Internet searches

One of the most fascinating things about the Newsnight debacle of recent days was the fact that the complainant (Lord McAlpine) was never named by Newsnight. A case of, at the very least, mistaken identity that had already been corrected in 1998 (Guardian, 10/11/12) became a re-sparked rumour. Even though those who made the piece knew that Steve Messham had named McAlpine, and should have checked the story, they didn't name him (perhaps due to the lack of evidence or risk that something else was wrong). But the story came out all the same.

It appears that twitter was the mechanism for the name being revealed, and then the internet in general added a whole load of extra texture. I'm still not entirely sure how the name was added to the story. Ian Overton of the Bureau of Investigative Journalism revealed the name to Michael Crick, and the next day trailed the Newsnight story on twitter (without the name). Crick phoned McAlpine who apparently said he would issue a libel writ, and Crick tweeted this, again naming no names.

In the Guardian's story on McAlpine looking to see who to sue for libel, media lawyers argue:

'the BBC could be successfully sued for libel if McAlpine can prove that Newsnight aired enough material for some viewers to "reasonably infer" that he was the unnamed politician.

Steve Kuncewicz, a lawyer specialising in media and social networks, said: "If the BBC has said something that doesn't name him directly, but allows people to put two and two together, then there would be a risk in that.'

I was watching Newsnight and couldn't work out who they were talking about. However, some people worked out the name from the previous stories, presumably through a web search, and they named McAlpine on twitter. And this is where the workings of the internet come into play to make sure we all knew… At this point if you were to look at twitter's trending list, or search for tweets on Newsnight, you would have found a name. If you went to google and typed in Newsnight paedophile, the rest would be filled in without typing. (' Funny that when I type "Lord Macal" into google, autocomplete brings up "Lord MacAlpine paedo". ' talkboard entry posted in the afternoon before the Newsnight broadcast). People were then being directed to blogs and conspiracy theory sites as the real news had created a gap that they wanted to fill. And then it became even more silly, with Philip Schofield downloading one of the lists (presumably from a conspiracy theory site) and presenting it to the Prime Minister.

This is, of course, a problem not restricted to this story. The workings of Google, twitter and so on, are guiding people to particular versions of the facts. As Google's autocomplete 'predicts and displays search queries based on other users' search activities and the contents of web pages indexed by Google' it may introduce a path dependency: if two words are searched for together many times, then the autocomplete will suggest the combination to others, and so the search can become even more common.

At the same time, much of the information on the web is unverified and/or undated. While we can say that people ought to know the difference between a genuine news site and a talk board, some of the news sites can also sow misinformation. Notwithstanding the fact that some newspapers have different levels of sticking to the truth, it's also the case that stories stay online permanently and one doesn't need to go to a library to check old copies. Indeed, if they're not dated or the date's not obvious it may even look like today's news.

A great example of this came when I found a talkboard with mention of a riot in Stoke. This was a couple of years back, and the poster had quoted the Daily Mail about the riot, said to be yesterday, and then argued that all the media except the Mail were not reporting it in order to stop any people going to cause more trouble. The usual arguments about 'PC gone mad' were there – if it was white people rioting it would be on the BBC – and then other people had responded with similar.

However, there was no riot that weekend. Presumably, someone searched Google for 'Asian Youth Stoke' and got this Daily Mail story at the top, with no date and so thought it had happened yesterday. In fact it was news from 2001.

The question needs to be asked: does the democratisation of information, and the combined effects of google searches, twitter trends, undated and unverified information help us better get to the truth? Or do the biases of path dependencies and people looking for the information that they want to find mean that truth is less likely to rise to the top?