This is a follow up to a previous post discussing the pros and cons of anonymity in e-participation. These are the main points found there

Greater anonymity may increase uncivil behaviour and the use of offensive words[21,22].
Greater anonymity may reduce comment quality[23].
Greater anonymity may reduce trust, cooperation and accountability[22].

Conversely

Greater anonymity may increase participation[25] and engagement[26].
Greater anonymity may yield more information[23] and produce more honest[26] and original ideas[29].
Greater anonymity may produce more equal[30,31] interactions leading to free discussion of controversial issues.

In this two-post series we consider two further papers on the subject, Ruesch & Märker 2012 – Making the Case for Anonymity in E-Participation and Moore 2016 – Anonymity, Pseudonymity and Deliberation: Why Not Everything Should be Connected. The main theme we wish to discuss is the idea that a sweet spot may exist in the anonymity spectrum, a solution that is optimal in balancing conflicting properties at each end.

Making the Case for Anonymity in E-Participation

We review the arguments made in the first paper. This paper conducts an analysis around the real world participatory budget of the German city of Gütersloh. Although the title suggests a position in favor of anonymity we note that the term anonymity is subject to gradation. This gradation of anonymity is the spectrum that we referred to above, where the balancing between opposing arguments occurs. The discussion is applicable to online forums, social networks and the web, so general points are first reviewed.

General arguments

  • The self-control argument

[..] there is strong evidence for the thesis that anonymous communication tends to cause more offensive communication; real name policy in turn facilitates civilized communication. Being identifiable by others seems to encourage self-control, decrease personal defamation and thus facilitate the creation of “communities of trust”.

  • The legal argument

Proponents of real name policy thus point to the importance of clearly identifiable profiles in order to be able to hold Internet users legally accountable for their words and deeds.

  • The online=offline argument

If we are accountable with our real names offline, why should we not have the same rule online?

Conversely

  • The open participation argument

forcing users to provide their real name will exclude many from participating in the forum or social network. This case is probably most clear when looking at authoritarian regimes where real name policy will exclude anyone who is critical of the government.

  • The freedom argument

[..] users are able to speak more freely and less self-censored and coloured by groupthink [..] when not having to reveal their real identity.

  • The privacy argument

[..] users must have the right to decide if certain political or other opinions are to be openly accessible, especially regarding the fact that the
spread of information is almost not controllable once published on the Net.

E-participation specific arguments

Most of these general points should be familiar as they are closely related to observations made in the previous post. Let’s turn now to arguments that are specific to e-participation as practiced in the case study described by the paper

so far the question of online anonymity has almost exclusively centred on social networks and unmoderated online forums. The area of e-participation has to date remained largely unexplored in connection with anonymity and real name policy. Therefore, the present paper sets out to relate the ‘anonymity debate’ explicitly to e-participation, [..]

The arguments are presented as rationale (pros) and then objections (cons).

Rationale 1: Real name policy and request for personal data ensure that only [eligible] citizens [..] participate thereby enhancing representativeness and in turn legitimacy.

Rationale 2: Real name policy and request for personal data ensure a high quality of dialogue by preventing offensive comments from anonymous citizens.

Rationale 3: Real name policy and request for personal data ensure transparent communication, thereby strengthening democracy.

These points are countered by the following objections

Objection 1: Real name policy and request for personal data distract from issue-related dialogue and thus degrade the quality of discourse due to biased, person-focused perception of messages.

Objection 2: Real name policy and request for personal data violate citizens’ right to privacy.

Objection 3: Real name policy and request for personal data cause time and cost intensive administrative problems.

Objection 4: Legal, administrative and usability problems caused by real name policy and request for personal data result in negative media attention and negative public perception.

Objection 5: Real name policy and request for personal data cause usability problems which act as a barrier to participation and thus lead to a decline in participation.

There is significant overlap between the general arguments made regarding online discussion and the internet in general with those made in the specific case of e-participation listed above. For example, the general self control and quality arguments map naturally onto Rationale 2 and Objection 1. Similarly, the open participation and freedom arguments map onto Objection 2.

Integrity preserving pseudonymity

The authors of the paper make the case that the Objections to real name policy are stronger than the Rationales, based on the Gütersloh experience. But as we said above, anonymity is a spectrum and we need not restrict ourselves to only two possibilities.

there are various ‘compromises’ between anonymous communication and real name policy, ranging from no registration at all (complete anonymity) over registration with pseudonyms, registration with real but unverified name or registration with hidden real name plus pseudonym, to registration with verified name and possibly also personal data.

Here we suggest one such possibility that can be interpreted to exist at the midpoint: Integrity Preserving Anonymity. With integrity preserving pseudonymity citizens are required to validate their real identities in order to participate in the platform, but said identities remain secret and private. Importantly, this privacy is maintained with respect to the general public but also with respect to the institutional authorities which manage the participation process.

With this approach, citizens still maintain a degree of identity in the sense that their contributions can be linked to their pseudonyms. But these pseudonyms cannot be linked to real world identities, beyond the required fact that the corresponding real world identity must be eligible to participate. This combination of characteristics is what makes the solution a midpoint with respect to anonymity. Identity is stronger than pure anonymity, defined by the set of publicly expressed comments and positions. But identity is weaker than real identity, pseudonyms cannot be linked to citizens.

Integrity preserving pseudonymity can be implemented with anonymous credentials, a technique from cryptography first proposed by David Chaum (CHA 85). An example of such a technique is described in our post Privacy-preserving e-participation with nMix and its followup, but we do not discuss this further here as it is the how rather than the why. Returning to the arguments above, we can revisit the pros and cons for this type of pseudonymity. The following table shows whether the Rationales and Objections offer suport for, or question, the use of pseudonymity, as well as the  two edge cases of pure anonymity and real world names. For simplicity we have given the arguments short names, and subsumed Objection 4 in Objection 3.


ArgumentReal world nameAnonymityPseudonymity
Legitimacy and integrityX
CivilityX/X
Communication transparencyXX
Issue-centric debateX√/X
Privacy, inclusion, freedomX
Administrative complexityXX
Usability and participationX/X

Cells are marked with  or X depending on whether the argument supports or questions the anonymity policy. Cases where the argument has positive and negative aspects with respect to the policy are marked with both. Here’s a review for pseudonymity

  • Legitimacy, integrity

By definition integrity preserving pseudonymity ensures only eligible citizens can participate and maintains integrity.

  • Civility

Pseudonymity has elements of both the real name policy and anonymity. On one hand, citizens cannot be identified entirely, so there is some opportunity for uncivil behaviour since participants are not fully accountable. On the other hand, pseudonymity does allow for the formation of some degree of identity based on the set of contributions that can be linked to the same citizen. This provides for a level of accountability that although smaller than that for real names is greater than pure anonymity.

  • Communication transparency

With pseudonymity citizens do not know what real person they’re talking to, so according to the definition in the paper this argument is negative (although points similar to those for civility could be made).

  • Issue-centric debate

This argument is ambivalent for the same reasons as those described for Civility. The existence of some level of identity may detract from a pure issue centric debate as would occur for pure anonymity.

  • Privacy, inclusion, freedom

By definition pseudonymity protects citizens’ real identity, and therefore also contributes to inclusion and freedom.

  • Administrative complexity

Pseudonymity requires citizen authentication based on real name and other personal information, so it will incur the same complexity problems as those for a pure real name policy. Additionally, pseudonymity requires a careful implementation with the right cryptography, incurring a cost that we can also classify under administrative complexity.

  • Usability and participation

This argument has both positive and negative aspects for pseudonymity. As stated in the previous point, the complexity stemming from real name authentication, especially if cryptographic protocols are involved, will decrease usability and therefore participation. On the other hand, some arguments that point to increased participation for anonymity apply to pseudonymity as citizens’ real identity remains protected. This point is related to the Privacy-inclusion-freedom argument.

A net evaluation of a pseudonymity policy depends on the relative importance that is assigned to each argument, simply counting the pros and cons is naive. If Administrative-complexity and Communication-transparency are considered less important than Legitimacy-integrity and Privacy-inclusion-freedom, then pseudonymity can function as an optimal balance with the best net result. This balance exists because pseudonymity maintains the important properties of pure anonymity while ameliorating some of its drawbacks, those that arise when citizens have no identity at all. This idea is echoed at the end (italics added) of the following remark in the paper

Although the case study based nature of this paper does not allow for generalizations, it still provides some strong indications that real name policy should be avoided in e-participation projects, and that negative effects of anonymity can be sufficiently accounted for by the use of pseudonyms and moderation

However, one must also acknowledge that the level of identity present in pseudonymity does present a greater threat to Privacy-integrity-freedom than pure anonymity, if inference attacks are conducted on pseudonyms’ linkable contributions.

Summary

We reviewed the general as well as the e-participation specific arguments for and against anonymity presented in Ruesch & Märker 2012. Most of these arguments could be related to pros and cons present in other sources as collected in the previous post. It was remarked that anonymity can be seen as a spectrum, with integrity preserving pseudonymity as a possibility located between the two ends. This possibility was then evaluated with respect to the presented arguments. Depending on the relative importance of each argument, it was suggested that this form of pseudonymity could be an optimal solution balancing conflicting requirements of e-participation systems[36].


(Chaum 85) – Chaum, David (October 1985). “Security without identification: transaction systems to make big brother obsolete”.

(Ruesch & Märker 2012) – Making the Case for Anonymity in E-Participation

[20] De Cindio, Fiorella. 2012. “Guidelines for Designing Deliberative Digital Habitats: Learning from E-Participation for Open Data Initiatives.” The Journal of Community Informatics 8 (2).

[21] Fredheim, Rolf, Alfred Moore, and John Naughton. n.d. “Anonymity and Online Commenting: An Empirical Study.” SSRN Electronic Journal. doi:10.2139/ssrn.2591299.

[22] Cho, Daegon, and Alessandro Acquisti. 2013. “The More Social Cues, The Less Trolling? An Empirical Study of Online Commenting Behavior.”

[23] Diakopoulos, Nicholas, and Mor Naaman. 2011. “Towards Quality Discourse in Online News Comments.” In Proceedings of the ACM 2011 Conference on Computer Supported Cooperative Work – CSCW ’11. doi:10.1145/1958824.1958844.

[25] Fredheim, Rolf, Alfred Moore, and John Naughton. n.d. “Anonymity and Online Commenting: An Empirical Study.” SSRN Electronic Journal. doi:10.2139/ssrn.2591299.

[26] Davies, Todd. 2009. Online Deliberation: Design, Research, and Practice. Stanford Univ Center for the Study.

[29] Connolly, Terry, Leonard M. Jessup, and Joseph S. Valacich. 1990. “Effects of Anonymity and Evaluative Tone on Idea Generation in Computer-Mediated Groups.” Management Science 36 (6): 689–703.

[30] Flanagin, A. J., V. Tiyaamornwong, J. O’Connor, and D. R. Seibold. 2002. “Computer-Mediated Group Work: The Interaction of Sex and Anonymity.” Communication Research 29 (1): 66–93.

[31] Klenk, Nicole L., and Gordon M. Hickey. 2011. “A Virtual and Anonymous, Deliberative and Analytic Participation Process for Planning and Evaluation: The Concept Mapping Policy Delphi.” International Journal of Forecasting 27 (1): 152–65.

[36] This is reminiscent of the privacy-integrity tension found in secure voting, and indeed there is overlap on the privacy part as it relates to freedom.