Why Democracy Must Die

Representative democracy is a lossy compression algorithm for public intelligence. We reduce millions of minds, local contexts, and competing values into campaign theatre, party branding, and occasional votes, then act surprised when the output is stupid. This is not just a leadership problem. It is an interface problem.

The irony is that worldwide we have millions of highly intelligent people with brilliant ideas that have the potential to reshape every facet of how we live. These same people almost invariably feel disenfranchised and apathetic, while a tiny minority controlled by vested interests drives us further into the mire.

Democracy in its current form is failing. The open question is what a better interface to collective intelligence looks like now that we have the tools to build one.

The Fear Machine

Humanity is controlled through fear. Just as a small child learns how to manipulate its parents, political leaders know they can persuade us to follow a course by making us afraid of the alternative. Intelligence doesn’t make us immune. We get steered into ratifying policies that serve a powerful handful while the rest pay the bill.

History is full of nations unjustly waging war, supported by majorities made afraid of the consequences of not doing so. The pattern repeats because the interface rewards it. Fear is the cheapest way to collapse a complex question into a binary choice, and binary choices are all the current system can process.

The fear machine used to need infrastructure: broadcast towers, printing presses, state censors. Now we carry the distribution system in our pockets and do much of the work ourselves. The frightened audience volunteers to pass the fear along.

We can sequence a genome in hours but we still choose our leaders the way we did before electricity existed. A species with 21st century tools and 18th century institutions, wondering why nothing works.

It is why democracy must now die.

Back to the Cave

The justification for this is simple.

I wouldn’t suggest electing one person to do the thinking for the other 99, then spending years watching them serve whoever funded their campaign. I’d want the knowledge in that cave put to work. Give someone a task they can do well. Hold them to it. Delegating work is one thing; surrendering our part in deciding what matters is another. Strip the ceremony away and the arrangement we’ve inherited looks insane.

Forum Government

Forum Government requires a public evaluation process against a set of explicit core values. These should reflect our highest potential and could include innovation, empowerment, open cooperation, self-sufficiency and sustainability.

Innovation means every policy is evaluated for whether it creates new capacity or just redistributes existing resources. Empowerment means decisions push power downward, not upward. Open cooperation means no backroom deals, no classified trade agreements, no policy shaped behind closed doors. Self-sufficiency means a nation that can feed and power itself before it trades with others. Sustainability means no policy passes that borrows from the future to pay for the present. These commitments can conflict. That conflict belongs in public, where the people living with the decision can argue about it.

You don’t vote for a person or a promise. You vote for values. Policy is generated by the public and evaluated against those values by qualified panels. The politician becomes a facilitator, not a decision-maker. The values are the constant. The policies are the variable. A government elected on sustainability can’t approve a policy that undermines it without violating the contract. The accountability is structural, not personal.

Start with a public Values Charter. Not vague virtue language, but a ranked set of governing commitments that the public can actually debate and ratify: sustainability, decentralisation, resilience, innovation, public health, whatever the polity decides matters most. Those values become the stable layer. Governments do not get to campaign one way and govern another without visibly violating the charter.

Policy proposals are then submitted into an open reasoning process. They are evaluated by rotating domain panels whose scoring criteria, conflicts, and assumptions are public by default. The public is not asked to pretend to be expert in every field. The public is asked to decide what matters, inspect the reasoning, challenge the assumptions, and judge whether the proposal actually serves the values it claims to serve.

To stop capture, the process has to be adversarial by design: rotating panels, published scoring, conflict disclosure, public challenge windows, mandatory review after implementation, sunset clauses for major policies. If a system claims to represent collective intelligence but cannot expose its own reasoning and revise itself under scrutiny, it is just elite rule with nicer typography.

The obvious objection is that this sounds like technocracy with better branding. It isn’t, or at least it does not have to be. Technocracy asks you to trust experts. A values-based system asks experts to show their work inside boundaries the public has already chosen. The public sets the constant and the experts argue over the variable, which is a very different political structure.

When I first wrote about this as Forum Government in the early 2010s, the technology to implement it was theoretical. It isn’t anymore.

We have built extraordinary tools for gathering information, connecting people and directing their attention. Then we pointed much of that machinery at advertising. Some of the most powerful communications systems in human history, deployed to sell shoes and keep people angry. I’d like to put that engineering to work on decisions we actually have to live with.

Give the values something difficult to do

Imagine a coastal town with a $10 million adaptation budget. This is a fictional exercise, with invented costs. One neighbourhood floods repeatedly. The initial options are an $8 million barrier, a $10 million voluntary relocation programme, or $6 million of drainage work with the rest reserved for maintenance and further mitigation.

Everyone can agree that safety, fairness, and sustainability matter. They can still disagree fiercely about what to build. Some residents want to stay. Others want a way to leave. Businesses need access. People outside the affected neighbourhood help pay. Each option has a different maintenance burden and uncertainty.

The Charter has not solved the decision. It has given the disagreement somewhere explicit to happen.

Start with the option list. Residents must be able to challenge it, because the person deciding what can be considered already has power. Engineers should explain what each option is expected to do and what could change that assessment. Residents should be able to describe consequences the model leaves out. A spreadsheet does not decide which attachment to a place matters.

Then identify who actually decides. Is the forum advising a council, or has it received authority over this budget? What rule turns the discussion into a decision? What is recorded when a minority disagrees? Who implements the result, and what triggers reconsideration if the cost or effect changes?

A public scoring table is useful only if people can inspect its assumptions and challenge the weights. Otherwise the organiser can place a preferred outcome inside the formula and announce that the values produced it.

A majority is not the only test

Now change the problem. Suppose a majority wants to restrict an unpopular minority’s use of a public space. A high agreement score cannot by itself justify the restriction. The community’s rights commitments need a place in the decision, including a way to refuse an option and a way to challenge that refusal.

Or suppose an urgent evacuation leaves no time for the full forum. Someone needs temporary authority to act. The process has to say who, under what conditions, with what record, and when that authority ends. A system that works only when everyone has unlimited time to deliberate is incomplete.

These are different jobs: participation, effective action, and legitimacy. They can support each other, but none automatically supplies the others.

There is useful prior work. The OECD’s study of deliberative processes examines selection, facilitation, learning, transparency, and the relationship to an actual public decision. Stanford’s America in One Room work provides concrete deliberative designs to inspect. We should learn from those rather than pretend a new interface invented collective reasoning.⁴⁵

Who runs the forum?

Facilitators, moderators, software operators, and the people paying for participation all shape the process. An open-source interface does not make that power disappear. A public log does not ensure that a resident with a job and children has time to inspect it.

There needs to be practical recourse: challenge a distorted summary, introduce an omitted option, understand why a contribution was excluded, and ask for a conflict of interest to be reviewed. If a person has no usable way to do those things, transparency can become another form of theatre.

Generated participation makes this harder. Many accounts do not necessarily represent many independent residents. The Sybil problem predates today’s models; a signing key proves control of a key, not membership in a community. Identity checks, privacy, and appeals need deliberate choices at the stages where they matter.⁶

Tools such as Polis can show patterns of agreement and disagreement. Their clusters are analytical outputs, not natural political tribes or an automatic mandate. A summary can help people find the discussion while still needing a way to preserve dissent and correct a distortion.⁷

This is why I would start with a bounded local decision. Record who participated and who could not, what people understood, how disagreement was represented, and whether the eventual action followed the agreed process. Look at the cost and the result after implementation. A pilot should be allowed to teach us that part of the design needs changing.

The proposal still uses public participation, deliberation, ratification, and accountability. Those are democratic ideas. What I want to retire is the assumption that our current representative interface is the final form those ideas are allowed to take.

The Window

For all our recent technological advances, we humans are a pretty pathetic bunch. We have promoted ourselves to our level of incompetency and are quite simply unfit to be in control of this planet. This is what so many of us know in our heart of hearts. It’s not that we’re all rotten, it’s just that the way we govern ourselves belongs in the dark ages and it’s hard to change.

The death throes of the old, tried and false have begun. A window still exists to turn our future around: build the mechanisms for collective intelligence now, while we can test them, challenge them and change them. Waiting for the next leader to be the exception is how we keep renewing the same contract.

This does not start at constitutional scale. It starts with local pilots: digital civic experiments, policy simulation on bounded problems, issue-level delegation, values charters for organisations and communities before nations. The tools exist. The question is whether we use them before the window closes.

We already use explicit values and constraint systems everywhere serious complexity appears. We just refuse to do it in politics, where the consequences are largest. That refusal will look increasingly absurd as our tools for collective modelling, simulation, and coordination improve. The deeper question is not whether democracy deserves to survive in its current form. It is whether we can build a better decision architecture before collapse makes the choice for us.

For there will surely soon come a time when the metaphorical 100 will need them.

References

1. Arrow, K. J. et al. (2008). The Promise of Prediction Markets. Science, 320(5878), 877-878. ↩

2. Posner, E. A. & Weyl, E. G. (2018). Radical Markets: Uprooting Capitalism and Democracy for a Just Society. Princeton University Press. ↩

3. Blum, C. & Zuber, C. I. (2016). Liquid Democracy: Potentials, Problems, and Perspectives. Journal of Political Philosophy, 24(2), 162-182. ↩

4. OECD, Innovative Citizen Participation and New Democratic Institutions (2020). ↩

5. Stanford Deliberative Democracy Lab, America in One Room: Democratic Reform. ↩

6. John R. Douceur, The Sybil Attack (2002). ↩

7. Computational Democracy Project, opinion groups in Polis. ↩