New coins on the auction block (2): Arab-Byzantine gold

I have been looking at Arab-Byzantine gold coins. The term “Arab-Byzantine” is used to describe a series of gold and copper coin types issued in the former Byzantine territories of Syria after the Islamic conquests of the 630s but before the introduction of a properly Islamic coinage by the Umayyad caliph Abd al-Malik in 696-697 AD. Although the terminology differs, three chronologically-successive coin types are recognised: (1) Pseudo-Byzantine (PB), (2) Umayyad Imperial Image (II), and (3) Standing Caliph. Most of the coins known today are copper, although some are gold. Gold coins are exceedingly rare and fetch correspondingly high prices when sold at auction.

Over the past 10 years, 12 previously unrecorded Pseudo-Byzantine and Imperial Image gold coins have appeared for auction. They are tabulated below. The prices are extraordinary, particularly for the Imperial Image coins. Arab-Byzantine gold coins were designed in imitation of contemporary Byzantine solidi, which typically sell at auction for prices under a thousand USD, orders of magnitude less than their Arab-Byzantine counterparts.

CoinAuction dateAuctionTypePrice (USD)
130 Nov 2010NGSA, 6, lot 285IINot known
225 Apr 2012Baldwin’s, Islamic 19, lot 7II177,283
226 Apr 2018Morton & Eden, 92, lot 12II184,406
322 Apr 2013Morton & Eden, 63, lot 6PB76,124
422 Apr 2013Morton & Eden, 63, lot 7PB60,899
59 May 2013Baldwin’s, Islamic 24, lot 3999PB49,804
524 Nov 2014NGSA, 8, lot 226PBNot known
69 May 2013Baldwin’s, Islamic 24, lot 4000II249,019
710 Sep 2015Album, 23, lot 68PB50,000
810 Jan 2017CNG, Triton XX, lot 1137PB42,500
822 Oct 2020Morton & Eden, 107, lot 1PBNot sold
914 Jan 2018Baldwin’s, New York 14, lot 21PB95,000
108 Jan 2019CNG, Triton XXII, lot 1226PB110,000
118 Jan 2019NYS, XLV, lot 314PB65,000
1218 Nov 2019NGSA, 12, lot 158PB50,461

Provenance information is scant. The catalogue entries for the second sale of coins offered twice usually reference the first sale, but nothing before that. The catalogue description for the Imperial Image coin sold by NGSA in November 2010 states that it was the first such coin to appear for auction in 24 years, confirming the rarity of Imperial Image coins and explaining the high price achieved by the NGSA coin and two similar ones in 2012 and 2013. So, no coins for 24 years, and then 3 in a space of 4 years with nothing said about provenance. The two Pseudo-Byzantine coins sold in the Baldwin’s May 2013 Islamic Auction were said to be from the Horus Collection, “formed over the past 35 years”, but nothing more. Perhaps the coins had been acquired decades ago, or perhaps only months before the sale. There is no way of knowing. In January 2000, the Imperial Image coin sold as lot 4000 at the May 2013 Baldwin’s sale was exhibited in Abu Dhabi as part of a large private collection of Islamic coins.

Imperial Image coin

Arab-Byzantine gold coins are generally considered to have circulated in the area of what is today Syria and the immediately adjacent territories of neighbouring countries. None of the coins sold since 2010 has a recorded find spot, and for Arab-Byzantine gold coins generally only one has any kind of information about where it was found. It is a Pseudo-Byzantine coin acquired by Paul Bedoukian in 1965 as part of the “Daphne” hoard of Byzantine gold solidi said to have been found 5 km from Antakya (ancient Antioch) in what is today the southern extremity of Turkey (Miles 1967: 208; Metcalf 1980). This hoard was acquired on the market, however, so its integrity and locational information are inherently unreliable.

Let us remember what was happening in Syria while these unprovenanced coins were being sold. Heavy looting of archaeological sites started being reported in 2012 after the onset of civil conflict in 2011, with some of the proceeds going to support armed groups. By late 2015 at the latest, it was widely known that there was an illegal trade in ancient coins out of Syria and that it was an important source of revenue for Salafist-jihadist groups such as Da’esh. Thus, the provenance of any coin seemingly fresh to the market after 2011 should have been open to serious scrutiny before being offered for sale. Perhaps the coins were all researched, but if they were it is a shame that nothing was ever published. So, just where exactly did these new coins originate? A collector’s coin cabinet in Switzerland or a looter’s hole in the ground in Syria? Again, there is no way of knowing. But the fact that the coins did sell without any satisfactory account of provenance suggests that the buyers just didn’t care, which is disconcerting.

Provenance aside, the auction prices offer some important insights into the potential profitability of the coin trade for Da’esh and other armed non-state actors. Let us imagine a hypothetical Arab-Byzantine coin found late 2015 by looters in Syria and worth something like 100,000 USD on the European market. The looters might not recognise the coin for what it is, but the local dealer most probably would and be aware of its potential value. Say the local dealer manages to sell it for 25,000 USD and then pays 20 per cent tax on the transaction to Da’esh. That would be 5,000 USD for a single coin. If the dealer manages to sell it for more, the tax take would increase accordingly. A few more similar coins and the numbers start to add up …


Metcalf, William E. 1980. Three seventh-century Byzantine gold hoards, Museum Notes (American Numismatic Society) 25: 87-108.

Miles, George C. 1967. The earliest Arab gold coinage, Museum Notes (American Numismatic Society) 13: 205-229.