Moral panic over ‘Russian-linked’ accounts in social media has departed the realm of logic

(Irussianality) – As, I am sure, all of you know, a proper scientific experiment will have a ‘control group’. Say I have a new cancer drug. I can’t tell if it’s actually any good just by testing it. I need something else to compare it to. It’s only by means of the comparison that my results have any meaning. To see if the ‘independent variable’ is of any significance, you have to consider other possible factors which might be affecting the result. In short, you can’t treat a single phenomenon in isolation from everything else.

Bear this in mind, as we’ll come back to it later. But for now, let’s switch track and turn to the matter of ‘Russian interference’ in US politics. What have we learnt to date?

What we’ve learnt is that some ‘Russia-linked accounts’ posted messages about US politics, and paid for advertisements related to US politics, on social media such as Twitter and Facebook. Some of these messages were anti-Clinton and pro-Trump (along the lines of ‘a vote for Hillary is a vote for Satan’), but some were anti-Trump, and some were about completely different things altogether (Black Lives Matter and the like). For a sample, take a look here.

We’ve also learnt that an account is deemed ‘Russia-linked’ if it features even one of the following criteria: it was created in Russia; registered via a Russian phone carrier or email account; uses Cyrillic characters; the user regularly uses the Russian language; and the user has logged in from any  Russian IP address, even once. I’ve logged in to this site in Russia, so according to this definition you are reading a ‘Russia-linked’ blog. That means that if I make any comments about US politics, they will be added to the list of evidence of ‘interference’ by the Russian government.

Clearly, this is all a bit silly. But, let’s not worry about that for the moment. Let’s accept that some of the ‘Russia-linked accounts’ are indeed Russian, though we can’t tell that any of them are actually linked to the Russian government, and let’s accept that Russians are posting things about US politics. Does that amount to ‘interference’? And does it show that Russians are particularly noteworthy interferers, so noteworthy as to justify a vast witch-hunt?

Now, this is where the matter of comparison comes into play. Russians are posting stuff about US politics. But what about everybody else? Let’s face it, Russians are hardly likely to be the only ones. US politics interests people just about everywhere, and some of them no doubt have some strong views on it and may even have generated some commentary or memes or something else which they’ve posted on Facebook or Twitter. If you’re going to say that ‘Russian interference’ is especially prominent and dangerous, you need something to compare it to. For instance, you might compare it to the complete total of all social media users. Are Russians posting substantially more about US politics than social media users as a whole? Alternatively, you could look at individual countries. What about Canada-linked users; Britain-linked users; French-linked users; Mexican-linked users; whatever? Have any of them posted stuff about US politics, bought political advertisements, and the like? And if so, do they do it more or less than Russia-linked users, in proportion to their numbers.

This matters, because if you were to do such a comparison and discover that, say, Canadian users were generating very similar stuff on Facebook and Twitter, and doing just as much compared to their overall numbers, then you’d have to start investigating ‘Canadian interference’. Or if you found the same with Brits, Germans, French, Mexicans, Venezuelans, whatever, you’d have to investigate British, German, French, Mexican, Venezuelan, etc interference too. And then, it would become obvious that Russian interference’ isn’t particularly abnormal.

Maybe it is. Maybe, ‘Russia-linked accounts’ have generated far more of the sort of stuff under investigation than accounts linked to other countries. But then again, maybe not. To date, I haven’t read anything which suggests that anybody has carried out the research to show which is the case. If that is true (and please show me if I’m missing something), then all the findings about Russian interference are utterly meaningless, as they lack any comparison. This is basic scientific method. Am I the only person to have thought of this?