The nym wars: how many identities are enough?
The biggest political question this year might not be national debts or the Arab Spring, but what form identity will take on the Internet in the future. As the Google+ service began demanding that people sign in with their legal names and suspending accounts believed to be in conflict with this policy, the “nym wars” broke out. Google is not alone in wanting to keep online identities strongly tied to legal identities: the National Geographic Society demands that comments on ScienceBlogs may no longer be pseudonymous, and Facebook’s Mark Zuckerberg has stated:
But can we live a human life with just one identity?
Who owns the agora?
The real struggle is not about anonymity, the ability to make comments not traceable to anybody, but pseudonymity, the ability to maintain a more or less stable persona that can interact with others without revealing all aspects of the self.
And it matters to society because, as Kee Hinckley puts it in an excellent essay on the topic:
the forum for public discourse is no longer the town hall, or newspaper, or fliers on the street. It is here on the Internet, and it is happening in communities like this, hosted by private sector companies.
If it was only a matter of local rules of social spaces that potential participants could choose freely between, it would be a non-issue. But these systems are becoming more than local clubs: Facebook, Google, Twitter and some blogging platforms are increasingly not only the medium through which discourse flows, but the identity providers for many more functions of life.
For example, my Google login ties together an email identity, a calendar, a Google+ identity, a number of documents, a blog, and membership of mailing lists. Other services rely on such identity providers to keep track of identities, outsourcing the hassle of connecting the disparate utterances and attributes of a particular voice in cyberspace with each other. This is normally useful both to me, Google and third parties. But I would be harmed much more if this identity and the connections it represents were to be dissolved than the other parties. The loss would not just be hassle, but loss of access to documents, information and – most importantly – social relations maintained through the medium of this identity. An on-line identity is simple from a software point of view, but can encompass a sizeable amount of personal effort, social capital and emotional value.
Controlling the possible identities of people is a profound power. From a purely economic perspective it can be good business: there are economies of scale to be the main identity provider to an identity metasystem. Being the organisation other organisations have to deal with in order to handle identities can no doubt can be profitable. Traditionally this has been the domain of governments maintaining legal identities, but with the emerging globalised fluidity they have a hard time keeping up. The legal identity mainly functions as a guarantor that breaches of contract or crimes can be pursued legally. That is its only advantage, and why it would be such a benefit for the would-be identity providers to have a link between online identities and legal identities: their identity systems would gain further value and power, improving their draw to the third parties who are the real customers.
But controlling identities allows control over who can express themselves in different forums, and how. Most trivially it can control access, but it can also prevent or encourage linking different aspects of a person’s life. We are all using different personas in different contexts, managing the links between them in complex ways. The selves we show to a partner, doctor, boss, friend, colleagues or gaming guild are all different – and usually need to be so. The person who behaves the same in bed with a partner as they do at work is a most queer character. There are also numerous public roles – as professionals, as interested public, as debating voices, as whistle-blowers and dissenting members of groups – that require proper control over what information and identities are linked together.
The list of groups disadvantaged or harmed by a policy that makes it hard to separate an online identity from their main identity is long. The concerns listed are by no means hypothetical – people have indeed been harmed by linking separate identities, or been scared to silence by the threat of someone linking them (again, see Hinckley’s examples). This is the most worrisome aspect of the power of identity providers: it can have chilling effects on discourse even without any action being taken.
I have earlier argued that pseudonymity is important for some functions of the open society: some things need to be said, but there is no need for them to be said by an identifiable person. In order to build credibility a voice also need to have a record and show that there has been a personal investment into the identity – full anonymity rarely produces any credibility. This is not just relevant for whistle-blowing or major criticism against power, but can be just as relevant in the dissemination and discussion of science.
But most fundamentally, pseudonymity is about reducing constraints on our authenticity (Hinckley):
Persistent pseudonyms aren’t ways to hide who you are. They provide a way to be who you are. You can finally talk about what you really believe; your real politics, your real problems, your real sexuality, your real family, your real self.
Do no evil?
The resolution to the struggle over what kind of identity management will dominate the future is far away. It is a complex problem with both economical, political, cultural and technological aspects, and billions of stakeholders. But it is also an ethical problem: what should an internet company unwilling to do evil do?
First, recognizing that there is a profound problem is useful. Thinking, like Zuckerberg apparently did, that human social identity is simple or can be reliably changed because of software, means running headlong into conflict. At best it will be just the internet company that will be hurt.
Second, at worst a simplistic view of human identity can lock us into inhuman identity systems. If we are forced through law, economics or technology to act according to a rigid model of identity that does not suit us we will suffer. This suffering will be in domains that can truly hurt us: having one’s identity, voice or social links circumscribed are among the worst things that can happen to a person. Identity systems must allow open-ended and creative identity creation. Just because there is no menu option for a social relation doesn’t mean it is invalid.
Third, any entity that becomes the manager of identities gains a number of responsibilities to the people whose identities are being managed. It is holding more than data, it is holding important social connections. If it cannot be held accountable and there are no incentives for it to treat people well, then there is potential for great harm. Identity providers cannot avoid becoming social institutions, no matter what kind of organisation they are, and this entails that stakeholders must get a decisive say.
Fourth, the public also has a responsibility for maintaining the identity ecosystem. To not care about how the fora of social discourse function means allowing their structure will be determined by others. To not care about the structure of one’s identities means allowing others to define them – potentially severely reducing autonomy and authenticity.
Then we shall see face to face?
But maybe we should work to create a world where we do not have to show multiple faces? Maintaining a separation between multiple selves can be tiresome, and sometimes it is a relief when the mask slips and allows a glimpse of the vulnerable true self beneath. But that world would have to be a world of extreme tolerance, mental health, no crime, and no information asymmetries. Its inhabitants would not mind hearing frank discussion of themselves or their children, their confessions would be public, and their intelligence agencies would be acting in the open. We are very far away from it.
Aiming in that utopian direction – both institutionally and individually – might be a good aim. But that requires thoughtful effort, not a decree.