Remodelling Britain’s voting system for the 21st century:
Agile voting for an agile democracy
British parliament and its flawed voting systems have destroyed citizen engagement in politics, damaged the integrity of politicians, and failed to ensure Britain can deal with 21st-century challenges.
In response, technologists have developed blockchain voting apps to ensure British democracy is more transparent, accountable, and fair. Despite this positive move, these blockchain models echo many of the problems inherent in real-world voting systems, such as lack of representation, fake identities, and centralized power.
As a result, blockchain models are unlikely to spark the kind of revolution British democracy needs to thrive and prosper. For this to occur, it’s vital we combine “iterative” (i.e., agile) app development of collaborative games into our current representative democracy system. This would enable citizen-led policies to be incrementally built from the start, rather than market-led solutions being finalized at the end, without citizen input.
This “middle-out” approach not only has the potential to improve citizen engagement and collaboration in politics but also cultivate the kind of agile democracy Britain needs to address complex challenges post-Brexit.
Welcome to the new age of Agile Democracy.
Problem 1: electoral voting
Most people think Britain and the US are democracies. Both have representative governments that run free and fair elections, uphold freedom of speech, ensure all citizens have civil liberties and are equal under the law. Yet both countries are run as electoral oligarchies, with power in the hands of wealthy elites and immigrants scapegoated when policies turn sour (e.g., Trump’s Muslim ban, May’s Windrush fiasco). A quick glance at the meaning of democracy tells us we no longer live in one, if we ever did (Devenney and Woodford, 2017).
While some British citizens may have been wooed by the promise of “taking back control” of parliamentary sovereignty post Brexit, recent political fallout after the referendum suggests control is unlikely. The reality is far more complex. Daily functioning of British parliament is a chaotic affair, randomly flip-flopping between the needs of centralized HQ (Whitehall & Westminster), Big Finance, vested-interested corporations, and popular media, with citizen needs virtually overlooked (Barnett, 2017). Such traits are perhaps not surprising given the high levels of corporate psychopathy in politics and the actions taken to maintain power (Boddy, 2011).
To add to the chaos, Britain faces entirely new challenges post Brexit (e.g., automation, immigration), which will require more citizen engagement and internal flexibility within the parliamentary system. Such demands are a problem for parliament as its current structures are deeply inflexible, having changed little since its inception in 1688. Forged from an unwieldy aggregate of monarchical absolutism, aristocratic privilege, and capitalist energy, British rule remains a system of privilege, consisting of a cabinet government accountable to a parliament of Commons and Lords under the crown (Barnett, 1997).
Britain’s parliamentary system is kept in check by single-winner elections supported by a plurality voting system, aka first-past-the-post (FPTP). In FPTP, voters mark on a ballot the candidate of their choice, and the candidate who receives the most votes wins (Figure 1). Under this winner takes all system, a majority rule is generally the outcome, at constituency and national level (Johnston, 2007). FPTP may have worked decades ago when there were only two parties to vote for (e.g., either Conservatives or Labour). But at a time when citizens increasingly vote for smaller parties, FPTP neither allows people to choose different candidates from within their chosen party or have their choice even represented in parliament (Figure 1) (Hix et al., 2010).
Problem 2: Commons voting
Those who support the flawed FPTP voting system also have to contend with a highly competitive and corrupt voting game played out in the Commons. Indeed, any two-party House of Commons voting system is effectively flawed for two main reasons. First, it presumes democracy is only about ideological clashes between “left” and “right”, yet evidence shows left and right are highly amorphous concepts that mean different things to different people at different times (Wheatley, 2015). Second, the government can only be successfully prosecuted by one political party in control, however, history shows effective government and democratic argument involves the entire public sphere (Hutton, 2010).
In addition to the flawed notion of political binaries and one-party rule, there are complex technical issues of passing legislation (Figure 2). At present, literally thousands of pieces of legislation must be amended otherwise they will be redundant after 29 March 2019 when Britain leaves the EU. With Brexit consuming the lion share of government and parliament’s time, the huge task of amending laws will be rushed and mistakes “undoubtedly” made. The effects of which are unlikely to be noticed until it’s too late. As a consequence, British businesses face a “storm” of flawed and poorly scrutinised laws, rushed through by government ministers with minimal parliamentary oversight post-Brexit (Chapman, 2018).
To ensure the British government and parliament is resilient to new challenges post-Brexit, reforms are needed. Reforms attracting the most support involve the adoption of a multi-member constituency system (aka proportional representation) (Winter, 2016). The problem is that despite the need for reforms to bring about change, there is no evidence proportional representation would reverse the kinds of problems inherent in Britain’s two-party political system (Hix et al., 2010). This is likely because both electoral voting systems are prone to systemic weaknesses that are accelerating over time.
Voting is never truly representative. We assume voting is fair because it attempts to reflect and represent the total population. However, thanks to the quirky British voting system, votes do not translate to an increase in the number of parliamentary candidates representing them (Great Britain: Ministry of Justice, 2008), despite an increasing number of voters choosing to vote for smaller parties such as the Green Party and the centrist Liberal Democrats (Figure 1). This was also seen in the Brexit vote, which initially recorded 3.8% difference in favour of leaving the EU, yet more recent analysis of polls found a majority prefer to remain (Low, 2016).
Voting is a competitive game. We assume competition determines the best and brightest candidates to represent us in parliament. However, the need to win votes ultimately means beating someone else. Under this competitive system, candidates form tribes and viciously enter all-out wars against each other to claim victory. To win, candidates game the system to gain an edge. This means voting is a zero-sum game: there are always winners and losers. A good example of this winner takes all game is the rise of UKIP, and the Conservative-led call for a Brexit referendum to heal increasing dissent on EU membership. Ironically, instead of healing dissent, Brexit has caused more internal cross-party conflict than ever (Shipman, 2016).
Problem 3: Blockchain “peer-to-peer” voting
Technologists have recently questioned whether digital “peer-to-peer” systems could boost relations between citizens and their representatives, thereby increasing transparency, accountability, and fairness. In this new model, dubbed liquid democracy, every citizen can give feedback on policies as they change over time. Even if citizens are unable to vote (from time pressure or lack of expertise) voters can select a personal representative who can cast a proxy vote on their behalf. All votes can then be publicly verified, distributed, and stored on a decentralized registry or “blockchain” which ensures protection and security of data. This means the entire voting process can remain simultaneously anonymous yet absolutely transparent, which is essential given the current challenges of voter fraud and corruption levelled at elections (Ramos, 2015).
One recent example of blockchain voting is open source platform Sovereign by The Democracy Earth Foundation. The Sovereign app allows citizens to make proposals that are debated and voted on by the community. Despite its potential to increase citizen engagement in policy decisions, blockchain voting is deeply problematic. As discussed earlier, all voting systems (whether single or multi-member) have to overcome at least three challenges. First, officials running digital systems need to insulate themselves from potential voter fraud and corruption. Second, even assuming every voter is honest, voting rules themselves can be faulty (e.g., a third-party candidate with little support can influence the outcomes of two main candidates). Third, relates to voter identity. In real life, voters use official IDs and are prosecuted for fraud. However, on the internet, citizens can vote several times simply by creating multiple fake identities.
Internet security threats involving fake identities are referred to in computer science as a “Sybil attack”. Even if a single user presents multiple identities, it can adversely control a substantial fraction of the entire system (Douceur, 2002). This creates huge problems for internet voting systems such as Sovereign, and cryptocurrency consensus algorithms such as Bitcoin and Ethereum. Two methods, “proof-of-work” and “proof-of-stake” have emerged to solve this problem: proof-of-work by showing you’ve done computation; and proof-of-stake by showing you’ve staked tokens as collateral. However, these two options only work because it’s impossible to fake either the computation and the tokens. For internet voting apps where prior wealth shouldn’t matter, both methods are costly and ill-suited (Sills, 2017). One way of preventing these Sybil (fake) identity attacks is to have certified agents verify identities (Douceur, 2002).
Proof of Identity
To verify identities, Sovereign developed a technique of “proof-of-identity.” This requires each voter take a video of themselves, stating their name and other personal data. Other members then compare two randomly selected videos to discern whether both are identical. While this may sound good in theory, in practice such systems are not failsafe. Studies find people are excellent at identifying familiar faces at low-quality but bad at recognising and/or matching unfamiliar faces (Hancock et al., 2000). One study of people judging photographs found “when the target was present, subjects picked the correct person on only about 70% of occasions. When the target was absent, subjects nevertheless chose someone on roughly 30% of occasions” (Megreya and Burton, 2006). In addition to achieving poor accuracy, these kinds of cognitive biases also promote two undesirable outcomes:
1. Out-Group Homogeneity effect: The tendency of people to confuse individuals from other ethnic groups with one another, aka “they all look the same to me” phenomenon (Ackerman et al., 2006). In the context of Sovereign, this may lead to people subconsciously discriminating against minority groups..
2. Change blindness: The inability of people to detect obvious changes after a transient event, such as an eye-blink (Simons and Rensink, 2005). Again, in the context of Sovereign, unless members apply specific attention to superficial changes in facial features, changes are likely to get overlooked.
These biases when combined would see minority groups disenfranchised and certain results rendered true when they’re actually false (i.e., false positives). Thus, instead of empowering citizen representation, the Sovereign could likely take away the rights of real people to vote. Designers of Sovereign have attempted to limit bias by requesting people avoid “eyeglasses, hats, makeup or masks.” Regardless of whether these rules are even enforceable, there are several ways voters can create multiple identities that easily elude human detection (e.g., beards, hairstyles). Even if humans could accurately identify fakes every time, the chance certified agents would be examining two videos of the same member simultaneously is virtually nil.
Recent improvements in machine learning suggest AI could do a better job at detecting Sybil (fake) identities than humans. However, in reality, the same person with similar facial expressions can appear strikingly different with changes in light source (Figure 3). Problems of light variation are exacerbated by additional factors such as facial expression and aging (Belhumeur, 2005). While neural networks make it easier for machines to learn how to recognize and account for light variation, vast time is required for training the system and it’s not always perfect (Prado et al., 2016). Even when timing and accuracy is solved, AI will still need to overcome the wicked problem of recognizing different facial expressions and aging effects (Sharif et al., 2017).
When looking at Sovereign, it’s not clear developers have done enough to remove problems inherent in our current voting system. Many problems are replicated and, in some cases, amplified (particularly those related to discrimination and bias). In this brief review, voting doesn’t appear to be transparent, accountable, or fair under either FPTP or blockchain conditions. In light of challenges, serious questions need to be answered: (a) can we honestly say the British government is effective and well-coordinated, (b) are official decision-makers properly held to account, (c) are our parties and the government machine too easily influenced by vested interests, and (d) do citizens feel their diverse needs and views are even taken in account?
When pondering the above questions, it’s clear real world and blockchain voting systems are flawed. However, we don’t need to throw the baby out with the bathwater. A more pragmatic approach would be to identify best features from both approaches and fuse them into a new system, rather than replacing one flawed system with another. Just as our democratic forefathers did in the Enlightenment. One possible solution is to combine iterative (i.e., agile) app development of collaborative games into our current representative democracy system. This would enable citizen-led policies to be incrementally built from the start, rather than market-led solutions being finalized at the end, without citizen input (Beck et al., 2001). The following section examines the kinds of agile ingredients that may be required to build such as system, and the potential benefits they offer British democracy in contrast to current flawed voting systems.
Solution: agile voting apps using game theory
One way to steer British democracy towards more cooperative ends is to build game theory into the voting system. What makes game theory so useful is its ability to apply memorable stories to communicate complex decision-making between people (aka “games”). The most famous game is the “prisoner’s dilemma” in which two prisoners must each decide whether to plea-bargain by giving evidence against the other.
A more obscure game, the “stag hunt” considers two hunters who must each decide whether to hunt the stag together or hunt rabbits alone. The stag hunt offers players two potential scenarios: (a) either both hunters combine efforts to hunt the stag, or (b) each hunter individually hunts rabbits. Each would prefer to cooperate in hunting the stag, but if the other player’s motives or actions are uncertain, the rabbit hunt is a risk-free alternative. The moral here is that half a stag is better than a pair of rabbits. The stag hunt is central to many cooperative games, where players work with, rather than compete against, fellow players to achieve a shared outcome. Players typically play against the game rather than each other, favouring cooperation over competition. For example, in Pandemic, players must collaborate to stop and cure different disease strains. Another game such as In Dead of Winter, is set in a zombie apocalypse. Here, the game is more complex, requiring players to achieve both a communal victory and a personal goal.
These collaborative gaming systems are not only seen in the gaming world. Many sectors such as technology and healthcare are experienced rapid growth due to the implementation of agile collaboration. Since the 1980s, GlaxoSmithKline has employed agile methods to accelerate the R+D of pharmaceutical drugs. Harvard Catalyst has applied similar methods to accelerate biomedical research. Apple now connects 1,000s of users via their ResearchKit platform to streamline clinical trials. These examples are just some of the success stories where agile tools are helping businesses navigate complex problems and build responsive solutions. Now it’s time British politics developed agile voting systems to fuel stronger relationships between people and politicians, so future policies are fit for purpose and resilient to change over time.
Agile voting systems: 5 key ingredients
1. Build social trust and reputation: Current voting incites cross-party friction. By contrast, cooperative games harness the power of reputation because citizens who can capture the attention of others ultimately gain more influence. This can be achieved by integrating social credit into the voting system, enabling people to focus on social activates they care about and be rewarded for their actions. In this agile system, citizens participate as little or as much as they like. Being rewarded for pleasurable activities also limits the potential for Sybil (fake) identities, because the amount of social credits earned by citizens is likely to be worth more than any incentive to create fake accounts.
2. Build cost factors into objections: Under competitive systems, it is cheap to vote against a policy, and you’re rewarded for it (e.g., increased media attention). This zero-sum-game of Punch & Judy creates more losers than winners, promoting social exclusion from programmes instead of driving inclusion to support future growth. Agile voting turns this on its head, making it virtually free to do nothing and expensive to object. This means voters reserve their objections for issues they’re passionate about and avoid excessive vetoing, so their opinions don’t get discounted by others.
3. Build crowdsourcing tools: Current voting systems engage citizens after market-driven solutions have been already put in place by politicians. Agile voting turns this on its head by adopting a citizen-based approach before any policy agenda is agreed upon by parliament. Integrating crowdsourcing tools into the system would potentially enable every citizen to join open and rapidly evolving interest groups and combine their talents to solve problems they care about. All solutions can be aggregated within the group with the most popular put to a final members’ vote.
4. Build lazy consensus: Introduce the default position of silent approval (aka lazy consensus). This process of not requiring explicit approval from other citizens before implementing new ideas will optimize development, as most policies are not of interest to most people. If any voter does object, a consensus-seeking process begins. Citizens then voice their concerns and listen to each other but try to avoid blocking the proposed action, which results in a stalemate situation
5. Build consequences into the system: Under competitive systems, unfettered growth is deemed good. However, due to resource depletion and commodity price hikes, companies are now moving towards “regenerative” manufacturing models where products are reused multiple times. Regenerative game plans can be integrated into agile voting systems, so citizens see the negative consequences of their actions and choose to make adjustments to initiate a reversal of trends.
Potential agile voting apps to solve political crises
- Bloodbath (Crime): A hypnotic game app turns children into psychopathic killers and citizens have to work together to stop the rising violence, gun crime, and carnage.
- Death Valley (Environment): A scorching heatwave and relentless sandstorm destroy agricultural systems and citizens have to combine available resources to grow food and survive.
- Holocaust (Nuclear warfare): Rapid proliferation of nuclear warheads by global superpowers demand citizens collaborate to stop destruction and slaughter on a mass scale.
- Blackout (Automation): A supercomputer has become conscious and citizens must forge partnerships to prevent it from shutting down the global communication system and taking control.
- Extinction (Healthcare): A deadly virus is rapidly killing men and making women infertile, forcing citizens to work together to eradicate it and avoid potential mass extinction.
- Deluge (Immigration): Fortified walls across Europe have been breached and citizens have to team up and devise intelligence strategies to keep mass migration under control.
- Bedlam (Jobs/Economy): Half the global population are unemployed due to mass automation and citizens have to combine forces to prevent an epidemic of mental illness and suicide.
- Flood (Housing): Rising global floods have prevented the building of new housing and citizens must work together to figure out a sustainable way to rebuild their homes and cities.
Benefits of agile voting apps
A number of features present in agile voting may provide key benefits:
- Agile voting helps cut through the noise: It’s no secret we’re drowning in a tsunami of data. Today’s challenge is not so much finding time to read everything, but rather deciding what to focus on. It’s simply not possible to acquire deep knowledge of all political topics. Agile voting apps potentially give us permission to be curators of our own destiny by focusing on things we love and are good at.
- Failure is the default mode: Citizens are incentivized to collaborate because they want to avoid a failed outcome. For example, in Bloodbath, citizens collaborate to stop rising youth crime, and in Extinction, citizens team up to avoid mass extinction. Unlike competitive games, everyone agrees the default outcome is bad, and are motivated to a collective goal. If one person loses, we all do.
- Bias is towards action: Agile voting games incentivize citizens by specifying undesirable outcomes beforehand, so they have no choice but to take action or lose the game. i.e., “unless you change your habits, bad things will happen”. This contrasts the competitive FPTP voting system which fails to engage many people because most already know their votes aren’t fully represented.
- Fixed timeline: In the real world, no one can predict disasters, and therefore people tend to avoid taking action as they can’t tell whether the threat is imminent, in the near future, or not at all. In agile voting games, players are driven to combine forces within a specific time frame to solve challenges, regardless of whether predictions are accurate or not.
Challenges of agile voting
No voting system is perfect. Human systems are complex and adaptive, making challenges impossible to predict or control. Nevertheless, some key challenges are more likely to emerge than others:
- Lack of representativeness: Some people aren’t able to advocate for themselves directly (e.g., young children, elderly folk, and citizens with severe learning disabilities).
- Hacking the system: Popular mainstream media apps, for example, could influence popular opinion on a topic by promoting or suppressing different types of content.
- Narcissistic leaders: Psychopathic individuals with powerful views and an audience of doting supporters with less social standing will always bias the system.
- Drop-outs: People who don’t feel they have power or can influence the system will leave. This means agile voting apps could cause certain levels of fragmentation.
Summary
British voting systems, whether real-time (i.e., FPTP) or digital (i.e., blockchain) are deeply flawed. However, there’s no need to eliminate certain useful features related to either system. A more pragmatic way forward would be to combine iterative (i.e., agile) app development of collaborative games into our current representative democracy system. This would enable citizen-led policies to be incrementally built from the start, rather than market-led solutions being finalized at the end, without citizen input. This middle-out approach not only has the potential to improve citizen engagement and collaboration in politics, but also cultivate the kind of agile democracy Britain needs to address complex challenges post Brexit.
Of course, the transition from competitive to cooperative voting will not be easy, as vested-interest corporations will want to preserve power. However, if Britain really wants to create a future where its entire population can thrive, radical changes to the current voting system are vital. We need to encourage people to work together: not by making things harder, but with simple alternatives. We need to design new agile voting games that properly acknowledge reputation as the powerful asset it’s become. And we need an agile voting system that doesn’t overwhelm the already heavily burdened attention span of most citizens, but rather respects their desire to focus on things they’re truly passionate about. And we need to do it now.
References
ACKERMAN, J. M., SHAPIRO, J. R., NEUBERG, S. L., KENRICK, D. T., BECKER, D. V., GRISKEVICIUS, V., MANER, J. K. & SCHALLER, M. 2006. They all look the same to me (unless they’re angry) from out-group homogeneity to out-group heterogeneity. Psychological Science, 17, 836–840.
BARNETT, A. 1997. Britain needs a written constitution [Online]. Available: https://www.theguardian.com/commentisfree/2016/nov/30/why-britain-needs-written-constitution [Accessed 3 May, 2018].
BARNETT, A. 2017. The lure of greatness: England’s Brexit and America’s Trump, Unbound.
BECK, K., BEEDLE, M., VAN BENNEKUM, A., COCKBURN, A., CUNNINGHAM, W., FOWLER, M., GRENNING, J., HIGHSMITH, J., HUNT, A. & JEFFRIES, R. 2001. Manifesto for agile software development [Online]. Available: http://agilemanifesto.org/ [Accessed 3 May, 2018].
BELHUMEUR, P. N. 2005. Ongoing challenges in face recognition. Frontiers of Engineering: Reports on Leading-Edge Engineering from the 2005 Symposium.
BODDY, C. 2011. Corporate psychopaths: organizational destroyers, Springer.
CHAPMAN, B. 2018. Brexit: UK businesses face storm of flawed legislation as ministers rush through laws, warn experts [Online]. The Independent. Available: https://www.independent.co.uk/news/business/news/brexit-uk-business-eu-withdrawal-law-legislation-changes-economy-a8296061.html [Accessed 3 May, 2018].
DEVENNEY, M. & WOODFORD, C. 2017. Why the US and Britain are not democracies [Online]. The Conservation. Available: https://theconversation.com/why-the-us-and-britain-are-not-democracies-71745 [Accessed 2 May, 2018].
DOUCEUR, J. R. 2002. The sybil attack. International workshop on peer-to-peer systems. Springer.
GREAT BRITAIN: MINISTRY OF JUSTICE 2008. The Governance of Britain: review of voting systems, the experience of new voting systems in the United Kingdom since 1997 The Stationery Office.
HANCOCK, P. J., BRUCE, V. & BURTON, A. M. 2000. Recognition of unfamiliar faces. Trends in cognitive sciences, 4, 330–337.
HIX, S., JOHNSTON, R. & MCLEAN, I. Choosing an electoral system: a report by the British Academy Policy Centre. 2010. British Academy.
HUTTON, W. 2010. Them and us: changing Britain-why we need a fair society, Hachette UK.
JOHNSTON, L. 2007. Politics: an introduction to the modern democratic state, University of Toronto Press.
LOW, A. 2016. Brexit is not the will of the British people — it never has been [Online]. London School of Economics. Available: http://blogs.lse.ac.uk/brexit/2016/10/24/brexit-is-not-the-will-of-the-british-people-it-never-has-been/ [Accessed 2 May, 2018].
MEGREYA, A. M. & BURTON, A. M. 2006. Unfamiliar faces are not faces: evidence from a matching task. Memory & cognition, 34, 865–876.
PRADO, K. S., ROMAN, N. T., SILVA, V. F., BERNARDES, J. L., DIGIAMPIETRI, L. A., ORTEGA, E. M., LIMA, C. A., CURA, L. M. & ANTUNES, M. M. 2016. Automatic facial recognition: a systematic review on the problem of light variation. International Conference on Human-Computer Interaction. Springer.
RAMOS, J. 2015. Liquid democracy and the futures of governance. The Future Internet. Springer.
SHARIF, M., NAZ, F., YASMIN, M., SHAHID, M. A. & REHMAN, A. 2017. Face recognition: a survey. Journal of Engineering Science & Technology Review, 10.
SHIPMAN, T. 2016. All out war: the full story of how Brexit sank Britain’s political class, HarperCollins UK.
SILLS, K. 2017. Sovereign and the problems with internet voting [Online]. Ordinary Times. Available: https://ordinary-times.com/2017/09/17/sovereign-and-the-problems-with-internet-voting/ [Accessed 2 May, 2018].
SIMONS, D. J. & RENSINK, R. A. 2005. Change blindness: past, present, and future. Trends in cognitive sciences, 9, 16–20.
WHEATLEY, J. 2015. Politics is too complex to be understood just in terms of Left and Right [Online]. Available: http://blogs.lse.ac.uk/politicsandpolicy/politics-is-too-complex-to-be-understood-as-just-either-left-or-right/ [Accessed 3 May, 2018].
WINTER, O. 2016. Our voting system isn’t just unfair — it’s tearing our country apart [Online]. Huffington Post. Available: https://www.huffingtonpost.co.uk/owen-winter/voting-system-uk_b_9391418.html [Accessed 3 May, 2018].
