Informing Knowledge4Policy with an Understanding of our Political Nature
“Understanding our Political Nature”, a recent EU study into knowledge, reason and policymaking, brings interesting, and sometimes challenging, ideas to Knowledge4Policy.
The decision to build the Knowledge4Policy platform (“K4P”) was taken in 2016, the year of Donald Trump’s election to the US Presidency and the UK’s decision to leave the EU. Around then, and possibly not by coincidence, many more people discovered what psychologists had discovered decades ago: people do not always think rationally.
“Humans do not always think rationally. This is not necessarily problematic. What is problematic is to base politics on the assumptions that they do”
The above quote is from “Understanding our Political Nature”, a July 2019 report from the Enlightenment 2.0 research programme, managed by the EU Commission’s Joint Research Centre, which publishes Knowledge4Policy.
Writing it brought together 60 experts in the fields of behavioural science (psychology, neuroscience, anthropology, economics, cognitive-linguistics), social science and the humanities (history, political science, public policy studies, philosophy of science) from around the world to help policymakers better understand:
- how and why emotions, values, identity and reason affect how we think, talk and take decisions on political issues
- the implications for policymaking.
Each of the report’s seven Chapters — spanning disinformation to emotions, via values and identities — first sets out key scientific insights before outlining implications for policymaking.
While each Chapter is fascinating and relevant to K4P, today I’ll focus on those themes most relevant to our current priorities: making K4P more accessible to policymakers, and building successful online communities.
Evidence-based or Evidence-informed?
The report challenges the traditional view of “evidence-based policymaking” in several ways.
To begin with, it points out that the scientists’ role in policymaking — traditionally seen as neutral and ‘unsullied’ by politics — is unrealistic at best, and that the phrase “evidence-based policy” is actually harmful, as it obscures the political and values trade-offs involved in policymaking. It reflects a simplistic view of policymaking as linear, informed exclusively by value-free science: disinterested, impartial, objective, rational and morally neutral.
While this is exactly what most people (including myself) who studied science believe it should be, the reality is more complex. Values can enter any stage of the scientific process, from the moment a researcher explores the background interests animating the field, through to the moment they draw conclusions and frame their results.
“no such thing as a neutral frame; something is included at the expense of something else… Framing of a policy problem & decisions on what evidence to commission or take into account … are in fact political. We need to be more transparent about the role of values in science“
This does not mean science cannot be trusted or that the scientific method is faulty. It does mean that we should be more transparent about the role of values in science.
The report favours the term ‘evidence-informed policymaking’ as it better reflects the fact that policy decisions must also account for non-scientific factors (particularly values) in a very non-linear system “with multiple actors, institutions, overlapping phases and feedback loops”.
This has important implications for both the JRC and the K4P platform, which hosts scientific teams (‘Knowledge Services’) explicitly set up as value-free and policy-neutral. If their work is more political than realised, how we do we reflect that in K4P’s content strategy, architecture and interactive features?
how we do we reflect this in K4P’s content strategy, information architecture and interactive features?
The report suggests several answers, but poses K4P just as many questions. In the interests of transparency, here are some of them.
Openness, Transparency, Trust … and K4P Closed Groups?
Trust, derived from openness and transparency, are central themes of the report, and the subject of an entire Chapter.
They are also central to K4P’s DNA, as (I hope) demonstrated by our Medium Publication.
Our information architecture, moreover, ensures visitors can easily find the evidence underlying the ‘policymaker-friendly’ briefings and syntheses developed for policymakers by the EC’s scientific services.
However, these themes run counter to one of the features most often requested by our Knowledge Services: ‘Closed Working groups’, where they and their scientific partners can work together in privacy using modern online tools.
‘Working together in privacy’, of course, is what experts do every time they assemble in a university lecture hall to discuss their work, so they’re simply asking for better digital tools to do their current job. Nevertheless, this runs squarely into new Commission rules on expert groups designed to promote lobbyin gtransparency, so we are consulting the Commission’s legal experts.
Communities for Collective Intelligence
Closed Working Groups are a special case of a more general feature we have begun developing for Knowledge4Policy: online communities.
As the report points out, policymaking is a collaborative, group enterprise… and not a very effective one.
“policymaking is to a large extent driven by collective processes… this does not inevitably lead to better decisions as groups do not necessarily collaborate effectively”
The solution permeating the report is to develop online communities to improve diversity of views and create higher ‘collective intelligence’.
“a collective’s intelligence is more than the minimum, maximum or average intelligence of individual group members. Collective intelligence is a property in itself”
So what is an online community?
Definitions vary, but my favourite is simple: “a place where members apply their collective intelligence to solve each members’ problems”.
The Knowledge Services have been requesting this from the outset, as online communities offer a better way for our scientists to work with more scientists from more organisations to create better knowledge together.
Understanding audiences’ needs
And yet the report makes me thankful we didn’t rush into building them.
The web is littered with failed communities. Some reasons are technical, many more are organisational, but most are underpinned by a failure to understand and serve their (potential) members’ needs.
If your community is not in the Top 3 Places your audience needs to be, they won’t become active members
Communities demand a lot from their members. While many people read dozens of websites and enewsletters, most do not have time to contribute to many online communities. If your community is not in the Top 3 Places your audience needs to be, they won’t become active members.
And while we launched with a pretty good idea of what scientists wanted from K4P, we found no existing audience research into what policymakers need from a site bringing together policymakers and scientists to promote evidence-informed policymaking. So we commissioned our own:
And what did this research say about policymakers joining online communities? That they would not… unless it was a “closed working group”.
Policymakers are unlikely to join anything other than Closed Groups of members they trust. But Closed Groups could undermine public trust in government and science
I hope you enjoy the irony as much as I do. Policymakers are unlikely to join anything other than a closed working group, where the level of trust between members is high. But closed working groups are exactly what the EC’s transparency rules don’t like… because they risk undermining public trust in government and science.
More Challenges…
This report illuminates other difficulties we’ll face building online communities bringing scientists and policymakers together.
Generally, the best way to build an online community is to find an offline community — regular attendees to an annual conference, or members of a working group — and offer them an online venue to continue their conversation. It’s not enough, but it really helps.
But what happens when your offline community isn’t functioning well? The report explores many well-known reasons for the failure of ‘collective intelligence’ policymaking processes, notably unequal distribution of key information, failure to value expertise, and social phenomena such as groupthink and group polarisation.
Waving an ‘online community’ wand will not magically make these problems disappear
Remember, these are reasons offline policymaking groups fail to develop good policy. Waving an ‘online community’ wand will not magically make these problems disappear.
These problems, however, will make building functional online communities more difficult — the ability of online spaces to magnify groupthink and group polarisation, for example, is now clear.
… and Opportunities
The report’s authors seem fully aware of the scale of the challenge.
“Thinking collectively can significantly improve the quality of political decisions but only if collaborative processes are carefully designed”
Moreover, all the tactics the report sets out to improve ‘collective intelligence’ policymaking are relevant to Knowledge4Policy — for example:
- “increase the diversity of views”: online communities on K4P will, thanks to its focus, bring in a much more diverse participant base than that usually found in ‘Brussels Bubble’ policy circles
- “favour independence of thought in the conversation”: this is a direct response to the problem of groupthink: while skilled and resourced community management will be essential, K4P’s emphasis on evidence will naturally favour independent thought
- “promote decentralised inputs”: allowing members to submit relevant knowledge to K4P has long been a key demand of our Knowledge Services
- “objectively aggregate and synthesise knowledge”: aggregation has been our focus for the first year, resulting in a silo-free knowledge base which nevertheless provides Knowledge Services with high-granularity knowledge management specific to their knowledge domain. The new knowledge interfaces mentioned above, moreover, should soon allow them to better present syntheses of that knowledge.
In summary, the report endorses the idea of online communities on K4P, sets out some of the challenges we will face, and provides some guidance on meeting them.
But then it complicates things further…
Engaging citizens
The report reinforces the view that all levels of government should better involve citizens in its policymaking process, and sets out a range of proven citizen engagement methods.
While citizen engagement in policy development is not a new idea, the report also argues for the active involvement of the scientific community.
“It is not enough to simply strive to bridge the science / policy divide. The future is not through advisers and external counsel but through involvement in a co-creation process… allow interested parties, including citizens, stakeholders, experts and policymakers to work together” — Vladimír Šucha, Director General, Joint Research Centre
While this could be achieved with K4P online communities, it introduces a new audience. And that could be a problem.
Is K4P the right place for citizen engagement?
While I’ve personally been advocating and exploring citizen involvement in EU affairs since 2002 (blogplug), I am not sure that K4P is the right venue, as:
- any website targeting many diverse audiences will reach none of them well
- we are currently building K4P for policymakers and scientists — two very diverse audiences, as our own audience research confirmed
- we therefore already have our work cut out for us. Adding a third audience — citizens — risks the entire project.
Of course, noone would ever be excluded from K4P — as set out earlier, openness and transparency are core values for us. The question is whether we should change our architecture, design and content to help “citizens, politicians and experts to engage on an equal footing”… in the process risking creating a site which is less effective for everyone.
The report gives good reasons to try
For each problem involved in bringing citizens into the policymaking process, the report sets out proven solutions such as deliberative democracy, participatory design, citizen assemblies and citizen engagement methods.
Most are offline processes, but they need an online counterpart, where “argumentative- and/or vote-mapping software to visualise conversations can help increase clarity, visualising arguments, common ground or diverging views.”
Such an approach could also be useful to policymakers, who would appreciate an overview of where there is scientific consensus, and where there is none.
In summary, integrating citizen participation risks making K4P’s content strategy, architecture and interactivity — already complex to serve policymakers and scientists — even more difficult to get right.
On the other hand, it offers a real opportunity to bring science, policymakers and wider society together, which would:
“help citizens and policymakers make sense of complexity in policy and societal issues, increasing understanding about the trade-offs that need to be made for policy-options… provide a better understanding of why people may have different perspectives and capture more accurately citizens’ values, interests and expectations … and enhance representative democracy in the face of populism and public distrust”
It’s an alluring prospect, but such ‘mission creep’ must remain a longterm goal as it cannot be done without significant resources and deep partnerships with organisations able to bring citizens into the conversation.
My instinct, when reading the report, is therefore to stay on course: first, get online K4P communities working for scientists; then find ways of bringing policymakers on board; and only then widen the audience further.
What do you think?
We’d love your thoughts, so don’t hesitate to share them by Responding to this post, submitting your own or by Tweeting to @EUScienceHub.
