35 of 40 famous megaliths are dated only by association.Only Nan Madol has a direct date on its stones; the ages of the Great Sphinx, Carnac, Sacsayhuamán and others rest on nearby material, texts or style. We rank the 10 where a first direct date would matter most.Megalith leads →
Lead
The Great Sphinx enclosure has never been dated directly.A paired chlorine-36 test against Khafre-era quarry faces could show an older age, or confirm the accepted one.Sphinx case →
News
Monte Verde, the textbook vindication, was reopened in 2026.The challenger's own radiocarbon dates match the older age, and the key layer was destroyed decades ago.Monte Verde →
Pattern
Most celebrated vindications hold at their core.7 of 10 hold, usually with later refinements; 3 are still contested. All 7 refutations held. (Corrected after outside feedback: we had wrongly counted refinements as partial failures.)Observation O1 →
Tentative
“It can't be that old” objections fail more often.25% of objections that proved wrong vs 9% of those that held: the predicted direction, not significant (p = 0.19). The other five hypotheses need more resolved cases.Hypothesis H4 →
Null
Two database scans found nothing new.No clustering among 173,946 radiocarbon dates. The 10 early-Americas candidates were contamination, miscoded samples or database errors; 4 remain unexplained only because their site reports were never published.Discovery track →
Every claim, at the depth it claimed
VindicatedPartly vindicatedOpenUnsupportedRefuted
Across: year the claim was made. Down: age claimed (log scale), over the geological epochs. Bars show claimed ranges; hover a refuted case for where it really belongs. Claims with no figure in years (e.g. Piltdown’s “Early Pleistocene”) are listed in the cases table only.
Time to verdict median 12 yrs to vindicate · 2 yrs to refute
Pre-registered hypotheses published before any data
Hypothesis
n / needed
Status
H1
The size of the leap predicts the fate
4 / 12
awaiting data
H2
Independent dating methods predict vindication
14 / 20
awaiting data
H3
One team and closed material predict refutation
14 / 20
awaiting data
H4
The objections that proved wrong were mostly priors
60 / 30
consistent
H5
Vindication is slower than refutation
14 / 20
awaiting data
H6
A proponent's outside stake is a warning sign
14 / 20
awaiting data
H7
Many megaliths are dated by association, not directly
White Sands footprintsHuman footprints in an ice-age New Mexico lakebed, dated to ~23,000–21,000 years ago; the dates are disputed.
Medium costOSL-date sediment directly above and below several track horizons (single-grain quartz or feldspar), blind, at a lab outside both camps, with water-content history modelled.
6.5/8
1.3×
5
no
2
Jebel IrhoudSkulls once read as Neanderthal; dated in 2017 to ~315,000 years and proposed as earliest H. sapiens.
Medium costAttempt enamel protein sequencing on an Irhoud tooth, as done for Homo antecessor (Welker et al. 2020), to help place Irhoud relative to Neanderthals and H. sapiens.
6/8
1.6×
9
no
3
Madjedbebe (Malakunanja II)Tools in a northern Australian rock shelter that may put people in Australia 65,000 years ago.
Medium costSearch archived Phase 2 material for identified charred Pandanus endocarp or intact hearth charcoal, then AMS-date it (ABOx-SC) at two labs; a 1989 hearth sample there already failed pretreatment at ANSTO.