An Atlas of Money · All entries
Varna gold and spondylus
Attested Little disputed
Open on the mapThe oldest large assemblage of worked gold (about 3,000 objects), buried with Spondylus shell ornaments in a Chalcolithic cemetery.
No state involved
No one's debt

Claims and evidence
The world's oldest large assemblage of worked gold (c. 4600–4200 BC). Attested Little disputed
- Ivanov & Avramova, Varna Necropolis (2000)
Spondylus and Glycymeris valves were 'the earliest universal pan-European means of payment'; with weight-controlled gold, the earliest standardised value system. Weak lead Strongly disputed
Contested by Chapman, Fragmentation in Archaeology (2000): Spondylus as tokens of social relations, not currency
- Todorova (2002), p. 181
The museum catalogue of Balkan metals reads the Varna gold in a social, symbolic and religious context, alongside the first copper mining. Attested Little disputed
“which are contemporary with early copper mining and metallurgy in the Eastern Balkan region, have led to a new understanding of the role of gold in a social, symbolic and religious context”
Alexandrov et al., Gold & Bronze: Metals, Technologies and Interregional Contacts in the Eastern Balkans, NAIM catalogue (2018)
Kostov proposed a weight unit, the 'van' of 0.415 g, and expressed the gold bracelets, diadems and pectorals as multiples of it. Supported Disputed
Contested by the atlas re-test below
“has revealed at least two minimal weight units of 0.145 and 0.415 g.”
Kostov, 'Prehistorical weight system among the golden objects from the Varna Chalcolithic necropolis', Geology and Mineral Resources 11(3) (2004), 25–28“The 0.415 g unit (van) is the mean of 752 small gold beads from grave 36, total 312.5 g.”
Kostov, 'Prehistorical weight system among the golden objects from the Varna Chalcolithic necropolis', Geology and Mineral Resources 11(3) (2004), 25–28“rounded to whole numbers and halves”
Kostov, 'Prehistorical weight system among the golden objects from the Varna Chalcolithic necropolis', Geology and Mineral Resources 11(3) (2004), 25–28
Re-tested on Kostov's own figures, whole multiples of the van are not supported. Half-van steps fit better than chance, but only if that unit is assumed in advance, and the fit vanishes with catalogue errors of a tenth of a gram. Supported Disputed
Contested by Kostov (2004, 2017)
“At the van itself (0.415 g) there is no sign of a unit”
Atlas re-test of Kostov's 2004 weights (Kendall quantogram and Monte Carlo), 16 Nov 2026“random weights fit some unit as well or better in 92% of trials”
Atlas re-test of Kostov's 2004 weights (Kendall quantogram and Monte Carlo), 16 Nov 2026“with ±0.1 g, in 6%”
Atlas re-test of Kostov's 2004 weights (Kendall quantogram and Monte Carlo), 16 Nov 2026“some 1,500 years before the first known balance weights (c. 3000 BC; Ialongo et al. 2021)”
Atlas re-test of Kostov's 2004 weights (Kendall quantogram and Monte Carlo), 16 Nov 2026
Radiocarbon dates place the Varna I graves at about 4600–4400 BC, earlier and narrower than once thought. Attested Little disputed
“Our determinations date Late Eneolithic graves from the Varna I cemetery to 4600–4550 to 4500–4400 BC.”
Higham, Chapman, Slavchev, Gaydarska et al., 'New AMS radiocarbon dates for the Varna Eneolithic cemetery', Acta Musei Varnaensis 6 (2008)