The first. Will cannot be measured - it can only be produced
A thermometer is accurate regardless of who holds it, because temperature exists on its own. Will is different: there is no "what humanity wants" lying somewhere and waiting to be gauged. There is only what a defined community has decided, by a known procedure, after discussing the matter. That is exactly why elections do not discover the will of the voters but create it: before the count it exists nowhere, after the count it exists.
Hence a procedure that produces will requires someone whose will it is. A polling company can assess the distribution of opinions, and no one forbids it to do so. It cannot produce will, because there is no bearer of that will.
An opinion can be measured without a people. Will cannot be produced without a people. And an opinion gets ignored; a will gets an answer.
The third. The unamendable core is not only ethics; it protects the instrument against capture
The five principles of the core read as values. Look at them as a technical specification.
"A vote cannot be bought, sold, accumulated or transferred irrevocably" is a requirement that the result of a measurement cannot be acquired. "Property is separated from the vote" is there so that weight in a measurement cannot be bought with money. "Power does not accumulate, a mandate does not become an office" is there so that the instrument is not captured from within by those who maintain it. "Leaving is free, and expulsion does not exist" is there so that the denominator cannot be cleared of those who are inconvenient. "The people speaks but does not compel" is there so that the instrument cannot be turned into a weapon, and therefore so that both sides would use it.
Now ask: what protects an instrument belonging to a foundation from capture? The good faith of its board. Good faith is not inherited, is not verifiable from outside, and does not survive a change of board members.
The core is the only design known to us in which the "buy it" attack has nothing to target: there is nothing to buy, because the vote is inalienable. But for such principles to exist, a document standing above the management is needed - that is, a constitutive text that the operator itself cannot rewrite.
A live test of this argument
It is more honest to name it outright than to hint at it.
The largest project for confirming the uniqueness of a person built without a people is World, formerly Worldcoin, created by the company Tools for Humanity. It is set up as a product: confirmation of uniqueness by iris scan, its own token, a reward for passing the check. Technically the problem has been solved, and solved seriously - this is not amateur work.
Then comes the part we cite this case for. Regulators saw not a civic institution but a commercial collection of biometrics, and in three jurisdictions this was set down in decisions anyone can read.
- Spain. On 6 March 2024 the Data Protection Agency ordered the company Tools for Humanity, as a precautionary measure, to stop collecting and processing personal data in the country and to block what had already been collected. The company challenged the measure, and the Audiencia Nacional upheld it.
- Brazil. In January 2025 the National Data Protection Authority prohibited offering cryptocurrency or any other reward for the collection of biometrics, and on 25 March 2025, by a decision of its Board of Directors, rejected the appeal, keeping in force a fine of 50,000 reais for each day of resumed processing. The ground for the decision was precisely the payment: Brazilian law does not recognize consent bought with money as consent.
- Kenya. The High Court, in a decision of 5 May 2025, found a breach of the Data Protection Act of 2019 - the mandatory impact assessment had not been carried out - and ordered the irreversible deletion of all Kenyans' biometric data within seven days.
And not one body in any country treats that registry as a source of measurements of public significance.
There, the instrument works. Its readings carry no weight.
We do not consider that project to be acting in bad faith and do not dispute its technical side, much of which is done better than ours. We point to something else: the same instrument, built without a people, met exactly the fate predicted by the requirements of a constitutive text and of having no owner. It has an owner - which means it can be bought. It has users, not a people - which means the company sets the denominator. And to the question "who authorized you to ask" it has no answer.
The fourth. Who will answer the fortieth question
The instrument needs answers regularly and for years. Ask yourself what will make a person answer the fortieth question in the twelfth year, if the previous thirty-nine changed nothing visible.
Benefit? The return on a single answer is near zero - that is Olson's problem from the annex Why the Count Is Broken again, only now it strikes at the working of the instrument itself. Money? A paid answer devalues the measurement, and paying hundreds of millions of people for ever is impossible. Compulsion? Excluded by the core, and rightly excluded.
What remains is belonging. People turn out to vote although one vote almost never decides anything and taking part is individually irrational. They turn out because it is theirs, and because not turning out means dropping out of the "we".
A user base does not produce that. A people does. An instrument running on voluntary participation for decades requires not motivation but belonging - and the reason for that is an engineering one, not a poetic one.
Who needs this most already today
Not the comfortable resident of a country with working institutions - they have all of this.
According to the Office of the United Nations High Commissioner for Refugees, at the end of 2025 about 118 million people in the world were forcibly displaced, and registered stateless persons numbered about 4.5 million; their actual number is higher, because no one counts the unregistered. The figure changes every year, and we give it with its date so that the reader can see how current it is.
For these people, a confirmable "I exist, there is only one of me, and this is verified outside the state that does not recognize me" is not an abstraction but what they lack.
The injured party - those who do not yet vote and those who are not yet born - has no representation in any political system in the world. Usually the conversation ends there: naming the harm is easy, but there is nothing to repair it with.
Here there is something to repair it with. It follows not from good intentions but from the structure of the text. Article 5 of the Declaration says: "No decision of the Earthlings people is to shift adverse consequences onto future generations." Article 5 belongs to the unamendable core (Article 11), and the principles of the core are not put to a vote at all - this is not a question of a raised threshold, it is a question that is not asked. A decision that contradicts the core is void from the moment it is taken, and neither the passage of time nor its implementation cures it.
Out of this norm a standing procedure may grow: an assessment of the effect of every significant decision on those who come after - with a published opinion and the right of any participant to demand such an assessment. So far as we know, no such procedure exists today in any state or in any international organization. The closest to it are the Future Generations Commissioner for Wales and the Committee for the Future in the Parliament of Finland, but neither assesses every significant decision or gives any participant the right to demand such an assessment. Nor do we have one yet: we have the norm from which it follows, and those are different things.
And one more thing has to be said, or the picture will look prettier than the truth. There is no court behind this prohibition. The Earthlings people has no constitutional court, and we do not pretend otherwise; there is no one to enforce compliance. The force of the norm lies elsewhere: a breach is visible to everyone, neither the passage of time nor implementation cures it, and in any case of doubt the provisions of the core are construed in favour of the person. Such protection has less power of compulsion than a court's but is easier to verify.
Even so, the position remains singular. If the people ever acquires a voice anywhere outside itself, it will be, so far as we know, the only party at the table whose founding rule expressly forbids it to win at the expense of the generations that follow. Not because we are kinder than others, but because the very text that constitutes us forbids it.
Nine ways in which this fails to come about
This section is not here for the sake of balance. Every one of the outcomes listed is possible, some are likelier than the good ones, and for each it is stated what is being done and what we do not guarantee.
1. Taking sides in party politics
What happens. The people - or a noticeable part of it, or its leadership - publicly takes one of the sides in some country's domestic political divide. From that day the instrument is, for half of society, the enemy's weapon, and its readings are rejected along with it.
Why this is fatal. A tool of this type cannot be restored: the case of Doing Business shows that trust is lost entirely and at one stroke.
What is being done. The ban on taking positions on the domestic politics of states is written into the Declaration, not into a set of rules. The people has no opinion about parties, candidates or elections, anywhere, ever.
What we do not guarantee. We cannot prevent participants from speaking on their own behalf, and we must not do so. The risk that someone else's statements will be attributed to the people remains.
2. A question composed to fit the answer
What happens. The wording of one question turns out to be leading - by design or through carelessness - and this comes to light.
Why this is grave. What is devalued is not one measurement but the method: all past and all future ones fall under suspicion.
What is being done. Separating those who formulate from those who manage; publishing the question in advance; the adversarial stage with publication of objections; the ban on edits after voting begins.
What we do not guarantee. That the procedure will withstand pressure if the answer turns out to matter very much to someone very powerful.
3. Low turnout
It stands third, but in importance it comes first.
What happens. The people has a hundred million participants, and three per cent take part in a measurement. The opponent gets the perfect reply, and a fair one: "you speak for a hundred million, and you asked three in a hundred".
Why this is the most dangerous of all. A measurement with low turnout is worse than no measurement at all: it adds no proof but undermines what has already been accumulated.
Where the real cure lies. Not in reminders or appeals - in the question. A person answers when the question touches their life and when something depends on the answer. Hence the two requirements: a planetary question is put at the point where the consequence lands in individual lives, and not where institutions negotiate about it; and no period consists of planetary questions alone - they are always accompanied by questions of the internal life of the people, whose consequence follows at once and for certain.
The other side of this has to be said too, because it matters just as much: if the questions cease to touch people's real lives, interests and difficulties, the point of voting will disappear entirely - and low turnout will be the participants' correct answer, not their fault.
What else is being done. The share of those who voted is always published, and published first, including when it is low. Turnout will not be hidden behind absolute numbers.
What follows from this. The share of those who answer is the chief indicator of the project's vitality, more important than the number of participants. A people of ten million with a turnout of forty per cent weighs incomparably more than a people of a hundred million with a turnout of four per cent.
What we do not guarantee. That a suitable question will always be found. The boundary is deliberately narrow, and questions that are at once planetary and close to individual lives are fewer than one would like. We shall not widen it for the sake of turnout: a people that has started asking about anything and everything in order to get a better hearing has ceased to be what it was created for.
4. Shallowness
What happens. Measurements turn into polls: millions of answers given in ten seconds, without any examination of the question.
Why this is bad. Such a result is no better than an ordinary poll and deserves no greater trust. Scale without deliberation is not the voice of a people but a sum of reactions.
What is being done. The two layers: the panel gets to grips with the question and formulates the options, and the people votes on the prepared options.
What we do not guarantee. That panels can be assembled, that people will agree to spend weeks on the examination, and that the prepared options will be read.
5. The "foreign interference" frame
What happens. A transnational community of tens of millions of people is declared an instrument of someone else's influence. Taking part in it is restricted by law.
Why this is likely. It is an expected reaction, and it has its own logic: an organized transnational presence always raises the question of whose interests it serves.
What is being done. The people holds no positions on the internal affairs of states, funds nobody's politics and supplies a tool, not an opinion. Everything the people does is public and in its own name.
What we do not guarantee. Nothing. That decision is not ours to take.
6. The registry seen as a threat rather than a resource
What happens. A verified database of millions of unique people is perceived as a security problem rather than a public good. There are precedents for restricting similar uniqueness-verification projects in particular jurisdictions - see the World case above.
What is being done. Raw biometrics are not retained, identifying data are not held in the registry, there is no reward for passing the check. The architecture answers the known objections of that kind.
What we do not guarantee. That this will turn out to be enough.
7. Capture of the instrument from within
What happens. Control of the instrument - the choice of questions, the timing, the interpretation - becomes concentrated in the hands of those to whom the result matters. Formally everything is honest; in fact the instrument serves one group.
What is being done. Separation of those who formulate from those who manage, selection by lot instead of appointment, revocability of mandates, publication of the whole trail. The design of the people resists this more strongly than is usual: the core is not put to a vote, a mandate does not become an office, leaving is always possible.
What we do not guarantee. That the protection is absolute. It is not.
8. Success that corrupts
What happens. The instrument becomes influential. The influence begins to seem too valuable to risk for the sake of neutrality. The temptation arises, just once, not to publish an unwelcome result, just once not to ask about what would divide the people, just once to touch up a wording.
Why this is worth recording in advance. Because it happens not to bad people but to those who value what has been achieved, and it is always recognized after the fact.
What is being done. The duty to publish a result whatever it contains, and the full trail of every measurement. The sign of trouble is visible from outside: measurements stop producing unwelcome answers.
9. No one will ask
What happens. The instrument is built, works flawlessly, publishes measurement after measurement - and no one uses them. No citations, no requests, no embedding.
Why this is the most likely of the bad outcomes. It requires no one's hostility. Indifference is enough.
What is being done. The chronicle of non-answers is the only answer to indifference that we have: it turns silence into a recorded fact. It works slowly and may fail.
What we do not guarantee. Nothing. Use cannot be compelled.
What this costs and how to check that we are wrong
The price of the road chosen is not a small one. Slower growth: joining a people is harder than registering. A proposal that is harder to explain. An open legal question that the opponent makes use of. The wariness of states, which a foundation would not have faced. And voluntary limits - no positions on domestic politics, no compulsion - which a foundation would not have taken on. The trade is deliberate, and it may turn out to be a bad one.
So let us also name the condition under which we would be wrong. If in ten years a verifiable poll by some foundation or company is regularly cited by treaty bodies and states, and no one asks "who authorized you" - then the people was superfluous, and we chose an expensive road where a cheap one would have served.
How the measurement itself works, what requirements stand over it and which parts of what is described have already been built are in the document Civic Voice.
