the workshops themselves have been excavated: a million square metres of jade debris, blanks, rejects, whetstones and flint gravers at Deqing. A second and separable sub-claim — that Liangzhu is much older than its Neolithic date, reaching toward a pre-Flood horizon — is also weak, and for the reason given in its own section below: these are stratified grave goods with radiocarbon and an independent speleothem chronology, not a free-standing monument that a later ruler could have appropriated.

Related arguments: out-of-place artifacts (dates under skepticism) — see also whether this reconstruction's witnesses are securely dated · the age of the Kailasa temple at Ellora · the "lathe-turned" pillars of the medieval Deccan · the vimanas and divine weapons of the Sanskrit epics · the Ta Prohm "stegosaurus" carving · Angkor's engineering and the antiquity of the Khmer capital · predynastic Egypt's hard-stone vases · the Great Sphinx's water-erosion redating · the Osireion at Abydos · the Giza radar scans and the "second Sphinx" · Sacsayhuamán's polygonal walls

See also: the Watchers as teachers of the arts (assessment: contested), the mechanism this argument would need · Liangzhu on the timeline · the Jishi Gorge outburst flood, the other East Asian entry.

The claim

Two things about Liangzhu jade are true and are the whole basis of the argument. The first is the material. Nephrite sits at about 6–6.5 on Mohs' scale and is exceptionally tough as well as hard; it does not flake like flint and it does not chisel. Nothing softer than quartz will mark it. The second is the work. From the Yangtze delta between about 3300 and 2300 BC come squared cong tubes with true bores, bi discs drilled perfectly round, and surfaces polished to a mirror — some nine-tenths of all surviving Neolithic cong and bi, produced by a culture with no metal.

The showpiece is the object nicknamed the "king of cong," excavated in 1986 from Tomb 12 of the Fanshan cemetery at Yuhang and now in the Zhejiang Provincial Museum: 8.9 cm high, 6.5 kg, lying beside the head of the man buried there among 641 other jades. On its four vertical channels are eight complete versions of the Liangzhu deity-and-beast emblem, the shenren shoumian — a plumed human figure astride a tusked, clawed monster — each about 3 cm by 4 cm, executed in shallow relief for the masses and in incised line for everything else.¹

It is the incised line that generates the argument. Fang Xiangming, who directs the Zhejiang Provincial Institute of Cultural Relics and Archaeology and has measured the piece, reports that within a single millimetre the engraver laid five to six lines that do not overlap; the eyeballs of the beast are two to three millimetres across; the mean width of an incised line is under 0.2 mm. He calls it micro-carving, and a modern viewer needs magnification to read what is there. From this the argument infers what it takes to be the only sufficient cause: a rotary machine, or a tooling tradition inherited from somewhere the Neolithic could not have reached on its own — on this timeline, the arts the Watchers are said in 1 Enoch 7–8 to have taught, surviving into the post-Flood world.²

Who advances it

The claim belongs to the general out-of-place-artifact literature rather than to a single author, and it travels in the popular form "no Stone Age hand could do this." It earns a page because it is genuinely advocated, because it is one of the few places where the reconstruction this timeline explores makes contact with a testable material question, and because the strong version of it is not silly: it asks what the artisans actually had, and until 2005 the published answer to that question was demonstrably incomplete.

The evidence for

The difficulty is not exaggerated. Every specialist account agrees that jade is not carved but abraded — ground away by loose sand harder than itself, carried in water or oil, on a tool of bamboo, bone or wood that leaves no mark of its own. The Smithsonian's account of a Liangzhu bi puts the sequence plainly: "Working jade is an extremely laborious process that involves both slicing and removing unwanted stone with powdered abrasive minerals, such as quartz, garnet, or corundum"; the central hole was made with "hollow tubular bits, aided by mineral abrasives and water … from both sides of the disk to avoid breakage"; and "lastly, the surface was polished with fine abrasives to create a high sheen." Nothing in that description is quick, and the Liangzhu output is not small.³

And the received account of the abrasives was wrong. This is the argument's strongest card and it should be conceded in full. Quartz sand had been "generally assumed to have been the major abrasive used in prehistoric China." In 2005 Peter Lu and collaborators analysed four ceremonial axes from Sanxingcun, Beiyinyangying and Liangzhu contexts and found them made of a corundum-dominant rock previously undescribed — corundum being Mohs 9, the second-hardest natural mineral, and its earliest attested use anywhere by some two thousand years. They then took a mirror-polished Liangzhu axe fragment of c. 2500 BC, cut slices from it, polished them with the best modern analogues of quartz, corundum and diamond, and compared all four surfaces under an atomic force microscope. The surface polished in antiquity matched the diamond-polished slice and not the others, on shape, on height histogram and on the height–height correlation function alike; the corundum grains in the original stand at a root-mean-square roughness of a few nanometres. Alluvial diamond sources lie within 300 km. Their conclusion: "Neolithic craftsmen of ancient China were certainly using corundum and very possibly diamond about two millennia before anyone else was known to have done so."⁴

So the argument is right that the standard picture needed revision, and right that these objects imply capacities nobody had credited to the fourth millennium BC. The question is what kind of capacities.

Adversarial assessment

What holds. The precision is measured, not asserted; the material is intractable; and the "they only had sand" reply that the claim is usually met with is itself false. An argument that forces a correction of the textbook account has done real work even when its own conclusion fails.

The tool-marks are readable, and they are the marks of a hand. This is the load-bearing objection, and it is positive evidence rather than an appeal to what Neolithic people "must" have done. Tool-marks on worked jade survive and can be told apart. Sax, Meeks, Michaelson and Middleton examined eight nephrite artefacts spanning Hongshan and Liangzhu, the Western and Eastern Zhou, and the Ming and Qing, confirmed the material by Raman microscopy, and — by taking moulds of the tool-marks and imaging them in a scanning electron microscope, a method developed for Near Eastern quartz cylinder seals — established criteria by which six distinct working techniques can be recognised. The same team went on to date the arrival of the true rotary incising wheel in China from those criteria. The point for this argument is the negative one: a rotary wheel is visible in the marks it leaves, it is not visible on Liangzhu jade, and the techniques that are visible there are string and blade sawing with loose abrasive, tubular drilling, grinding, and hand incision. If a machine had cut the emblem on the Fanshan cong, the evidence for it would be on the object, where anyone can look.⁵

The workshops have been excavated, with the tools and the failures in them. Between 2017 and 2019 the Zhejiang institute and the Deqing county museum worked a site complex at Zhongchuming, about 18 km from Liangzhu City: twenty-four artificial earthen platforms over roughly a million square metres, radiocarbon-dated to c. 4800–4500 years ago, from which came some three thousand pieces of jade raw material, blanks, half-finished objects and broken rejects, together with the toolkit — whetstones, grinding stones, drills, chalcedony and flint. It is the largest Liangzhu lapidary complex known, and it is not alone; a jade workshop with cutting, drilling, abrading, polishing and incising tools was identified north of Nanjing at Dingshadi in 1997. This is what a craft industry looks like in the ground: raw material, waste, mistakes, worn abrasive stones, and a settlement of ordinary rank tied to the capital. A machine technology, or a technique received whole from elsewhere, does not leave a debris field of hand tools and failed blanks.⁶

The disputed question is which hand tool, not whether a hand tool. Fang Xiangming — the same excavator who supplies the millimetre figure the argument leans on — attributes the fine incision, including the laying-out of the design, to gravers of flint, and names the jade-working sites from which such flint gravers have been recovered: Mopandun at Danto, Dingshadi at Jurong, the Tangshan Jincun and Zhongjiagang deposits at Liangzhu itself, and Zhongchuming at Deqing. A rival proposal, prompted by the shark teeth found at Yaoshan and elsewhere in Liangzhu contexts, is that a tooth was hafted and used as a burin; a third points to quartz crystal. Under magnification the incised lines are not smooth grooves but diagonal parallel striations, and some are visibly built up out of short interrupted strokes of uneven width and depth — the signature of a point dragged repeatedly by hand, and the one thing a machine would not produce. The debate among specialists is about which hard point was held; nobody working on the material proposes that no point was held.⁷

What the numbers actually demand. Six non-overlapping lines to the millimetre is a spacing of about 0.17 mm, with individual lines averaging under 0.2 mm wide. That is fine work; it is not a scale that defeats the unaided eye and a braced hand, and it is the ordinary range of pre-industrial engraving, seal-cutting and die-sinking in several unconnected traditions. The claim is sometimes pushed further — a widely repeated popular figure has the narrowest lines at "less than 0.01 mm," ten microns — but that number is not the archaeologists' measure and nothing should be built on it. The honest statement of the puzzle is Fang's: an emblem 3 cm by 4 cm, eyeballs two to three millimetres across, five or six lines to the millimetre, and no metal in the culture that made it.²

The trajectory is a craft tradition's, not a technology's. The capacity does not appear from nowhere and it does not behave like transmitted knowledge. It grows out of an older lithic industry in the same valley — the corundum axes of Sanxingcun date to c. 4000–3800 BC, a millennium before the Fanshan cong — and it is carried by a specialist industry inside a state that lasted the better part of a thousand years, with hereditary workshops, graded settlements, and single elite tombs holding six hundred jades. Then Liangzhu collapses about 2300 BC and the technique goes with it: nothing of comparable fineness is worked in China again until the Shang, more than a millennium later. Skills die with the institutions that pay for them. That is exactly what a long apprenticed craft does under state patronage, and exactly what an inherited machine technology would not do.⁸⁴

What is genuinely still open. Lu and colleagues found no vestigial abrasive left on the polished surface and closed with the caution that "further studies may be needed to establish the specific tools and techniques used to create these lustrous surfaces"; the identity of the incising point is unsettled; and no full replication of the Fanshan emblem at its own density has been published. These are open details inside a process that is understood, and they should be reported as such rather than either suppressed or promoted into a mystery. An unanswered question about which stone was in the artisan's hand is not evidence that no artisan's hand was involved.⁴

The dating sub-claim, kept separate

The manufacturing question and the antiquity question are different, and conflating them is how this class of argument usually goes wrong. Suppose the manufacturing claim failed entirely, as it does; one could still ask whether Liangzhu itself — or the technique it implies — is much older than the conventional date, far enough to reach a pre-Flood horizon. What would have to hold for that? The real precedents that make such a move live elsewhere are monument usurpation, renovation counted as construction, and reused foundations: mechanisms that work on architecture, where a later ruler can inherit and re-inscribe someone else's building.

None of them transfers here. The jades are portable objects recovered from sealed burials in a stratified settlement sequence — Majiabang to Songze to Liangzhu — dated by radiocarbon on associated organic material, with the whole assemblage typologically internal to that sequence. There is no earlier structure to appropriate and no way for a grave-good to be older than the grave without leaving the stratigraphy visibly disturbed. For this antiquity sub-claim — the charge that Liangzhu's measured date is wrong by millennia — the choice of Flood date is moot: whether the Flood is placed at the Masoretic c. 2348 BC or the stronger Septuagint c. 3300–3000 BC, Liangzhu's own conventional c. 3300–2300 BC range already straddles the long chronology's horizon and lies largely before the short one's — so the claim at issue is not that the jades reach an undated deep past but that their measured date is wrong by millennia, and that is what the independent chronology below defeats. The Flood date is not incidental, though, to a different and stronger form of the argument — that Liangzhu inherited a pre-Flood capacity rather than possessing an anomalous date — which the next section takes up. And the terminal date is fixed by a method wholly independent of the site: uranium–thorium dating of stalagmites from two caves in the same monsoon domain places the collapse inside a wet interval between 4345 ± 32 and 4324 ± 30 years BP, a precision of about thirty years, matching the clay layer over the abandoned city. Converging methods with different failure modes are the condition under which a date is secure, and they converge here. The antiquity sub-claim is weak — not because the consensus says so, but because the specific mechanisms that could unseat a date do not apply to this kind of evidence.⁹

Where the Flood date is not moot: the inheritance reading

There is a second and stronger form of the manufacturing argument on which the Flood date is not incidental, and it deserves to be stated at its best. That form does not claim Liangzhu is secretly older than dated; it grants the conventional date and claims instead that the capacity was inherited — that the abrasive arts were known before the Flood and carried through it into the post-Flood world, the mechanism this reconstruction locates in the arts the Watchers of 1 Enoch 7–8 are said to have taught. On that reading the date the argument needs is the one that leaves Liangzhu standing after the deluge, so its artisans can be heirs and not casualties — and the choice of reckoning decides whether the reading is even coherent.

The wiki's stronger yardstick, the Septuagint long chronology (Flood c. 3300–3000 BC), is the one that makes it live, and the fit is worth conceding plainly. On the short Masoretic reckoning (c. 2348 BC) Liangzhu's own c. 3300–2300 BC range falls almost entirely before the Flood, which turns the culture into an antediluvian society a universal deluge should have ended — not an heir but a casualty, and one that on the premise ought not to have left this stratigraphy intact. On the long reckoning the Flood falls at the very start of the Liangzhu range, the Fanshan great cong of c. 2500 BC sits a comfortable half-millennium after it, and the abrasive tradition's own history straddles the horizon in the very pattern an inheritance story predicts: the corundum axes of Sanxingcun — the earliest attested use of that mineral anywhere — fall c. 4000–3800 BC, before the Flood, and the technique climaxes at Liangzhu after it. Granting the premise, that is a real pattern, not one the argument invented.⁴

And yet the verdict does not move, because the evidence that defeats the manufacturing claim is untouched by which reckoning is chosen and lies wholly on the post-Flood side of both. What the inheritance reading needs is a capacity beyond Neolithic reach — a machine, or a finished technology received whole. What the post-Flood ground at Liangzhu holds is the opposite: tool-marks diagnostic of a dragged hand-held point, with no rotary wheel among them;⁵ a lapidary complex at Zhongchuming of c. 2850–2550 BC — itself after the long-chronology Flood — full of blanks, rejects, worn whetstones and the failures of people learning a hard craft;⁶ and incised lines built up from short interrupted strokes of uneven depth, the signature of a hand and not of an heirloom. One concession is owed in the other direction: the subsidiary point that the capacity "grows out of" the older Sanxingcun industry cannot be pressed while the premise is granted, since a global Flood falling between the two would sever exactly that continuity — so that defeater is set aside here, and the load-bearing ones do not need it. The know-how that could, on this reading, have survived the Flood is at most ordinary skilled abrasion, which is not the beyond-Neolithic capacity the argument requires and which the delta demonstrably worked out for itself, in its own rivers, after the water went down. The Flood date thus decides whether the inheritance reading is coherent; it does not rescue it. The manufacturing claim stays weak on evidence that no choice of Flood horizon disturbs.

Assessment

Weak — as the claim that Liangzhu jade required machine tooling or a technology beyond Neolithic capacity. The manufacturing process is understood in its main lines and is attested at every step by material that has been dug up: abrasion rather than carving, with loose sand harder than nephrite; string and blade sawing; tubular drilling from both faces; grinding and polishing with corundum and, on the finest surfaces, diamond; and hand incision with a hafted hard point, flint on the excavator's reading. The tool-marks are diagnostic and show no rotary wheel. The workshops, the raw material, the blanks, the rejects and the abrasive stones are in the ground at Zhongchuming and Dingshadi. And the capacity rises and falls on the timescale of an institution, not of an inheritance.

Two things are owed to the argument. Its premise about the difficulty is correct, and pressing it is what produced the corundum-and-diamond result — a genuine discovery that pushed the earliest known use of both minerals back by two millennia, and that a complacent defence of the received account would have missed. And its central observation stands undiminished: five or six incised lines to the millimetre, cut without metal, on an object whose emblem no living viewer was meant to examine closely. What the evidence will not support is the inference drawn from it. The correct reading of that emblem is not that someone else made it; it is that these people were extraordinarily good at what they did, over many generations, with materials they found in their own rivers — which is a more interesting result than the one the argument was after, and unlike it, is what the ground actually shows.

Sources