The Bunnik Hoard 2023

Research History

In October 2023, Reinier Koelink and Gert-Jan Messelaar, using metal detectors, discovered a significant Roman coin hoard in the municipality of Bunnik (province of Utrecht, the Netherlands). The hoard consists of 404 silver and gold coins of British and Roman origin, with the youngest coin dating back to the reign of Emperor Claudius (41–54 CE). This is the largest Roman coin hoard ever discovered in Utrecht and the first mixed composition hoard found in mainland Europe. The coins are believed to be connected with the Roman conquest of Britain and are of great historical, social, and scientific value.

The finders reported the hoard to the Archaeological Reporting Centre of Landschap Erfgoed Utrecht (Utrecht Landscape and Heritage Foundation), which cooperates with the national database and public platform Portable Antiquities of the Netherlands (PAN). The finders first contacted archaeologist Anton Cruysheer, who subsequently coordinated all activities related to the coin hoard with all parties involved. He also carried out the examination and identification of the coins, with support from Paul Beliën of the Numismatic Information System (NUMIS) of De Nederlandsche Bank and John Sills for the British gold staters. As a treasure find, the discovery was also reported to the municipality of Bunnik, which issued a formal acknowledgement to the finders and the landowner.

The coins were documented in PAN by archaeologists Wouter Hinrichs and Julius van Roemburg and then cleaned by Rasmus Thelen of the Restoration Atelier Restaura. After independent evaluation by two experts, the owners reached an agreement with the Rijksmuseum van Oudheden (the Dutch National Museum of Antiquities), where the coins are now on public display. The exhibition was curated by archaeologist and conservator Jasper de Bruin.

The National Cultural Heritage Agency conducted research at the find location to map the archaeological and landscape context of the coins. The research team consists of Tessa de Groot, Jan-Willem de Kort, Harm-Jan Pierik, Menno van der Heiden, Koen van Egmond and Willem Derickx. Emmy Lammertsma and Liesbeth van Beurden of BIAX Consult are working on the archaeobotanical analysis. And radiocarbon dating of the samples is done by Safoora Kamjan of the Centre for Isotope Research of the University of Groningen. A detailed report will be published in 2026.

Location and Context of the Find

The hoard was found in the municipality of Bunnik in a field far from known Roman sites. Bunnik lies along the former northern border of the Roman Empire, the Neder-Germaanse Limes (Lower German Limes). The coins were discovered at a depth of less than 30 cm below the surface.

The archaeological fieldwork was carried out to understand why the coins were in this location. Firstly, a coring survey was carried out to map the landscape context of the find. Samples were taken for archaeobotanical, geochemical and C14 analysis. Secondly, an excavation was carried out at the site of the find to collect further finds and document possible features. This was done in collaboration with the finders, who now also took care of the metal detection.

The analysis is still in progress, but the first results can be shared here. More coins have been collected, twenty-two denarii and one aureus, which fit seamlessly with the other coins in terms of dating. These come from the ploughzone and a recent arable layer underneath. The coins have probably been removed from their original context and dispersed by agricultural activities in (sub-)recent times. No other finds or archaeological features were recovered. Nevertheless, the composition and dating of the coins point towards a single and deliberate deposition. It is suspected that the coins were originally buried in a shallow pit and stored in organic material, such as a textile cloth or leather pouch, which has since decayed. This pit would have been dug in a fairly soggy area and close to a water-bearing channel. In Roman times, this zone was unsuitable for habitation and agriculture.

Dating of the Coins and Historical Context of the Conquest of Britain

The Bunnik hoard consists of 44 gold staters attributed to the British king Cunobelinus, 72 gold aurei, and 288 silver denarii, dating from approximately 200 BCE to 47 CE. The most recent coins in the hoard, struck in 46-47 CE, include silver denarii and two die-matched gold aurei of the emperor Claudius. Their identical dies suggest that they were part of a single batch, probably distributed as military pay. The pristine condition of these coins further implies that they were buried shortly after issue, having seen no circulation due to their use as military payment.

The gold staters bear the name of king Cunobelinus (c. 5-40 CE) of the Catuvellauni tribe, considered to be the longest-reigning British Celtic king. Four of these staters are regarded as posthumous issues, likely struck by Cunobelinus’s successors, Togodumnus and/or Caratacus, around 43 CE. Analysis by John Sills reveals a wide chronological range for these Celtic staters, suggesting that the hoard was not selectively assembled based on quality, gold content, or weight. This suggests that the staters were removed from circulation in a single event, consistent with the hypothesis that the hoard represents spoils from the early Roman conquest of Britain under Aulus Plautius (43-47 CE). The staters may have been distributed to the army as a donativum, a monetary gift often given to soldiers in the Roman army as a reward after a successful campaign.

In 43 CE, Emperor Claudius appointed Aulus Plautius to lead the Roman invasion of Britain, officially intended to reinstate Verica, the exiled king of the Atrebates and a Roman ally. Verica had been overthrown probably by Togodumnus and Caratacus, sons and successors of King Cunobelinus. Plautius commanded four legions supported by approximately 20,000 auxiliary troops, including Batavians from what is now the Netherlands. Plautius first defeated Caratacus on the River Medway, then Togodumnus on the River Thames – in both cases with the help of the Batavian forces, who were able to swim across these rivers in full body armour. After consolidating his position, Plautius sent a message to Claudius, informing him that he could make the crossing. Claudius did so in style – much like Julius Caesar in 55–54 BCE – with war elephants and heavy artillery. He then secured the territory as the Roman province of Britannia and appointed Plautius as its first governor. Later, Plautius celebrated a triumph in Rome alongside Emperor Claudius.

The composition of the hoard strongly suggests a connection to the conquest of Britain. Based on the location of the find site within the Limes region, a military connection is also likely. This region is known as the Lower German Limes, the northern border of the Roman Empire, which runs through the Netherlands from Zevenaar in the province of Gelderland, via the province of Utrecht, to Katwijk in the province of Zuid-Holland. The Lower German Limes played a crucial role in communication and served as a key logistical hub for the conquest of Britain.

Analysis and comparison

The only other stater of Cunobelinus known from the Netherlands was found on the beach near Katwijk. It is pierced and was probably used as a pendant. The Bunnik hoard is the first coin find in the Netherlands (and mainland Europe) to contain both British and Roman coins. This type of mixed hoard is uncommon in Britain as well.

The only comparable find is the 2019 Helmingham Hall hoard in Suffolk, England, which also included silver Roman denarii, gold aurei, and British gold staters of King Cunobelinus. This find was also discovered by metal detectorists, followed by archaeological research, which has yet to be published. Both hoards feature coins dating from circa 200 BCE to 46-47 CE, including a denarius of King Juba I of Numidia. And although the number of Republican coins in the Helmingham Hall hoard is significantly higher compared to the Bunnik hoard, the percentage relative to the total is similar. The table below compares the Helmingham Hall and Bunnik hoards in terms of both numbers and percentages.

While the Helmingham Hall hoard contains more silver coins from the Republic, the Bunnik hoard has a higher proportion of gold, representing greater purchasing power. In the first century, a Roman army soldier earned 225 denarii annually (soldier of an auxiliary: 187.5), paid in three stipendia, in January, May and September. A gold aureus was worth 25 denarii. The exact value of a gold stater is unknown, but we can make an estimation for the value in denarii, when we relate the weight and gold content of both staters and aurei in the Bunnik hoard. The 72 aurei, with a gold content of c. 99% have an average weight of 7.71 g. The gold content of the staters is c. 41.5%.

And the average weight of the 42 staters is 5.41 g and 1.33 g for the 2 quarter staters. The value in denarii per coin type in the Bunnik hoard is listed below:

With these values in denarii per coin type, we can now calculate the total number of denarii in both coin hoards and the equivalent annual salaries of a soldier:

Interpretation

The exact reason for the concealment of the Bunnik hoard cannot be determined. However, some insights can be drawn from the composition and size of the find as well as the location it was buried. The hoard, as described earlier, consists of three parts: British gold staters, Roman gold aurei, and silver denarii. The gold staters may represent war booty associated with the initial conquest of Britannia under Aulus Plautius (43-47 CE), possibly distributed as a donativum – a monetary gift commonly given to Roman soldiers as a reward following a successful campaign. The Roman aurei and silver denarii are likely linked to military payments (stipendia).

Unlike the Helmingham Hall hoard, the relatively large size of the Bunnik hoard – both in the number of gold coins and the equivalent of annual salaries – suggests that it was more likely a payment to a high-ranking official (such as a centurion) rather than a soldier. At the time, a centurion earned a significantly higher annual salary: 3,375 denarii (centurion of an auxilia cohort: 937.5 denarii).

We also cannot rule out the possibility that the Bunnik hoard belonged to a group of individuals rather than a single person – for instance, a group of soldiers and their families. Another possible reason for the concealing of the hoard could be a (religious) offering made by an individual or a group, perhaps as an expression of gratitude for a safe return after the first phase of the Roman conquest of Britain (43-47 CE). 

The composition, dating and limited spatial distribution of the coins probably indicate a single and deliberate deposition. The find location is striking: far from habitation and in a wet zone along a channel. Based on this, an interpretation as a ritual deposition or a find hidden for some other reason and no longer retrieved is the most obvious.

Conclusion and Future Publications

The Bunnik Hoard 2023, like the Helmingham Hall Hoard 2019, likely relates to the early phase of Britain’s conquest under Aulus Plautius (43-47 CE). The dates of the most recent coins in the hoard, both the British staters (c. 43 CE) and the Roman coins (46-47 CE), also point to this. Finally, the occurrence of die-identical aurei and the findspot in the context of the northern border of the Roman Empire (the Lower Germanic Limes) also support this conclusion.

While the exact reason for burial remains uncertain, the hoard’s size – equivalent to nearly 11 years' salary for a soldier – suggests it was likely associated with a high-ranking official, such as a centurion, or possibly a group of soldiers and their families, rather than the savings of a single soldier. It may have been a ritual deposition, the wealth of a centurion, or coins hidden for some other reason and never retrieved. Be that as it may, the Bunnik hoard still leaves room for the imagination.

The hoard is registered with PAN (Portable Antiquities of the Netherlands, hoard ID: PAN-S-280), NUMIS (Numismatic Information System of De Nederlandsche Bank, hoard ID: NUMIS-1175412), and the Coin Hoards of the Roman Empire project (Ashmolean Museum and Oxford Roman Economy Project, hoard ID: CHRE-21255).

Contact and Acknowledgments

For questions about this text, please contact

Anton Cruysheer: a.cruysheer@landschaperfgoedutrecht.nl

or

Tessa de Groot: t.de.groot@cultureelerfgoed.nl