The physiognomy of identity?
I was listening to BBC Radio 4’s Thinking Allowed programme the other morning, and it had a segment on the 19th century idea of physiognomy - the notion that a person’s inner morality and character is reflected in their physical appearance and manner. One of the participants in the discussion, Sharrona Pearl, suggested that the rise in popularity of the idea and it’s elevation to the status of a ‘science’ in Britain co-incided with mass urbanisation. The cities were filling up with people who had left rural villages where there were close-knit clans with strong trust relationships and they suddenly needed to make very quick trust decisions when surrounded by a multitude of strangers (to paraphrase). It struck me that the Internet is akin to that state now and we also need a framework by which to make fast trust judgements when we encounter people online, and so there may be much to learn from the history of physiognomy in this respect.
Of course, although it’s undergoing something of a revival, physiognomy itself is
at best a pseudo-science and has been entirely unsuccessful at providing a definitive set of rules for making trust decisions. Nonetheless, we make trust judgements based on physical appearance all the time, consciously or not. And, as Malcolm Gladwell argues in Blink: The Power of Thinking Without Thinking, in general it serves us pretty well. So, I wonder, could it be that when we design online systems we currently concentrate far too much on authentication and not enough on codifying and presenting what is known about an identity?
Let me explain what I mean.
A couple of weeks ago I had the pleasure of talking to social engagement expert Steven Clift at the Guardian’s Activate 2010 conference. Steven has spent the last fifteen or so years setting up local online discussion groups to get neighborhoods more involved with each other and with the wider community. One of his messages at Activate was that ‘civility’ matters and that the only way to ensure it is to make ‘real names’ the default identity in any online space in which government and citizens engage with each other. (He also acknowledges that anonymity is also sometimes important, but only in exceptional cases such as whistle-blowing).
Now, I buy that argument. It’s crucial that discussions are conducted with civility in order for debates to have half a chance of being fruitful and inclusive (look no further than the UK Government’s Spending Challenge fiasco this week!). However, I think there are many more kinds of online identity than simply real names and anonymous, in fact I think there’s a nuanced spectrum of identity that needs to be far better reflected in the way people are presented online. I’m sure there must be a load of research in this area (and would love to be pointed at it), but, thinking about it independently, I think there are 3 basic kinds of identity, each with two variants.
The basic types are:
- Anonymous
- Pseudonym
- Real Name
Anonymous can be named with a default, e.g. Guest; or a customized name ‘on the fly’; pseudonyms can either hide the person’s identity, or be well known (or at least an open secret); and real names can be verified and unverified. On top of these basic types, there are a range of attributes that contribute to the identity: attributes that relate to the platform being used, such as number of posts, date joined, reputation, etc.; attributes that derive from the identity’s social graph, connections to facebook and linkedin, etc.; and attributes that are contributed and modified by other people.
Then there is the tension between how the person would *like* to be perceived and what their behaviour identifies them to be; and also not just whether the identity is trustworthy, i.e. whether the identity and the person behind the identity are congruent, but also whether what the identity says is truthful, i.e. whether the person behind the identity is knowledgeable in the subject they are talking about and whether they have a track record of honesty or not.
So, there is a huge challenge in defining what pieces of information should contribute to a meta-identity, not the least of which is deciding which pieces of information are equivalent to each other in the codification. However, there is an even greater challenge in figuring out how to display all that information visually in such a way that a viewer can process it and form a judgement about it in an equivalent amount of time that it takes us to form an initial opinion from a person’s face and demeanor.
Perhaps every time an identity, an avatar, is presented online it should be accompanied by a second image - the image of a face, wrinkled by lines of trust, behaviour and social standing…
