good experiments vs bad science (or a great evening at #nesta #theexperimentalsociety #rri)

I wrote earlier this evening during the event itself on some subjects touched upon by speakers and audience both at Nesta’s HQ, as the meet-up was unspooling the engaging thoughts of those who attended.

The best of the evening were the measured and tantalising responses from all parties, around the broad sweep of ideas Nesta, its people and the frame employed laid out for those attending.

A lot of the thoughts and even preoccupations of the many specialists and private sector participants present seemed to focus on the dichotomy of (suddenly, perhaps; or maybe relatively, now) acceptable government (ie public sector activity) versus unacceptable private.  In the camp of unacceptable private we had the large American tech organisations: the Chinese and other more distant powers didn’t seem to get a mention, however.  Maybe latent, or patent, anti-Americanism still rules some roosts in some parts of our country.

I tried to ask a couple of questions during the couple of hours we were there, but time constraints made this impossible.  So here they are, in the body of this post instead.  Anyone who’d like to engage with their thrust is most welcome to do so: and I promise, I don’t bite.  Not unless bitten.  And if distractingly, even a bite may be welcome …


Three observations I’d like to focus on, anyhow.

  1. Our attention was drawn to the fact that ID cards – in particular biometric ones – have spread where legislatures are essentially toothless, and have not where they are more inquisitive and vigorous.  One example given was the UK, though I find it difficult to understand the argument for two reasons: firstly, there are plenty of Western democracies which use fingerprints and now digital signatures for their ID cards: I’m not sure, then, about the specific technologies, but I am sure that democracy is not an impediment to their adoption; secondly, meanwhile, the UK makes good use I am sure of a far more sophisticated, outsourced, privatised and perhaps just-as-well ID card that is the Android or iPhone smartphone.  Even before fingerprints were used to unlock their entity for users, their digital footprint, and therefore their ability to track and predict citizen behaviours, was well developed.  So whilst Spain and other countries use a national ID system which measures and defines everyone with the same yardstick, the costlier your phone, the more the UK tracks you.  A hidden tax on the well-off, perhaps?
  2. The second observation I have relates to the clear reservations expressed that large private-sector corporations are gathering more intelligence these days about the component members of society than even the NSA, GCHQ and the security agencies of many other Western democracies.
    1. To be honest, whilst Edward Snowden’s revelations obviously horrified many people, there is an element of the panopticon about their impact: the important thing about CCTV, phone-tapping and the intervention of our electronic communications isn’t that it happens all the time (it may, I don’t know for sure, and probably never will) but that it happens enough for us never to be sure it isn’t happening.
    2. In this sense, modern security apparatuses, their parent the panopticon, and – bewilderingly I am certain for their proponents – open-source software communities, are all examples of how a million unseen eyes, but maybe essentially unseeing too (no one can ever watch a million eyes to check whether they are watching back or not), can help to self-regulate extremely complex circumstances.
    3. And similarly, in this sense, it doesn’t matter whether experiments on society are carried out by the precursors to Facebook, Apple, Amazon and Google which are our Western governments of centuries ago, of recent time, and of some uncertain time into the future – or, instead, by the vilified corporations already mentioned: the important issue here isn’t who does what, but – rather – how they involve the citizen in what they do, if indeed the citizen is ever to be involved.  If Facebook wants to experiment on me and my feed, and won’t tell me, and won’t inform me of the results in a form I can process to my heart’s content, I may be sufficiently clued-up to be angry with the corporation.  However, I will be just as angry, and I consistently have been, when my elected government acts in exactly the same way.  Facebook et al are reflections of behaviours laid down a long time ago.  It is, after all, Western legislatures which comfortably continue to promote the legal figure of the corporation, a figure which has existed for centuries.  They may very well be right to allow it to exist, but what they can’t do – what we as citizens can’t do! – is then complain about what we have long ago agreed to.
  3. The third point I’d like to make relates to the word “experiment” and the idea of “bad science”.
    1. One attendee accused Facebook’s voting experiment of exactly that: bad science.  He said (correct me if you were there, and feel I am misstating) that the process lacked, for starters, a hypothesis.  I am young (figuratively and professionally) and green in these matters right now, but having just started a Master in Criminal Justice at Liverpool John Moores University, I realise that the concept of ethnography – widely used in the field, and as scientific in its processes as any more traditionally understood scientific method – would suit Facebook’s datasets right down to the ground.
    2. If Facebook did not use an ethnographic approach in what it did in 2010 and more recently, it maybe did “bad science” after all.  But if its data crunchers were to stumble across the concept of ethnography, as I have happily done, and even though inexpertly still, I am sure that perfectly and satisfactorily valid experiments could be carried out by any number of Facebooks.
    3. Finally, an anecdotal observation on my part in relation to the whole idea of experiments: kids love experimenting.  If, as adult voters and citizens, we prefer to reject or skirt around the term, or feel uncomfortable in its space, perhaps the real reason is that as adults we find ourselves not doing the experiments but, rather, having the experiments done to us.  There is a substantial difference as far as potential buy-in is concerned.  We need to establish this difference, its impact and how to take advantage of it, if the experimental society is ever to properly flourish.

“the experimental society”

As I write these words, I am attending the Nesta event “The experimental society”, at its headquarters in London.  

The comms tech being used – easily visible to both speakers and audience – is really nice: poll and question software called, which even allows for spontaneous thought-up answers to add to the standard multi-choice responses.

The debate has already touched on the privacy side of ID cards where politicians reframe studies whose results they don’t like; on the right a society – more importantly, the private sector – has to conduct the often hidden experiments being discussed tonight; specifically, quite a lot of debate around Facebook’s manipulation of people’s streams around election times; and particularly on the allegedly “poor science” nature of such social experimentation.

I have to say as a newly recruited ethnographer in the context of Criminal Justice, and as a citizen never instinctively in favour of the Facebooks of the world, to condemn such beasts as having conducted bad science shows a gently limiting acceptance of what the scientific method can mean.

A large scale ethnographic experiment may not have been what FB actually did on the occasions described this evening, but with more work and preparation, and an understanding of the various options too, it would be very easy to do good ethnographic science – that is to say, deliberately going in without a prior hypothesis to hand – whilst generating the kind of data which FB functions with every day.

My impulse in all these things is never to stop nor attempt to control what anyone does, but work from a promoted and shared culture and wider environment which aims to better encourage societal participation in both experimental design and the understanding of the end results themselves.  Never even attempt to stop stuff full-stop: rather, look to create a far more collaborative and cooperative societal Petri dish out of which this experimentation will then emerge quite naturally.

To paraphrase Peter Levine: work to a Good Democracy where inclusiveness and efficiency are balanced – not only in well-understood political debate but also, now, in experimental societal-wide innovation and implementation.