WTF happened?

Well, after a couple of weeks’ consideration, just another Steven Bradbury phenomenon, something else altogether, or perhaps just a re-run in more virulent form of the fake news and social media negative campaigns seen in the 2016 Presidential election.

The range of explanations is wide. The first, and most obvious, is that it’s hard for someone like Bill Shorten, who is neither liked nor trusted to sell a big agenda. The second – the success yet again of negative campaigning. In this case part of the evidence for this explanation is reminiscent of the Bush campaign, magnified here by social media,  in which the Democrats were accused of planning a ‘death tax’ – a US focus group tested concept the Liberals stole holus bolus. This influenced people who would never be affected by inheritance taxes even if they were reintroduced. The same applied with franking credit and tax policies where the 20% who would have been affected were successful in terrifying the 80% who wouldn’t have been.

But perhaps the most significant factor was the amplification of negative online campaigning which has been oft-mentioned but little studied. However, a lot of light has been thrown on the subject by a paper (The New Age of Propaganda: Understanding Influence Operations in the Digital Age) in the Lowy Institute newsletter The Interpreter by Zac Rogers and Maryanne Kelton, Emily Bienvenue from the Jeff Bleich Centre for the US Alliance in Digital Technology, Security and Governance at Flinders University.

They write: “Influence operations in the digital age are not merely propaganda with new tools. They represent an evolved form of manipulation which present actors with endless possibilities – both benign and malignant. While the origins of this new form are semi-accidental, it has nonetheless opened up opportunities for the manipulation and exploitation of human beings that were previously inaccessible. Now conducted across the whole of society, we are only at the beginning of a new era of population-centric competition.

“With regard to propaganda, the fundamental distinction between the old and the new lies in the difference between participatory and passive forms of information consumption. In the digital age, when people post, comment, like, share, and search, we are participating in information processing and knowledge formation in a way we didn’t before. We are actors in our own information consumption, and this represents a subtle but important shift.”

They contrast the way we consume traditional media formats – largely passively – to the way in which social media engage, in the way students learn by doing, as participants in “a type of cognitive investment. People engage differently when they are themselves participants in the narrative. They experience the narrative as it’s developing – it becomes part of their lived experience. Posting, commenting, tagging and sharing – they are no longer at arm’s length from the subject matter.”

The approach resonates with what we know about how German listeners reacted to radio propaganda in the 1930s when radio was a much more participatory media by virtue of families listening around one family communication device and participating in the broadcast through discussion.

The Rogers, Kelton, Bienvenue publication coincided with a project involving live coverage of online campaigning in the recent EU elections. Andy Patel – writing on the F-Secure website said: “I recently worked with investigative journalists from Yle (a Finnish publication), attempting to uncover disinformation on social media around the May 2019 European elections. This work was also part of F-Secure’s participation in the SHERPA project which involves developing an understanding of adversarial attacks against machine learning systems – in this case, recommendation systems on social networks. My contribution to the project was to analyse Twitter traffic for manipulation and poisoning attacks.”

The team identified “suspicious tweets (which) originated from two accounts – NewsCompact and PartisanDE. These accounts link to each other in their descriptions. For instance, PartisanDE identifies itself as the CEO of NewsCompact. The language used in tweets published by these accounts is clearly not written by native English speakers. Both accounts receive a great deal of engagement from far-right Twitter profiles (many of which I’m familiar with from past research). They share a great deal of politically inflammatory content, racist content, and mistruths. The accounts also share many URLs to sources such as VoiceOfEurope, RT, Sputnik, and some other non-authoritative ‘news’ domains.”

What is clear from both the Flinders University and the Yle research is that negative campaigning – which has always been with us – has taken on new, insidious and more effective forms. In the recent Indian election Prime Minister Modi, spent – according to The Economist (25 May 2019) – 51% of his speeches attacking opponents, 18% talking about national security and only the remaining bit about positive policies. One can imagine how this was amplified by social media and ponder what a similar analysis of Scott Morrison’s campaign would reveal and how that was magnified by the many negative social media campaigns against Labor policies and imagined threats to Australia.

All the research indicates that it is unsurprising that looking at mainstream media is not the best way to find out what’s going on in elections. Indeed, shortly after being caught out by the Australian election result various Australian media outlets were touting the huge success of populist parties in the EU elections – largely on the basis of Nigel Farage’s UK success. Yes there was an upsurge in the populist vote although EU-wide social democrats, liberals, mainstream conservatives and Greens captured 75% of seats. If any one party enjoyed great success it was the Greens who came second in the European wide elections.

And as for the consensus on opinion polling in Australia – the explanations for their ‘failure’ are also not hard to find. Movements within the statistical margins of error; judgment errors in allocating preference flows; too much reliance on landline interviews; lower than expected turnout – indeed, it was surprising they got so close.