Last weekend I was a bit-part player in a lengthy Twitter back-and-forth about a coin, culture, and the communication of complex concepts in no more than 140 characters at a time. In many ways it was a debate about nothing of any profound consequence, unless what Icenian coins can tell us about parts of East Anglia in the Iron Age is something that really keeps you awake at night. But in some regards it struck me as what 2016’s been all about: the misrepresentation of information by people in a position of influence that is (or runs the risk of being) taken as the truth because it seems to form an authoritative opinion and that is a good deal easier to engage with and accept than a lengthier but more accurate account of the same information.
Let me say now, in case it isn’t as clear to anyone reading this as it is to me writing this, that this post is not in any way, shape or form accusing its primary named subject of anything malevolent or of deliberately misleading people for the purposes of a hidden agenda. Nor is it a redux of the classic academia vs. non-academia debate – I’m a doctoral-level student who’s also been accused (wrongly as it happens) of indulging in academia-bashing during one of my periods out of higher education, so I’m acutely aware of the scant value of the arguments one way or the other. Instead, what I hope to get across is that last weekend really brought home to me how in this supposedly post-factual age, in which “experts” are disparaged for their inaccurate predictions and perceived self-interest, it’s now contingent on those who work on complex issues to counter the narratives offered by certain people touting simple but incorrect explanations/solutions by making their own short, sharp and above all accurate contributions to debates.
It all began with a tweet in which Twitter powerhouse and from what I gather highly successful writer of books on historical topics Tom Holland announced that he’d bought a particularly interesting Iron Age coin. (Let’s skirt questions of whether the coin was legally discovered and was sold with full and accurate information about its provenance and say that we’ll accept that it was.) There followed soon after two further tweets: one positing that the apparently lupine iconography on the reverse of the coin might be related to the Norse myth of Fenrir, and a second in which it was asserted that many Icenian coins imply that the people ‘may well have been culturally Germanic, not Celtic, speaking proto-English’.
Friends, acquaintances and academics whose work I admire took him to task for various shortcomings around the use of the word ‘culturally’ and whether the iconography of wolf-with-associated-figures was not more likely to relate to the legend of Romulus and Remus. As you can imagine, the debate shot off in all sorts of directions, got a bit heated at times and also encouraged at least one online crazy to briefly enter the fray with some bullshit “contribution”. And that was just the threads of replies I was a participant in, as elsewhere Mr Holland was getting a much easier time of it, allowing him to say some things that were even more sketchy and worthy of challenge (and a few frosty one-word answers for good measure):
Where I weighed in was in order to highlight the problem – and I stand by saying it’s a problem, rather than an issue or whatever as a very positive-minded former colleague of mine once insisted we spin these things as – of oversimplifying and overstating what a scholarly source actually says, and how this wasn’t simply because Twitter’s famous 140-character limit prevents the articulation of nuanced interpretations. The above tweets demonstrate both Mr Holland’s extrapolation of one particular point drawn from what I’ll show below to be a methodical (if controversial) piece of research while simultaneously omitting the finer points of what it has to say about the relevant chronology, as well as the mistaken conclusions some of his followers arrived at as a result. Tongue partly in cheek, I offered the following alternative (but still not problem-free) wording for the first of his tweets shown above to square his interpretation with what the authors of his more critical replies were pointing out – unfortunately it didn’t prompt a response from Mr Holland (I’m sure things would have turned out differently if there was an edit tweet function):
Frankly, none of the above would be such a problem if the tweets came from a weekend warrior tweeting to an audience of his best mate, uncle and some people who support the same sports team, but Mr Holland’s a big deal – he’s got a blue tick, and is followed by over 60,000 Twitter accounts. So if he distorts things or gets things wrong in what he posts, that’s seen by a lot of people. And if I was working on Early Anglian Norfolk, I’d be pretty hacked off right now that for the next couple of years the first question I could most likely expect after giving a paper on my research would be “But didn’t Tom Holland say that all this happened before the Romans came?”
At the time I took the first of the above screenshots, mid-Abu Dhabi Grand Prix on the Sunday afternoon, the above tweet had been liked 38 times and retweeted 10 times, and this remains true at the time of writing. The original tweet, meanwhile, has amassed 157 likes and 41 retweets. Even though from personal experience I know that every retweet doesn’t come with a like, if we focus on the latter then that’s a considerable audience who have presumably actively engaged with the content of the tweet(s) and been moved to register their agreement with it by means of liking it. Moreover, the retweets mean an audience reach in excess of Mr Holland’s already substantial following. Which is why it’s pretty important that what publicly-respected figures, especially those who undertake to publish works of scholarly rigour, communicate in the media is accurate.
Now I don’t profess to be any kind of expert on Pre-Roman Iron Age East Anglia, but I did happen to know the source Mr Holland was using to make this claim about ‘Germanic’/’proto-English’ language. I also happened to know, thanks to a Facebook post, the means by which he most likely discovered that source (and unwittingly or otherwise the inspiration for his referencing of culture). A few months back, a friend posted a link to this short magazine article about a hoard of Icenian coins written by Chris Rudd, the numismatist from whom Mr Holland purchased his example. In it, Rudd mentions ‘the growing belief that Iceni culture was influenced by Germano-Belgic culture’, a position he backs up through reference to his earlier work hypothesising that ‘the wolverine imagery of Norfolk Wolf gold staters was probably related to Norse mythology’, plus another piece of published research offering ‘persuasive evidence of the Iceni’s linguistic links with ancient Germanic names […] formed over 400 years before Angles and Saxons settled in East Anglia’, as well as archaeological studies from the same collection arguing for complementary ‘cultural similarities’ between the Iceni and Germanic and Belgic peoples of the Late Iron Age.
The linguistic study in question is a 2011 chapter by Daphne Nash Briggs (no relation of mine, by the way) entitled ‘The language of inscriptions on Icenian coinage’, which was published in the BAR British Series volume The Iron Age in Northern East Anglia: New Work in the Land of the Iceni. By great fortune, its author has uploaded it to Academia for all to see (or maybe just members – I forget). It’s a really meaty bit of research, one that I’ve seen attract some comments on social media concerning the basis of some of its etymological analyses, but now’s not the time to deliberate on them. (UPDATE: One early comment responding to this piece on social media advised that none of the philological evidence nor Nash Briggs’ treatment of it is sufficient as to back up the study’s basic premise, so you may wish to read the following with this in mind.) The fundamental issue at hand is whether Dr Nash Briggs’ conclusions are equal to the emphasis that Mr Holland placed upon them – namely that they evidence a Germanic, not Celtic language spoken by the Iceni.
A rough and ready search of the Portable Antiquities Scheme database suggests that Dr Nash Briggs’ sample is exhaustive and thus the various terms she identifies are not a subset of a larger corpus. Consequently, in attempting to get a quick handle on the conclusions of her study, a good place to start is the table at the top of the second column on page 95 summarising the results of the linguistic analysis. It lists one name-type found on Icenian coin inscriptions (ANTEÐIO) as Celtic, another (-PRASTO) as a Latin loan-word, four as either Celtic or Germanic, and six as ‘Germanic?’ (with one of these, ESICO, having a question mark entered against it in the Celtic or Germanic column). These results in themselves run counter to Mr Holland’s claim for the Icenian vernacular to have been ‘Germanic, not Celtic’. Indeed, in Dr Nash Briggs’ own words (my emphasis added):
‘The presence in pre-Roman Britain of scraps of a philologically Germanic onomastic vocabulary does not of course mean that its users would necessarily have spoken the language from which they were conscripted […] nor would it necessarily have been the only language of Iron Age Norfolk, which it manifestly was not, since a standard form of Gaulish or Brittonic is recorded on some Icenian and almost all other inscribed ancient British coins and a majority of recorded ancient place-names, so was obviously familiar at least to all of ancient Britain’s political elites.’ (Nash Briggs 2011, 95-96)
The role of place-name evidence in this debate was another thing that interested me. The handful of Roman-era place-names recorded from the greater Icenian region are all of Brittonic rather than Germanic composition (Branodunum, Camboritum, Gariannum, Venta Icenorum: Rivet and Smith 1979, 274-75, 294, 366, 492 – to these can be added Epocuria, an Icenian ‘castellum‘ named in one of the recently-published Walbrook writing tablets (the relevant page reference eludes me sadly)). The record of these names during the Roman period, even as early as the first century CE in the case of Epocuria, does not compel them to be as old as coins produced in the middle years of the previous century (such as the Norfolk Wolf gold staters of the sort purchased and tweeted by Mr Holland), and Dr Nash Briggs’ dataset is derived from coins produced across a period of time perhaps in excess of 100 years, during which the Iceni experienced a number of episodes of stress, culminating in the establishment of Roman rule and with it the introduction of Imperial coinage.
In the latter part of her chapter, Dr Nash Briggs provides a fascinating discussion of the background to this period and the vectors by which various influences could have come into play. I harbour a secret crush for British Iron Age studies (don’t tell anyone), but some of what’s postulated in the absence of documentary testimony does terrify me when I compare it to what flies in early medieval studies. So it is with the argument (made by Chris Rudd in a work I must confess have not read) for ‘a series of increasingly unequal treaties’ made between the Iceni and their southerly neighbours the Catuvellauni, apparently during the first half of the 1st century CE, that are seen as the catalyst for the adoption of Catuvellaunian design motifs on many later Icenian coins, and more profoundly ‘the adoption of Brittonic and Latin vocabulary into the language of Icenian governance’ (Nash Briggs 2011, 98-99). Coin inscriptions would be an obvious indicator of this – but place-names? How much were they the products of the “language of governance” as opposed to local characterisations of local landscapes? Might we be better served rejecting such an all-or-nothing reading of a very restricted body of evidence?
Dr Nash Briggs’ concluding remark is measured in a way that Mr Holland’s tweet was not:
‘Unless and until a scrap of connected text is discovered to confirm its linguistic identity, evidence for use of a non-Latin, non-Brittonic, but possibly coastal West Germanic language in late prehistoric and Roman Norfolk must depend upon analysis of personal names, theonyms, and political titles, beginning with Late Iron-Age coin inscriptions.’ (Nash Briggs 2011, 99)
Important questions are posed by the study; I for one will be looking harder at the origin of the name-form DVRO (Nash Briggs 2011, 90-91) in the context of an article about the name of Dorking I’m aiming to write next year. Overall, Dr Nash Briggs makes a stronger case for bi-/multilingualism among the Iceni than for them speaking a wholly Germanic language in the first century BCE. Certainly, her ‘possibly’ is some semantic distance away from Mr Holland’s ‘may very well’, and her other statements similarly never go so far in their endorsement of the likelihood of a Germanic vernacular of the Iceni. Dr Nash Briggs’ paper has been viewed on Academia 166 times to date (two of which can be chalked up to me today), and I can’t imagine the printed version has received many more readers, whereas I’d bet the equivalent figure for Mr Holland’s tweet runs into the thousands. Undue credence has been lent to a well-argued but in some respects speculative theory, and this will serve to undermine the work being produced by those toiling away on research concerning Norfolk in the post-Roman period.
I’ve just seen Mr Holland’s been at in again, in long-form this time, in a Guardian article about cricket of all things, and how the concept of the Heptarchy might usefully be reemployed in a sporting context. No matter that the Heptarchy is not attested by the sorts of contemporary written sources you’d expect such a concept to feature in, and at best described a relatively short-lived reality – we’re talking cricket, dammit! Holland’s sources would appear to be the ambiguous testimony of late-ninth-century traveller Syrian Harun ibn Yahya (whose mention of ‘seven kings’ Holland transforms into ‘seven kingdoms’) and twelfth-century chronicler Henry of Huntingdon. Again, one one level this is completely trivial, but on another it’s the presentation of spurious information as solid fact to a wide audience (and to add insult to injury, Surrey gets made part of East Anglia!).
And so we return to the crux of the matter. Mr Holland’s clearly an accomplished and successful writer, and his most recent book seems to be his biography of King Æthelstan, published a matter of years after Prof Sarah Foot’s magisterial monograph on the same subject. While Prof Foot may not have any strong opinions on English cricket, she may be best placed to apply her tremendous expertise on early medieval subjects to a range of other contemporary issues, yet I can’t find her credited as an author of a single article on The Guardian website (there are a multitude of possible explanations for this, of course – perhaps she prefers The Times?). Fundamentally, how can serious researchers who are minded to make contributions to public debates apply the fruits of their research or their expertise in certain areas to compelling effect? How to turn long-form work into influential short-form content, be it as tweets, comment pieces, or content on other public platforms? And how to make that feeling of having weighed all sorts of possibilities to really get to the heart of a matter the thing people strive for over and above getting a half-baked opinion heard?
I’ve got to tread carefully as I don’t want to come across as advocating for those with expertise in Germanic linguistics or such like having the answers to all of the world’s ills. Give over! But… The contemporary world is an astonishingly complicated place – as 3 weeks travelling across the US back in the summer laid bare to me – yet the historical world was far from simple and so cannot be explained with easy, black-and-white explanations. It’s a huge leap from misrepresentations of Iron Age East Anglia to Brexit or Trump’s election victory, but those things are the end products of a longer trend. The nature of public discourse has changed, and having a command of all the facts doesn’t necessarily win the argument these days. This year we’ve seen and heard too many public figures denigrating experts while claiming “It’s simple really, the problem lies with …”, and however much a large chunk of society may think they need such purported explanations-cum-solutions, we can do so much better. This applies across the board.
There needs to be systemic change to return us to a situation where the facts matter, and presenting them correctly is more important than proffering partial analyses to get in there first or shout the loudest. This needs to start at the grassroots level and work its way upwards. Be good at what you do, what you post and publish, but also call out bullshit when you see it. If something’s wrong, or seems suspect, say so. Underline the reality of a situation, whether in terms of its complexity or the correct facts. But, crucially, do so in a way that’s quick and to the point (and don’t think I haven’t seen the irony in me stressing this after 18 paragraphs and well over 2,500 words). There’s a time and place for worthy read-this-30-tweet-thread-it-really-gets-to-the-nub-of-the-issue recommendations, but it smacks of the echo chamber and limited spaces within which the argument has already been won. We need to play the game of those who are influencing large numbers of people in sub-optimal ways, and win it by providing easily-digested chunks of accurate information upon which rounded opinions and conclusions can be founded. It’s not difficult to condense a lot of factual information into an easily-understood tweet or two with a bit of thought, especially if we know our material well enough. And more than ever, people deserve to be in full possession of the facts.
Nash Briggs, D., ‘The language of inscriptions on Icenian coinage’ in The Iron Age in Northern East Anglia: New Work in the Land of the Iceni, ed. by J. A. Davies, BAR British Series, 549 (Oxford: Archaeopress, 2011), 83-102
Rivet, A. L. F., and C. Smith, The Place-Names of Roman Britain (London: B. T. Batsford, 1979)
Tomlin, R. S. O., Roman London’s first voices: Writing tablets from the Bloomberg excavations, 2010–14, MOLA Monograph Series, 72 (London: Museum Of London Archaeology, 2016)