Public Administration CommitteeWritten evidence submitted by William Dutton (ST 02)

I would like to respond to “Question 6. Who is doing the strategic thinking on the UK’s role in an uncertain 21st Century?


Whitehall needs to reach across and outside of the Government for expertise and insight by using the Internet to source expertise, advice and insights of relevance to strategic thinking. Whitehall can foster bottom-up collaboration networks if it considers a number of guidelines:

Do not reinvent the technology.

Focus on activities, not the tools.

Start small, but in ways capable of scaling up.

Modularize, creating doable tasks.

Be open and flexible in finding and going to communities of experts.

Do not concentrate on one approach to all problems.

Cultivate the bottom-up development of multiple projects.

Experience networking and collaborating—be a networked individual.

Capture, reward, and publicise success.

1. The diffusion of the Internet and Web has greatly expanded the potential for distributed collaboration, such as through sharing documents, contributing comments and co-creating information, as demonstrated by the successes of open source software development (Weber 2004). A growing number of visionaries see these initiatives heralding a revolution in how organisations, including governments, will function, by tapping the wisdom of crowds—the idea that the many are smarter than the few (Surowiecki 2004; Tapscott and Williams 2006; Malone et al 2009). These visions have been defined as “crowd sourcing” and “mass collaboration”. However, the very notion of crowds and crowd sourcing is misleading.1 In order to capture distributed intelligence, networks of individuals must be cultivated and managed. As I have argued elsewhere (Dutton 2008), they are not crowds. Networking platforms and management strategies must be carefully developed to capture the value of distributed expertise.

Expert advice and insight versus citizen consultation

2. Discussion of crowd sourcing in the public sector, such as through conceptions of “Wiki government” or “collaborative democracy”, is complicated by the potential to blur distinctions between citizen consultation and expert advice. Noveck (2009: 17) defines collaborative democracy as “using technology to improve outcomes by soliciting expertise (in which expertise is defined broadly to include both scientific knowledge and popular experience) from self-selected peers working together in groups in open networks.” This is a useful definition, but does not make a sharp distinction between two very different roles that networking can play in government.

3. One is gauging opinion, which comes closest to “collaborative democracy”, that might ask citizens to respond to policy options, for example, on the basis of their experience. The other is engaging expertise, which might be based on scientific, technical, or experiential knowledge, such as being at the location of a problem. Using the Internet to gauge opinion is primarily focused on citizen consultation, while soliciting expertise online is focused on obtaining expert advice and insights. Understanding how to engage and respond to expertise can be as essential as consultation to the vitality of democratic institutions and processes, particularly with respect to strategic thinking.

4. The very legitimacy of decision-making in a liberal democracy depends on a government’s responsiveness to public opinion. Even statesmen, taking positions in opposition to public sentiment, would seek to carry the public. Governments in liberal democracies have traditionally viewed their citizens as constituents and have thus sought to gauge and consult public opinion. Public opinion polls, committee hearings, and consultation exercises are largely geared to understanding the balance of public opinion concerning policies and decisions. New information and communication technologies (ICTs), such as the Internet, enable more direct and frequent patterns of consultation.

5. However, citizens are more than constituents, whose opinions are equally legitimate. Citizens also have the potential to be experts on particular issues, where some citizens have more expertise than others, such as when they possess specialised knowledge or particular experience relevant to a specific subject. Viewed as experts, the challenge for government is not only to air public issues and gauge public opinion. The problem is also to find relevant experts, on the basis of merit and a spirit of voluntarism, wherever they live or work. The next problem is to find ways to bring their expertise to bear on a particular question in a timely and effective manner.

Expertise and the Internet

6. Experts are individuals who have gained the experience or skills to be judged as authorities by others knowledgeable about a particular area (QED). Citizens are not necessarily experts, but any given citizen might conceivably have expertise in some specific areas. Experts could be citizens of a particular nation, but many might not be. In any given area, not all the experts are citizens of your country. Moreover, not all experts are equal as some experts might be viewed as superior to others. In such ways, getting the advice of experts is very different from consulting the opinions of citizens.

7. The expert has long been a critical aspect of governance, from Machiavelli’s The Prince, written in 1513, to the present day, political scientists have wrestled with the role of experts in governance, particularly in democratic regimes (Benveniste 1977). Of course, expertise is already embedded in routine practice, such as when governments hire consultants, conduct studies, or build models to advise public officials on particular issues. However, the Internet provides mechanisms to create “distributed problem-solving networks”2 that can complement, if not substitute for, in-house expertise and paid consultants, to provide timely and effective advice in ways that could reduce the costs to government, while engaging citizens in new and meaningful ways in the process of governance.

Consider an increased use of collaborative tools for citizen sourcing advice

8. The Internet can be used to form various types of networks, which I have called “collaborative network organisations” (CNOs), that can perform a variety of functions important to bringing expertise to bear on policy and decision-making. There are a number of efforts to develop distributed problem-solving networks, aimed at networking distributed intelligence. Such networks should be used by Whitehall to harness distributed expertise, building on the lessons learned from early cases, including the key opportunities and risks to those who seek to employ them in the public sector. Specifically, I would recommend that governments should utilise CNOs to inform policy and decision-making at all levels, such as by nurturing a series of pilot projects, orchestrated by a set of platforms, guidelines and policies that will enable bottom-up initiatives that can engage experts, be well managed, and inform policy and decision-making.

Champions within government can start bottom-up CNO initiatives if they

9. Do not reinvent the technology. Tailor existing software, such as MediaWiki,3 rather than creating a homegrown system from scratch. The Internet is enabling increasing numbers of individuals to have access to the same content and the same tools. It is increasingly viable to use open source tools on the Internet and Web to build systems that are accessible to a wider population, and do not depend on in-house IT expertise and resources.

10. Do not focus on specific tools, such as Web 2.0 or social networking. Despite the popularity and cache of particular applications, it is important to focus on the activity that will be supported: sharing, contribution, or co-creation (Table 2, page 12) and bring the tools together to support it. Generally, most collaborations will want to move to the ability to co-produce documents, so tools that enable all of these activities are useful to build in from the start. The tools should follow from the activities you seek to support.

11. Start small, but with a design that is scalable. Many fears that surround distributed public intelligence stem from confusion over what it is, or major misunderstandings about what it will do. In discussing a policy blog with a senior official in a public agency, it became clear that he imagined a personal blog about what people did during the day. It became evident that he needed to see a mock up of the blog so that he would better understand what the proponents had in mind. In the realms of “crowd sourcing” and “mass collaboration”, advocates cannot assume that everyone understands exactly what is being proposed. By starting small, opponents and detractors can learn more about the activity and either worry less or actually become enthusiasts.

12. Modularize. Finding the right level for modularizing tasks is key in two respects. First, it regulates the difficulty of participation, shaping the success of any online collaborative activities. Asking questions or posing problems that are too difficult will undermine the likelihood of anyone participating. This is an art that will require experimentation. Secondly, by modularizing tasks, and focusing on specific issues that are aspects of a larger problem, the exercise reveals less about the nature of the big questions, which an agency may wish to protect. In such ways, modularizing tasks increases participation and lowers the risks of citizen sourcing of expertise.

13. Be flexible in where you go for expertise. For some questions, there may be a strong community of experts, making it most sensible to bring your question or problem to that community of users. In other cases, there may be no community, making it useful to use a platform, such as Intellipedia, to build a community around a particular area of expertise. However, it would be a major social undertaking as it would require government to attract users to this platform and to build a community of users around it. An alternative is for government to go to communities of users with relevant expertise. Of course, both strategies could be pursued simultaneously.

14. Do not concentrate on one solution to all problems. Wikipedia covers a wide range of topics, but it is creating and maintaining an online encyclopedia. It has a clear focus. As Wikipedia began to be used for reporting breaking news, the team set up a separate space, called WikiNews.4 Different communities of users are likely to frequent these different parts of the wiki, with each facing different problems, such as the priority placed on the timeliness of the news, as compared with encyclopedia entries, for which timeliness is less critical.

15. Cultivate the bottom up development of multiple projects. With top management support, many open technical platforms, and a set of policies and guidelines for users, a government could cultivate the development of a wide array of CNOs. Some large corporations have literally hundreds of CNOs within the firm. Each needs a champion, and these champions cannot be dictated from the top down. Networked individuals need to sense a value in networking on particular topics, and some need to take leadership roles, such as being a champion for a particular initiative. For such reasons, their formation is an emergent phenomenon, but one that can be cultivated by top management support and policies and guidelines that are welcomed by the users. Anyone in government should know that they might be able to go to the Internet or to a particular community on the Internet to get information, ask a question, or to solve a problem.

16. Get personally involved in distributed collaboration as a “networked individual” and encourage your colleagues to experience this process. It is particularly important that managers or professionals within a particular institutional setting gain this experience. This is how people learn how to use networks and capture the value of CNOs for your organiSation.

17. Finally, it will be important to capture, reward, and publicise best practice—success stories. These will help shape support for distributed networking, and also provide lessons for those leading other networks. This need not be a difficult job of documenting case histories, but simply acknowledging success stories that others can see as easily as by clicking a mouse and going to the Website.

October 2011


Benveniste, G. (1977), The Politics of Expertise, 2nd Edition. San Francisco, CA: Boyd & Fraser.

Dutton, W H. (2008), “The Wisdom of Collaborative Network Organizations: Capturing the Value of Networked Individuals”, Prometheus, 26(3), September, pp 211–30.

Dutton, William H. (2010), “Networking Distributed Public Expertise: Strategies for Citizen Sourcing Advice to Government”. One of a Series of Occasional Papers in Science and Technology Policy, Science and Technology Policy Institute, Institute for Defense Analyses, Pennsylvania Avenue, Washington D.C., August, 2010. Available at SSRN:

Malone, T W, Laubacher, R, and Dellarocas, C. (2009), “Harnessing Crowds: Mapping the Genome of Collective Intelligence”, MIT Sloan Research Paper No. 4732–09, 3 February. Cambridge: Center for Collective Intelligence, Sloan School of Management, MIT. See:

Noveck, B S. (2009), Wiki Government: How Technology CaN Make Government Better, Democracy Stronger, and Citizens More Powerful. Washington DC: Brookings Institution Press.

Surowiecki, J. (2004), The Wisdom of Crowds: Why the Many are Smarter than the Few and How Collective Wisdom Shapes Business, Economies, Societies and Nations. New York: Doubleday.

Tapscott, D, and Williams, A D. (2006), Wikinomics: How Mass Collaboration Changes Everything. New York: Penguin.

Weber, S. (2004), The Success of Open Source. Cambridge, Mass.: Harvard University Press.

1 In this sense, I would not agree with the thrust of Malone et al (2009). There is not a strict dichotomy between hierarchy and crowd sourcing as CNOs are managed networks albeit not strictly hierarchical in all activities.

2 This label is derived from terminology of InnoCentive, and adapted by Paul A David, where it defined a project at the Oxford Internet Institute, entitled “The Performance of Distributed Problem-Solving Networks”.



Prepared 20th April 2012