Citizen law making vs. legal illiteracy

Raluca Onufreiciuc
Raluca Onufreiciuc
Oana Olariu
Oana Olariu


Law-making should not exclude citizens. The following paper analyses how citizens’ engagement in policymaking fails because of citizens’ illiteracy in law and administration knowledge and provides an updated scenario for building common knowledge through discourse production along policy drafting. In a context of a huge appetite for active participation in the decision-making processes, it is of vital importance to collect people’s views and gain their confidence and support. Creating a constant habit from citizen participation is not only bringing added value in law making, but it is also a guarantee that they will meet the needs of citizens and therefore generate public commitment. Although citizens can contribute with their input, they have limited understanding and control over the data they provide and the results, often remaining detached from the very mission and scope of such involvement.

Key words: e-democracy; public participation; civic engagement; illiteracy; citizen law making.


Engaging citizens into policymaking is considered the next milestone in national and city administration, in accordance with practices established by such initiatives as Open Government Partnership (Faria, 2013). Until now, several iterations created a pool of international experience in conducting such maneuvers related to collaborative law and policy making.  The iconic case of Iceland drafting its constitution by making use of collective intelligence in 2011 represented an ice breaker for such trials previously discarded as utopic or unfeasible. Other initiatives followed, for instance, in Italy (Cindio & Stortone, 2013), California (Nelimarkka et. al., 2014: 446), Finland (Aitamurto & Landemore, 2015), or Mexico City (“Crowdsourcing the Mexico City Constitution”, 2018), to name just a few.  However, most of them faced great challenges due to incomplete processes design.

Engaging citizens into policy making is supposed to raise the quality of public policies by making use of crowd intelligence, and to connect citizens to public decisions, which, in turn, is considered to curve the probability of state capture by groups of interests and influence. It is also considered to be the best approach in diminishing the gap between the structures of traditional power and the citizens they should serve. That, however, enforces a different relation between the two actors, changing the subordination and hierarchical approach to one based on partnership structures, which are more horizontal in nature.

I. Methods of engaging citizens

There are two main veins for engaging citizens into policy making.  One is crowdsourcing and the other one is commons-based peer production. Both of them have their shortcomings, as well as their strengths. Crowdsourcing is more often used in policy drafting, compared with commons-based peer production, because it makes it easier to enact the principle of inclusiveness. Crowd intelligence is valuable because of the cognitive diversity it provides, adding new complexity layers to the future regulation that’s under construction. The most general definition would place crowdsourcing as an open call for anybody to contribute online to a specific mission. Tasks could vary in accordance with the end result that’s aimed and could cover everything from tagging pictures for rescue operations or disaster management to providing ideas or bits of information for solving R&D problems. The main purpose is to allow an actor to tap into a large pool of ideas and information which would be difficult to access otherwise. When institutionally enacted, it is mainly used for brainstorming problems and solutions. The institution has the initiative to organize and shape the processes and the institutional team is responsible to make use of the gathered information and to build it into a policy.

A policy making cycle takes several stages. At first, problems are identified and relevant data is gathered.  Different options and proposals are drafted, based on analyzed information. A consultation is run afterwards, then the policy is drafted, evaluations are made, decisions are taken and the process moves to implementation phase (Aitamurto, 2016). Citizens are engaged in brainstorming problems and possible solutions, then asked to take part in consultation, as well as in evaluation, to select the most frequent preferences. This is a simple way to ensure public involvement and citizens’ direct voice in public decisions. Instead of a passive citizenry, an enhanced citizen participation is based upon well-informed citizens in what regards decision-making processes and the belief they are capable of making a positive change.

However, the process gets usually stuck because traditional governmental structures lack the ability to absorb crowed intelligence products. As Aitamurto (2016) points out, based on the Finnish experiment to engage citizens into off-roads traffic bill, crowd logic is sharply contrasting with the logic of policymaking. Traditionally, the input on which a law is based takes a synthetized form. The input provided by crowds is atomic in nature, large and unstructured, highly diverse. Also, traditionally, the quality of the input is uniform or at least known in terms of how it varies across different degrees of expertise. However, the input provided by citizens varies in quality to unknown degrees. Probably a cornerstone difference, the input usually provided for compiling a law or a regulation is fitting the general legal and policy environment. The citizens’ input, based on personal experience, is generally unfitting within the broader picture. The Finish experiment failed to reach its results because the evaluation given by experts proved to contrast with the evaluation given by citizens, due to a biased community which took part in the process. Avoiding bias is a difficult if not impossible desiderate in crowdsourcing processes, because participants are self-selected. They share, as such, strong motivations and are directly interested in the outcome, acting upon very specific group related objectives. In addition, the Finnish officials did not know how to analyze and make sense of the highly diverse and contradictory crowdsourced data (Aitamurto, 2016).  On the other hand, in analyzing if educated elites participation in law making and adherence to rule of law, Ogun State from Nigeria carried out an experiment to see in which manner this could “affect” the vital component of achievement of good governance (Nwogwugwu & Ajayi, 2015: 61). The role of the educated elites involvement in the processes of law making was to permanently check and guarantee that the right laws are made by the legislators of Ogun State and public officials have a strong adherence to rule of law taking into consideration their general low commitment.

The other method of engaging citizens in policymaking is based on collaborative writing and drafting projects, much in a wiki style approach. It is defined as commons-based peer production (Benkler, 2002) and allows citizens to cluster in developing policy fragments, reviewing them and compiling the parts together. The method was used, for example, in California, in 2014-2015, as well as by the White House in 2015 (Aitamurto, 2016) and was part of the citizens focused local government’s strategy in Madrid, between 2015 – 2019. However, the method proved to hit very low in citizens engagement, failing to enact the principle of inclusiveness.

An overview on the two methods shows a sharp complementarity. While crowdsourcing ensures large crowds participation and cognitive diversity, it fails into channeling the input towards a synthetized and structured body of knowledge, fitting the larger picture of policies, regulations and laws. Commons-based peer production, by the other hand, because of the deliberative processes built in, is successful in channeling, synthetizing, structuring and fitting knowledge appropriately, but lacks engagement and cognitive diversity, because of the higher entry barrier it imposes. Ideas contribution is a much more accessible task than a structured problem solving activity. Because across online environments, 90% of visitors are passive observers and only 10% engage (Aitamurto & Landemore, 2015), high entry barriers are hindering the whole process.

II. Participatory vs. direct democracy

Before moving forward in proposing a solving scenario, it worth pointing out that the two methods are differently positioned across the spectrum of participatory versus direct democracy. Participatory democracy is focused on engaging citizens into public decisions making, but the power structures are those of representative democracy, none the less. That means that citizens’ feedback is used to attune public decisions to the general will and make better, more qualitative decisions. Due to the general lack of trust in institutions and less than obvious institutional accountability, participation is hindered because citizens do not trust their opinions will have real impact over how regulations and bills are going to be passed in the end. Ideas generation is the most accessible citizens’ engagement process with the lowest entry barrier, so participatory democracy should rank high on inclusiveness. However, because traditional power structures are not adjusted to absorb crowd knowledge, citizens often see indeed their feedback lost and the value of participation erodes. One way to avoid this scenario is to find a constant and suitable online framework to generate targeted feedback to all community’s actors.

Direct democracy, by the other hand, changes the traditional structures of power, by allowing citizens to directly make public decisions. Participatory budgets, for instance, are exercises comprised within direct democracy umbrella. Apart from other philosophical, ideological and procedural distinctions, direct democracy strategies seek to ensure the value of participation by vouching citizens’ impact over shaping their social phenomena. Yet, because direct action is often complex and subtle in nature, since affecting a part is affecting the whole, the entry barriers are very often too high for common citizens. One example from US policy making that generated a large body of literature, refers to the four sets of actors that can be a policy influencer: The Average Citizen seen as a median voter, the Economic Elites and the Mass-based or Business-oriented Interest Groups (Gilens & Page, 2014: 564). In testing their level of influence, it turned out that “the preferences of average citizens are positively and fairly highly correlated, across issues, with the preferences of economic elites.” On the same page, organized interest groups “have a very substantial independent impact upon public policy being often deployed against proposed policy changes”.

As such, a seemingly paradox is enacted. Participatory democracy has low entry barriers, but brings too little in stake for citizens to get interested in, while stakes are high in direct, but entry barriers are also high. It could be a step forward in exiting this circular track to design a hybrid process in collaborative law making, that would build on the strengths of both perspectives.

What makes crowdsourcing great for engagement is the easiness and accessible nature of brainstorming processes. What makes commons-based peer production efficient is the deliberative component that’s built in. According to communication theories, policy making may be defined as a structured process of contextual knowledge management in normative discourse production. The vast literature on knowledge and discourse production (Van Dijk, 2005) offers an instrumental definition. Knowledge is as such defined as a set of shared beliefs complying a set of norms which are acceptable and sufficient within a community. Translating this definition within policymaking domain, crowdsourcing is seen as an insufficient process for knowledge building. The general citizens’ community acts on its specific norms in terms of acceptable and sufficient criteria for an idea to be expressed. However, experts and officials obey to different criteria which are perceived to be acceptable and sufficient. Therefore, the knowledge cannot be produced. The same is true for commons-based peer production, as the low number of citizens engaged in such initiatives is symptomatic for how little intersection there is between sufficient and acceptable criteria used by common citizens versus other actors, as officials and experts.  One example of online platforms which takes into account the knowledge disparity of its participants is LiquidFeedback. This free software for political opinion formation and decision-making combines aspects of representative and direct democracy (Wikipedia). Its adoption in some of the German counties of Friesland and Rothenburg has allowed the public authorities to consult their citizens on a wide range of issues. As stated in “The Principles of LiquidFeedback” (Jan Behrens, Axel Kistner et. al. 2014), the platform was not intended to be a general information platform, but “an additional communication channel between citizens and their administration or voters and their representative”.

However, although research shows that there is no actual knowledge without contextual integration, integration of knowledge is, of course, a pervasive process (Goldkuhl & Braf, 2001). As digital communication proliferated, a lot of interest focused on knowledge integration within virtual teams (Alavi & Tiwana, 2002), allowing to draw some lessons for how to produce knowledge in citizens online participation, and especially within processes related to policymaking. The most prominent downside of citizens crowdsourced lawmaking initiative is that it lacks a deliberation component, which is essential for contextualizing knowledge and for knowledge transfer. Adding such a component would allow citizens to readdress their opinions in the light of a broader image and to further contextualize the data they provide, which is based on personal, and therefore, limited experience. One particular case is the long tradition of the “tyranny of participation” from Burundi. Closely related to the country’s rural population, this tradition depicted that a meaningful collaboration and a constant dialogue with citizens could become “painful” in terms of poverty and self-exclusion.

Such a component would bring the phase of citizens’ ideation, closer to citizens’ problem solving processes and group decision making (GDM). There is also a vast literature on GDM (Herrera-Viedma, Martinez, Mata, & Chiclana, 2005; Pérez et. all., 2018; Pérez, Cabrerizo, Alonso, & Herrera-Viedma, 2013). The most appropriate model could be the Group Decision Making with non-homogeneous experts, since crowds are made by citizens with various levels of expertise. The model consists in three phases which are repeated similarly to the rounds in any negotiation process. If the problem is not known, but only the symptoms, than the first round is dedicated to finding consensus over problem definition.

No matter the specific subject, the first phase consists in mapping attitudes towards the topic under discussion. All participants are ascribed with levels of significance, in accordance with their expertise.  All opinions are computed and those which are strongly contradictory with those expressed by the majority are highlighted. All persons are then provided with feedback and helped to readdress their opinions, until a certain level of consensus is reached. Then, participants move towards solution selection phase. By ascribing levels of expertise, it’s possible to offer different types of feedback to those engaged in the process. It’s not probable that high level experts would make an opinion without through consideration, but this is actually the case for most low level experts.

Providing specific feedback to these participants helps them learn about the issue at hand, changes their perspective over it and channels their opinions towards the main consensual line.

By merging GDM in crowdsourced policymaking processes, citizens are engaged in deliberation and pragmatic contextual knowledge building. As GDM processes became mainly automatic, with no need for moderators, they could easily be morphed into collaborative law making procedures. In such a manner, citizens’ legal and administrative illiteracy would be not just reduced, but the process would have a transformative impact over participants.

Of course, further analysis should be done in respects with the main aim of GDM, which is to achieve consensus. In terms of reducing the knowledge and attitude gap between experts and citizens, they could smooth the way especially because they are designed to mold opposite perspectives into a common ground, which is needed in terms of participatory democracy. However, it’s questionable if taming opinions which fell far from the majority points of view is a desirable outcome.


In tapping into their citizens’ wisdom, public authorities should not forget that they have the same goals: improving public service deliveries and policy projects. Citizen engagement regarded as a top-down initiative requires cities or municipalities to provide the necessary tools in order to get them involved in decision-making whereas citizen participation involves high levels of inclusiveness and therefore awareness. These mutual initiatives cannot be boosted if citizen participation platforms and projects do not evolve and adhere to technological innovations like artificial intelligence and NLP. But these kind of policy processes to collect citizens input are not efficient if they lack a technological and collaborative framework. As a consequence, they are supposed to give more time for meaningful interaction with citizens and undoubtedly improve the relationship between them.

Aitamurto, T. (2016). Collective intelligence in law reforms: When the logic of the crowds and the logic of policymaking collide. In 2016 49th Hawaii International Conference on System Sciences (HICSS) (pp. 2780-2789). Retrieved from http://thefinnishexperiment.com/wp-content/uploads/2016/01/CollectiveIntelligenceInLawReforms.pdf.
Aitamurto, T., & Landemore, H. E. (2015). Five design principles for crowdsourced policymaking: Assessing the case of crowdsourced off-road traffic law in Finland. Journal of Social Media for Organizations, 2(1), 1-19.
Alavi, M., & Tiwana, A. (2002). Knowledge integration in virtual teams: The potential role of KMS. Journal of the American Society for Information Science and Technology53(12), 1029-1037. doi:10.1002/asi.10107
Behrens, J., Kistner, A., Nitsche, A., Swierczek, B. (2014). The Principles of LiquidFeedback, Retrieved from: https://principles.liquidfeedback.org/.
Benkler, Y. (2002). Coase’s penguin, or, Linux and the nature of the firm. The Yale Law Journal, 112(3), 369.doi:10.2307/1562247
Cindio, F., & Stortone, S. (2013). A two-dimensional space to frame participatory initiatives and platforms. In Crowdsourcing the Mexico City Constitution. Retrieved from https://oecd-opsi.org/innovations/crowdsourcing-the-mexico-city-constitution/.
Crowdsourcing the Mexico City Constitution. (2018). Retrieved from https://oecd-opsi.org/innovations/crowdsourcing-the-mexico-city-constitution/.
Faria, C. F. S. de. (2013). The open Parliament in the age of the internet: Can the people now collaborate with legislatures in law-making? [E-book]. Madrid, Spain: Câmara dos Deputados, Edições Câmara. Retrieved from  http://bd.camara.gov.br/bd/handle/bdcamara/12756.
Gilens, M., & Page, B. I. (2014), Testing theories of American politics: Elites, interest groups, and average citizens. Washington, DC, USA: American Political Science Association.
Goldkuhl, G., & Braf, E. (2001). Contextual knowledge analysis-understanding knowledge and its relations to action and communication. In Second European Conference on Knowledge Management Proceedings (pp. 197-208).Retrieved from https://pdfs.semanticscholar.org/acc3/a056842823ab129b8bb5c5b8241fda3e2051.pdf.
Herrera-Viedma, E., Martinez, L., Mata, F., & Chiclana, F. (2005). A consensus support system model for group decision-making problems with multigranular linguistic preference relations. IEEE Transactions on fuzzy Systems13(5), 644-658. doi:10.1109/tfuzz.2005.856561
Nelimarkka, M., Nonnecke, B., Krishnan, S. Aitamurto, T., Catterson, D., Crittenden, C., & Goldberg, K. (2014). Comparing Three Online Civic Engagement Platforms using the ‘Spectrum of Public Participation’ Framework.In Internet, Policy, and Politics Conference on Crowdsourcing for Politics and Policy (pp. 1-22).
Nwogwugwu, N., & Ajayi, F. (2015). Educated elites participation in law making and adherence to rule of law in Ogun State (2003-2011). Kuwait Chapter of Arabian Journal of Business and Management Review,  5(4), 55-69.doi:10.12816/0019025
Pérez, I. J., Cabrerizo, F. J., Alonso, S., & Herrera-Viedma, E. (2013). A new consensus model for group decision making problems with non-homogeneous experts. IEEE Transactions on Systems, Man, and Cybernetics: Systems44(4), 494-498. doi:10.1109/tsmc.2013.2259155
Pérez, I. J., Cabrerizo, F. J., Alonso, S., Dong, Y. C., Chiclana, F., & Herrera-Viedma, E. (2018). On dynamic consensus processes in group decision making problems. Information Sciences, 459, 20-35. doi:10.1016/j.ins.2018.05.017
Van Dijk, T. A. (2005). Contextual knowledge management in discourse production. A new agenda in (critical) discourse analysis, 71-100. doi:10.1075/dapsac.13.07dij

Raluca Onufreiciuc
Oana Olariu

Related posts