From Kilwa to Australia, almost certainly not direct

By way of a medievalist break from all the strike woe, some time ago now I stubbed a note to talk about a rather surprising coin find. Since this means dealing with this blog’s most controversial subject ever, medieval knowledge of Australasia, I tread with care, but I also discover that not only have several others trod there before me, but one of them was Dr Beachcombing, once proudly included on the blogroll here, so it’s probably going to be OK. These are (some of) the controversial objects.

Copper-alloy fals of Sulaiman ibn al-Hasan struck at Kilwa Kiwisani 1294-1308

Copper-alloy fals of Sulaiman ibn al-Hasan struck at Kilwa Kiwisani 1294-1308, found on Wessel Island, Australia, 1944. Image released to the press by the Powerhouse Museum, Sydney on August 22, 2013, here reproduced from AFP, “How 900-year-old African coins found in Australia may finally solve the mystery of who arrived Down Under first” in Daily Mail, Science & Technology Section, 22nd August 2013, online here, but visible on the Powerhouse Museum’s site with better captioning here

Copper-alloy fals of Sulaiman ibn al-Hasan struck at Kilwa 1294-1308

Another fals of Sulaiman ibn al-Hasan from the Wessel Island find, reproduced this time from Ian McIntosh, “Life and Death on the Wessel Islands: The Case of Australia’s Mysterious African Coin Cache” in Australian Folklore no. 27 (Riga 2012), pp. 9–26, online here, at p. 14

These were found in 1944, in the Wessel Islands, which lie off Australia’s Arnhem Land, in its north-east, about 400 km south of Indonesian Papua (which becomes important). They were not then examined until 1979, and it’s possible, just possible, that another has been found on another island about 100 km away since then.1 But when they were examined, they caused a minor stir because they were struck in a place I’ve written about here before, the Sultanate of Kilwa in modern-day Tanzania, somewhere between 1150 and 1330.2 Now, Kilwan copper coins are by no means rare and they’re not really valuable, but context is everything: they are common in and around Kilwa, obviously, and are otherwise found in Zimbabwe now and then, and Oman once or twice (itself a thing to explain, but Kilwan gold coin, which seems to have been the international coinage of the Sultanate, is only found in Arabia so that some of the small stuff should have gone there too is actually one of the least weird hypotheses about these coins).3 Otherwise, they didn’t travel, and that they turn up so very far out of that orbit takes some explaining.

The Wessel Islands are the long thin offshore strand at top right of this map, at least as it loads for me

Now, the person who pointed me at the story in 2018 is very interested in global contacts in the Middle Ages and was, I think, rather hoping that someone who knew the coins might tell a story from them that would serve that. Alas for that person, I cannot, and the person who has mainly investigated the find, an anthropologist by the name of Ian S. McIntosh, has not wanted to either. This didn’t stop journalists trying to drum up excitement about possible medieval contact with Australia, of course, and even McIntosh is interested in, “a more accurate portrayal of Australian history that is currently allowed in textbooks,” suggesting that he would not mind a radical new finding.4 Nonetheless, he is not (or was not in 2012 and 2013, when for some reason his work hit the news) wedded to any particular theory; in fact he has five hypotheses for how the coins might have got there and of them, the idea of medieval contact between Kilwa and Australia is the least likely.5

Ruins of Kilwa Kiwisani, Tanzania

Ruins of Kilwa Kiwisani, reproduced from Teo Kermeliotis, “Ancient African coins that could change history of Australia” in CNN, Inside Africa Section, 27th June 2013, online here

Now it is worth saying that it is not impossible, albeit there is no evidence, given that there was a ninth-century dhow wrecked off Sumatra and an established Islamic presence in Indonesia by 1400, that at some medieval point a Kilwa trading ship did wind up way way off course and make contact with what is, after all, a really big land mass (even if the Wessel Islands are a very small offshore part of it).6 But that is almost certainly not the story these coins tell, for two reasons. Firstly, there is a much better window after 1505, when the Portuguese sacked Kilwa, and 1514-15 when they set up in Timor in Indonesia, thus opening up a definite route between Kilwa and Indonesia that was in use by Islamic traders until the nineteenth century.7 Secondly, and much more importantly, the coins were found with a number of Dutch duits struck from 1690 to 1784.

Copper duit of the Dutch East India Company struck at Gelderland in 1690

Copper duit of the Dutch East India Company struck at Gelderland in 1690, found in the Wessel Islands in 1944 and now Sydney, Powerhouse Museum, N21359-1, image from “Ancient Coins in ArnhemLand” in Past Masters, online here

So at this point we can start to spin hypotheses. First, the coins were actually deposited all together. In that case they’d pretty much have to have been gathered together by an eighteenth-century Dutch trader, probably buying sea cucumber which was apparently a big export of the area then. Why he’d have paid in these is less clear; perhaps they might have been given as talismans to the Yolngu people of the islands.8 However, this does also involve explaining why such a Dutchman should have had, but then got rid of, some 600-year-old fulūs from Kilwa, as well as why his own small change went back a century. Apparently Dutch trade networks encompassed the Swahili coast and the Wessel Islands (which do after all lie off a bit of Australia with a Netherlands place-name), so it’s not impossible, but we might at least want to consider alternatives, especially since the coins, while hardly fresh, don’t really look like they’d been circulating for six centuries. The hypothesis that the coins were gathered separately, however, involves them being collected in the Islands. The advantage of this is that they don’t all have to have arrived there at the same time; the disadvantage is that it means someone thought they were worth keeping and then someone else did and then someone else did, combining Kilwa coins with Dutch ones at some point, but eventually nonetheless someone thought that rather than keep them any more they should all be buried. Also, we then have several separate acquisition moments to hypothesize, including possibly Muslim traders from Makassar as well as later Dutchmen, and Occam’s Razor would therefore favour the later deposition. And after that, the hypotheses only get more complex. The one of these that the Powerhouse Museum in Sydney, where these coins now reside, appears to be that the coins are all individual losses and found together ‘over a period of time’; this doesn’t appear to be how any of the other reports of their discovery have it and I don’t know whether they have information the web doesn’t, but if it’s so I’d still rather think of it as a single hoard found in parts, just because there is so much more to explain if it wasn’t.9

Jensen Bay on Djinjan Island, Australia

Jensen Bay on Djinjan Island, Australia, a possible find-site for the hoard, image from “Ancient Coins in ArnhemLand” again

Now, Professor McIntosh himself has some ingenious possible explanations of the deposition of this hoard, which rest on really good local knowledge of the islands and the Yolngu’s own stories. I’m not qualified to judge them, and they’re on the open web, so I’d invite you to go and have a look yourself; at least one of the possibilities involves shipwreck, plague, taboo and, again, sea cucumbers, which is a pretty rare combination I’d say.10 But the message I’d take from it as medievalist is that there is no explanation of the Kilwa coins in that hoard that doesn’t involve something pretty unlikely having happened. A Dutchman with a pocketful of medieval fulūs is unlikely; a multi-generation coin collection in the Wessel Islands seems unlikely to me as well, though I don’t know the area or people at all so could be very wrong; a succession of drop-offs of very very few low-value coins with no precious metal content, which were not then either melted down or pierced for use as ornaments, also seems like a lot of unlikelihoods; and you might therefore argue that a Kilwa ship fetching up in the Wessel Islands in some disastrous fashion isn’t any more unlikely. But to that I would say, firstly, that that still involves explaining why they wound up buried along with the Dutch coins, and moves the necessary six centuries of circulation and conservation into the Wessel Islands, a bit of a bigger ask. Secondly, the medieval deposition theory has the weakness that there is literally no other evidence to support it, whereas all the other hypotheses, however stretched, can be attached to things which we do know to have happened. And this is why I couldn’t give my colleague the story they may have wanted. But whatever the story of these coins actually was, which Professor McIntosh may yet discover I suppose, it must have been a pretty wild one! And with that thought I leave it to you.


1. Kylie Stevenson, “‘It could change everything’: coin found off northern Australia may be from pre-1400 Africa” in The Guardian 11th May 2019, online here; it is quite important to mention that at the end of some heavy lab work, the coin remained unidentified!

2. These dates come from “Collection of coins, photograph and documentation” in Powerhouse Collection, online here, whose precise identifications of the coins suggest that they have got more information than the 2013 publications on which this post mainly relies.

3. For the numismatics of Kilwa you still have to see G. S. P. Freeman-Grenville, “Numismatic Evidence for Chronology at Kilwa” in Numismatic Chronicle 7th Series Vol. 18 (London 1978), pp. 191–196, but cf. now Jeffrey Fleisher and Stephanie Wynne-Jones, “Kilwa-type coins from Songo Mnara, Tanzania: New Finds and Chronological Implications” in Numismatic Chronicle Vol. 170 (London 2010), pp. 494–506.

4. See Ian McIntosh, “Life and Death on the Wessel Islands: The Case of Australia’s Mysterious African Coin Cache” in Australian Folklore no. 27 (Riga 2012), pp. 9–26, which is my main source for this post. The quote is however from Teo Kermeliotis, “Ancient African coins that could change history of Australia” in CNN, Inside Africa Section, 27th June 2013, online here. McIntosh also has another relevant article, “The Ancient African Coins of Arnhem Land” in Australasian Science May 2014 (Melbourne 2014), pp. 19-21, online here, but I only found this as I was compiling these notes so haven’t incorporated it. It is available with a wealth of other relevant documentation from “Ancient Coins in ArnhemLand” in Past Masters, online here.

5. “Five separate hypotheses” in Kermeliotis, “Ancient African Coins”, just before an expedition to find out more; “a whole series of hypotheses” in AFP, “How 900-year-old African coins found in Australia may finally solve the mystery of who arrived Down Under first” in Daily Mail, Science & Technology Section, 22nd August 2013, online here, after the expedition, which I think must have been that documented in “Ancient Coins in ArnhemLand”. McIntosh, “Life and Death”, isn’t quite so methodical in its arrangement as to number its overlapping hypotheses.

6. Though as McIntosh points out, ibid. p. 10, they are kind of a “catching mitt” for anything being carried through that stretch of sea. For the dhow see Michael Flecker, “A ninth-century AD Arab or Indian shipwreck in Indonesia: first evidence for direct trade with China” in World Archaeology Vol. 33 (Abingdon 2001), pp. 335–354, and for late medieval Indonesia Barbara Watson Andaya and Leonard Y. Andaya, A History of Early Modern Southeast Asia, 1400-1830 (Cambridge 2015), pp. 82-129.

7. McIntosh, “Life and Death”, pp. 13-14.

8. Ibid., pp. 15-16, citing the diaries of one Matthew Flinders; the relevant extract is online here thanks to “Ancient Coins in ArnhemLand” again.

9. “Collection of coins”.

10. McIntosh, “Life and Death”, pp. 21-23.

2 responses to “From Kilwa to Australia, almost certainly not direct

  1. One of my funniest discoveries on yer internet was stumbling across a madman who attributed all mysterious international exploration, discovery and trade to “the Celts” by whom he meant the Irish. More recently I’ve seen remarks by someone who attributes everything interesting and ancient in Europe to the Serbs.

    So the explanation for these finds may well lie with a joint Serbo-Celtic ship which … (And afterwards discovered NZ.)

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.