In Conversation With: Paolo Spada

Each month, PlaceSpeak presents a Q&A with experts in public engagement and civic technology. 

This month, we spoke with Paolo Spada, researcher at the Empatia project. Paolo’s work explores how to optimize participatory systems, complex platforms that combine a variety of digital and in-person participatory processes. Paolo is a member of Participedia (CA), of the research board of the Participatory Budgeting Project (USA), a member of the World Bank’s Digital Engagement Evaluation Team, and the Democracy Matters team (UK). His current projects focus on how to promote better and more meaningful consent forms in digital engagement practices, gamification, and how to improve comments in online news by adapting deliberative practices.

Q: Your new project, Empatia, seeks to integrate multiple types of democratic innovations. What is the history of Empatia and why was it created?

A: Around 2012, we observed a trend emerging around cities utilizing multiple types of democratic innovations. The idea was that if one failed, the other ones could still engage people. So there was a lot of redundancy. Rio Grande do Sul, Brazil built one of the first major participatory systems in which more than 1 million people participated. This system was a combination of Participatory Budgeting (PB), democratic innovations, other forms of public consultation and other traditional forms that are typical of Brazil. In total, there were more than 12 digital processes combined. So we were observing this trend of “the more the better”, in part because it was the easiest way to quantify participation. If you can show you have 1 million participants, it’s a big deal and it’s easy to get funding.

This idea of summing many methods together is straightforward, but there’s an assumption that there’s always a positive interaction between different types of democratic innovations, and we knew that was not true. Take the history of PB. There were many cases in which the introduction of digital participation, instead of reinforcing the process, actually reduced its legitimacy. In Recife, Brazil, the people who traditionally participated face-to-face felt “cheated” because the introduction of online voting allowed many people who had never participated before to alter the decisions made by in-person participants over months, simply with a click. This also mattered because those who participated face-to-face tended to be low-income, while those who participated online were middle-class. So the digital divide, combined with a badly-designed process, generated a lot of criticism.

Given the trend of creating multi-channel participatory processes, we started to study the literature which set the theoretical basis of the Empatia project. We started to construct a highly flexible platform specifically designed to manage and optimize multi-channel systems that combined many different democratic innovations. We are still at the beginning of Empatia, but the City of Lisbon is now launching its new participatory system. For many years, they have had PB, Fix My Street, and some face-to-face public consultation. So they were combining these things already, but in an informal way. What Empatia did was to create a very simple integrated login which allowed people to go participate without having to login to three or four different services. This was the first very basic service. The second and more interesting case, which showcases the potential for these integrated platforms, was a problem with PB in Lisbon. In theory, PB is designed for large ideas: the city invests €5 million in it, and the minimum amount of money dedicated to each project is €50,000. But many people were putting forth small ideas valued below €50,000 and this generated a lot of “noise” for the city to sort through. So what the city is doing with Empatia is an ideation process. If you submit a small idea to PB, you get a notification and are automatically redirected to the other process. We are piloting it now following a design thinking workshop with city staff where we thought about the system together.

Q: What are the key challenges that participants and decision-makers face during the citizen engagement process?

A: For participants, the problem with these processes is a mismanagement of expectations. Participants tend to think that the processes will do more than they actually will, in part because the way we do engagement exaggerates the reality of these processes. We tend to promote citizen engagement as a cure for all ills and try to engage as many people as we can. In reality, the majority of these processes are actually reflexive games. It’s not really participatory democracy, and participants are not called to influence big decisions on the local or national scale. They’re mostly called to participate in a sandbox, where consultations may have impacts but don’t really impact the equilibrium of politics. Sometimes, they are even invited to participate in processes that don’t do anything or even the opposite of what they wanted. Participants are told that they can change the world, change their city, but what they end up getting is a dog park – and that leads to disappointment.

In cases where we’ve observed an impact, it’s difficult to determine whether the process has generated the impact, or an alignment of various factors: an innovative politician, active civil society, media attention, or others. But on average, many democratic innovations do not have a clear impact on public policy.

As for decision-makers, cities often introduce engagement processes not as the flagship element in a political process, but rather as minor, secondary projects that are injected to make people happy, or because there is a bureaucrat who is very interested, or as a pilot. So they’re often not well-funded, or introduced amidst very strong opposition, and have to be constantly justified. In many cases, the organizers cannot risk too much and are afraid of impact evaluation. They also really need to show results that are understandable, so they really push quantity (e.g. number of participants) in all possible ways, such as counting the same participant three times if they come to three different events. The objective there is to be able to respond to the opposition which says that you’re wasting money, “We’ve spent $20,000 but we have engaged 200,000 people.”

This is also why serious impact evaluations are rare. Everyone wants validation, but when you start demonstrating that the process was not as inclusive or as deliberative as they had hoped, then they step back. This is a problem because we lose a lot of opportunities to learn. There is very little research on failure or the negative impacts of citizen engagement. So one of my agendas is to push the idea of learning from failure.

Q: Local governments are increasingly using participatory budgeting (PB) as a way to engage citizens in the decision-making process. What has resulted in its increased popularity?

A: Between 1989 and 2000, Porto Alegre, Brazil was conducting a significant amount of the city’s investments through PB. They had also closed other channels of engagement, so anyone who wanted money from the city was forced to go through PB, which obviously pushed it to unprecedented levels.

Since then, it has spread around the world. PB is particularly powerful because there is a set amount of money from the beginning which people know that they have control over. They see the money spent if the city is doing its job, and they get immediate feedback. When you’re doing consultation and deliberation, even in the best example such as British Columbia’s famous citizen assembly, it took years to see feedback and in the end it failed. It’s really rare for participants in a public consultation to see the direct impact of their feedback. In PB, if you win — and it’s designed so that many people win — you can see the difference that you’ve made. Usually 10, 15 projects are implemented per neighbourhood, and the person who has proposed the project sees work starting within a year, maybe even less. That’s why it’s spreading like wildfire.

It’s also very easy to explain the process: “How would you spend $1 million?” It’s not like explaining the debt, or holding a referendum that would happen a year from now. In fact, it was originally designed by Brazil’s Workers Party as an anachronistic form of gamification — to hook people through PB and identify community leaders that they could then push to other roles within the party and more complex participatory processes. People like to decide how to spend money, and that’s why it’s such a powerful engagement tool. While PB has other flaws, it’s very easy to get people to participate.

Q: What do you think is the biggest barrier to legitimate citizen engagement or public participation processes?

A: It comes back to this idea of false expectations. In most situations, we risk reducing the autonomy of participants by telling them that they’re participating in a certain exercise, which in reality has a different objective than stated. Often, the objective of democratic innovations are to increase the trust of participants in the local government, for electoral reasons, to reduce protest, and more. They’re not meant to change the status quo of politics. These exercises are still valuable, but should be explained as a reflexive game in which participants learn how democracy works.

Another danger to the autonomy of participants comes from the way we design these processes in order to achieve their goals in a short amount of time. Deliberative processes use a lot of psychological mechanisms to create a fake environment because you want to prevent conflict in small group discussions. You want participants to be productive, and you want to raise their sense of efficacy. Facilitators are trained to diffuse conflict because if you only have an hour or 30 minutes to discuss a certain issue, you can’t go through the process of having the conflict explode, manage it, and then finding a solution.

In sum, to obtain “good” outcomes in a certain amount of time, we are really railroading and manipulating the participants. Often the best participatory processes reach the point where participants themselves “hack” them. They change the rules, reshape the agenda, take over, and do something creative with them, because that means that they’re regaining autonomy. But that requires time that often we do not have.

Q: We featured Prof. Graham Smith in our November 2016 Q&A. The two of you will be collaborating on a project aimed at increasing civility in online participation processes. Can you tell us about it?

A: This project started as a replica of an experiment I conducted in 2012. A group within the Italian Democratic Party was interested in conducting an online discussion on electoral reform, so we proposed an experiment in which a treatment group would hold the discussion on the Deliberatorium, developed by MIT researcher Mark Klein. It’s basically a platform that restructures discussion: participants have to add an idea, an argument in favour, and an argument against the idea. Then they can go deeper to another level — adding an argument in favour or against of an other argument. The control group was in a standard forum with multiple topics, and people could create new topics. One expectation of this project was that fewer people would stay engaged in the more structured discussion than in the regular discussion forum. Instead, surprisingly, the level of engagement and retainment was totally identical. We are still quite puzzled by the result. Maybe the higher cost of participating in the complex platform was balanced by the higher quality of discussion, maybe the result is due to the fact that the participants were party activists that were particularly motivated, maybe because our engagement campaign was particularly effective.

So the project with Graham Smith was borne exactly from my attempts to replicate this experiment. I proposed a project to improve online news comments because people cared a lot about this issue. We’re going to adapt this technology for something that will look like Monkey Cage blog (on the Washington Post). The control group would use the standard comments at the end of the blog. The treatment group would have a modified argument mapping system which would allow people to provide comments in the standard way, but also in the map as described before. We’re currently recruiting journalists to help us out, and are speaking with the Coral Project that has created Talk, a new forum platform designed in collaboration with a large community of journalists. Ideally we would like to use Talk as control group.

One of our prinicipal investigators on this project is Michael Morrel from the University of Connecticut, and he’s an expert on empathy. One of his theories is that inducing more empathy increases the quality of discussion. Thus in this project we will also test the impact of an empathy-inducing treatment. This approach, based on perspective taking activities, has been used successfully in psychology to deal with bullies. To our knowledge this would be one the first study to apply this technique to promote civility in news comment environments.