Martin Delius, former state representative of the Berlin Pirate Party
One of the thorniest questions the Berlin Pirate Party had to face as soon as it decided to adopt LiquidFeedback in January 2010 is that some of its members had created pseudonymous accounts on the platform. This created multiple problems. To being with, the party had to verify whether these pseudonymous accounts were owned by actual party members. Second, because LiquidFeedback (LF) allows users to vote, it was necessary to verify that each user had one and only one account.
Third, a decision had to be made on whether pseudonymous accounts could receive delegations or not. Because the software does not distinguish between pseudonymous and non-pseudonymous accounts this was a political, or human, decision. A discussion began within the party. Many arguments for and against pseudonymity were heard. Eventually (that is, almost five years later) it was decided that members could use a pseudonym in LF on one condition: they had to introduce themselves at a party meeting with their pseudonym. In this way, pseudonymous accounts with delegations could be held accountable for their decisions within LF, which functioned in Berlin as the equivalent of a permanent party Convention. As I noted in my previous post, because LF blurs the lines between the represented and the representatives, the delegating and the delegates, it is difficult to extend the anonymity of voting (the secret ballot) to such system. But it is also difficult to expect total transparency of its participants as some political decisions may be too sensitive to be exposed out in the open.
In order to strike a balance between the transparency we all expect of representatives and the private nature of voting the Berlin Pirate Party took two separate steps. First, it decided that LF could be accessible only to regular party members, that is, all political discussions and decisions were not going to be accessible from the rest of the Internet. (If this may sound obvious to many, it wasn’t for a party that had made of “radical transparency” in politics one of its talking points). Second, as noted, the party decided that those who wanted to use a pseudonym within LF had to be identifiable qua party members. It sounds like a reasonable compromise to me. Yet it is interesting that the developers of LF and the Berlin Pirates I have interviewed so far (all influential members of the party) found the pseudonymity issue… stupid. In other words, they all seem to agree on the fact that pseudonyms are pointless in a system like LF. Here is an excerpt of the interview with Martin Delius, former Pirate Party representative at the Berlin House of Representatives and recently reelected with Die Linke.
Q. There are many good arguments for transparency, including the one you just made—i.e. you cannot be deemed authorative and trustworthy on an institutional level if your identity is unknown. And there is no doubt that it is difficult to delegate a vote to somebody whose identity is concealed. However, there were some people in the Pirate Party who insisted on using a pseudonym within LF. There must also be some good arguments for pseudonimity…
A. Not one, not even one, and I will explain to you why. Regardless whether you use a pseudonym or not, if you actively and regularly take part in decision-making processes through such a platform your pseudonym will not rescue you from being a public figure. You will be identified somehow. Because you are not only taking part in the system through a pseudonym. You need to have connections outside of the system to get the majority, to get something done. Because the system won’t provide you with majorities, you have to fight for them. You have to use Twitter, mailing lists, public meetings and so on to get your arguments win. It won’t be enough to write a text and get somebody to vote for it. You may use a pseudonym, you may make it harder for other participants to participate in the decision-making process. You may even annoy most of the other participants by using a pseudonym. But that’s all you can do. You cannot protect yourself from being identified by using a pseudonym in a system that has been running for more than a week and that lets you take part in more than one decision-making process over a certain amount of time. You have no choice.
Q. But there are certain decisions on which you want to be able to express a preference without having to explain to anyone why you made that choice. For example, you want to be able to vote without being conditioned.
A. When you go to vote you talk to your dears, you talk to people you know. So you are never entirely free from conditioning.
Q. Yes, but one thing is to choose to be influenced, to ask for opinions, and another thing is to be blackmailed by those who may be in the position of doing it.
A. That’s not an issue specific to LD, it’s an issue for society. Yes of course you have to address blackmailing when it happens. But you can do that by being transparent. By having an unbroken chain of history you can prove that it has happened. For the first time, using a LD process such as LF, you can show to everybody that you have been blackmailed, that someone has been blackmailed. So that is an issue, but it is not an issue for ICT-based decisional processes.
Q. So you are saying that transparency protects the weaker subjects from forms of violence—whether material or symbolic—that may be exerted upon them. And that’s because uneven power relations preexist the advent of ICTs they have to be addressed within a wider societal context rather than assuming that online participation automatically levels the playing field.
A. Whether or not an individual is able to play an active role in a democratic society has always been an issue. Who is influencing me? If I say my opinion loud who will be arguing against me? Who is in a position of power to do something bad if I am arguing something he disagrees with? This has always been an issue, nothing has changed with the Internet. Now, more people demand to participate in decision-making processes at every level of scale. In a LD system, no one is interested in the members who are there just to vote. The interesting members are the ones who actively participate, the ones who get many delegations, the ones who participate in public discussions around the issue. The people who just vote are not interesting for someone who wants to meddle with the process.
[End of excerpt]
To back up his statement, Delius also made the example of how the Berlin Pirates were able to win a majority in Parliament for the approval of new legisltaion on data protection. This was possible thanks to a member of the Berlin Pirate Party who started an initiative in LiquidFeedback for decentralized data protection, which won the approval of the party members. Because the BPP elected representatives used this initiative in Parliament, the other members of the Berlin House of Representatives wanted to know who started it. For example, some journalists and other politicians wanted to know whether the author of the initiative was working for Google as Google might benefit from a decentralized and non-homogeneous legislation in the realm of data protection.
They were rightfully asking about his identity. After a discussion, we decided to open up the system and show the public this single decision-making process, from the initiator to the participants to the discussion in our caucus. Transparency International looked at it, Der Spiegel met with him, and unanimously decided that he is the best informed member of the party, he is not paid by Google, the decision-making process was valid, and everything was in a legitimate state. That made our argument as strong as it had never been. That’s why I am saying it is right to burden the participants of the system with the maximum of the transparency you can burden them with. For the sake of the system, for the sake of process.
Delius said. It seems to me that in this particular case it makes a lot of sense that the identity of the proponent is revealed. But there are other cases (e.g. let’s think for example about a law that regulates prostitution written with the contributions of those who exercise the profession) in which the need to identify the proponent may get in the way of a thorough debate. In other words, the danger that I see here is that this diffused parliamentarinism may end up investing aspects of social life that are refractory by definition to what is legally constituted.
It is true that a system like LF demands more accountability of its participants. But not all participants are ready to be accountable, and often for good reasons. So whereas Delius does not see even “one good reason” for why people should use a pseudonym in LF I see many good reasons for why they may not want to use LF to begin with. And this should be a reason of concern for a system that cannot claim to be truly democratic until it will not be viable for all sorts of people, including and perhaps especially those who cannot afford to freely express their opinion and who may have little time and resources to invest in online politics. Differently put, using a pseudonym within a system designed to produce a completely transparent decision-making process may not make much sense in the end.
But why should we assume that total transparency is always good in politics?
Transparency is certainly positive and necessary within an institutional context. But we should not forget that political parties function as transmission belts between society at large and institutional politics. The risk that I see here is that implementations of Liquid Democracy such as LF may tilt the balance between society and institutional politics too heavily towards the latter, forcing the many forms of political participation to conform to a standardized protocol–or else to be dismissed as insufficiently elaborated, opaque and ultimately incapable of meeting certain criteria.
Indcidentally, a similar process is at work in Italy, where the M5S has created a massively participated platform, Rousseau (see the first two posts on this blog) to allow its members to participate in the writing of laws. If there is nothing wrong in this, the emphasis on the importance of changing the law risks of channeling so much political activism towards crowdsourced forms of lawmaking. I personally see no point in changing the law if these changes do not stem from lived community experiences or from concrete activist campaigns. To be sure, this observation could be extended to any form of online political participation. For now, I should just notice that what we may call an emerging direct parliamentarianism brings with itself a series of “prescriptions” (such as the request of total transparency) which make sense only for those subjects who can accept full responsibility for their actions.