The Founding of a New Party and the Iron Law of Robert Michels

I. The gap and the obvious reflex

The first part of this reflection showed what caused the FDP to fail: not an election campaign, not a scandal, not a person, but the narrowing of the liberal idea to a client interest. It ended with a finding and an open question. The finding: Germany needs a strong liberal force, more urgently than ever. The question: what might it look like?

The obvious reflex is: then one simply founds a new, better liberal party. One that does not repeat the FDP's mistake. One that once again thinks of prosperity and freedom for everyone rather than for a clientele. One that takes Erhard's legacy seriously. Take the right programme, the right minds, the right name — and this time do it properly.

This reflex is wrong. Not because the programme would be wrong, or the minds, or the name. But because the form is wrong. A new party, however well it begins, is subject to the same laws as the old one. In twenty years it would stand where the FDP stands today — or sooner. Whoever does not understand this will repeat the mistake he believes he is avoiding. Before asking which party, one must ask whether a party can be the form in which a liberal idea can have an effect today at all. The answer this second part develops is: no. And the reason for it is more than a hundred years old.

II. The iron law

Robert Michels was no ivory-tower theorist. He had been an active member of the SPD and had experienced the party from within before turning his back on it. In 1911, in his work Zur Soziologie des Parteiwesens (published in English as Political Parties), he formulated what he called the iron law of oligarchy: It is organisation which gives birth to the dominion of the elected over the electors, of the mandataries over the mandators, of the delegates over the delegators. His thesis in its sharpest form reads: who says organisation says tendency towards oligarchy.

The thesis is not ideological but structural. It does not claim that the people in parties are bad. It claims that the organisation itself, independently of the intentions of its members, leads to the rule of a leadership stratum over the rank and file. Michels names three causes. First, technical necessity: large groups need a division of labour, the division of labour produces specialists, specialists accumulate knowledge, and the full-time functionary soon knows more than the voluntary member; the apparatus becomes indispensable. Second, the psychological component: the leadership wants to lead, and the mass allows itself to be led; the indifference of the rank and file is not a defect but a condition of functioning. Third, the leadership's taking on a life of its own: whoever once has power wants to keep it, and preserving one's own position becomes the real purpose, behind which the original aims recede.

What Michels observed in the SPD in 1911 can be found again in every party of the present. The SPD itself, whose leadership in 2024 pushed through a candidate for chancellor against the majority of the polls because it controlled the apparatus. The CDU, whose chairman emerged from intra-party networks while the members were allowed to put up posters. The AfD, which set out as a movement against the establishment and within a short time produced its own leadership oligarchy, which pushed out its founders one after another. And most clearly of all the Greens.

III. The proof of the Greens

The Greens are the most important case, because they are the only party that deliberately set out to refute the iron law. In the early 1980s they came forward with a whole arsenal of antidotes: rotation of office-holders, so that nobody would entrench himself in an office; separation of party office and parliamentary mandate, so that no concentration of power would arise; grassroots democracy and membership ballots, so that the members would decide and not the leadership; a collective dual leadership instead of a single chair; an imperative mandate binding those elected to the decisions of the grassroots.

Forty years later nothing of all this remains. Rotation was abolished because it was impractical — whoever rotates loses experience and influence, and the party loses its seasoned minds. Grassroots democracy has become the folklore of party conferences. Collective leadership was replaced by media-friendly lead candidates. The Greens today function like any other party, with professional politicians and professional careers. The experiment did not fail through negligence but because of the matter itself. Whoever wants to be effective in parliament must adapt to the rules of parliament; coalitions demand compromise, parliamentary work demands a division of labour and hence hierarchy, media visibility demands personalisation. The system shapes the organisation, not the other way round.

The lesson is hard, but it is unambiguous. It is not enough to found a party with better internal rules. The internal rules do not withstand the pressure of the system. The Greens tried every conceivable antidote, and each one failed. A new liberal party setting out with the best of intentions would have no tools at its disposal that the Greens had not already used without success. It would be subject to the iron law just as much as all those before it.

IV. Settling in

Alongside oligarchisation comes a second process, which works independently of it and reinforces it. One may call it settling in. Every party arose as an answer to a real question: social democracy to the exploitation of workers, Christian democracy to totalitarianism, liberalism to the restriction of freedom by state and church. None of these answers was wrong. The problem does not arise with the answer, but when the answer becomes an identity.

A party that has settled in no longer asks: what is the situation today, and what follows from it? It asks: what corresponds to our self-image? The FDP no longer asked where freedom is under threat today — it administered the answer formulated thirty years ago and equated freedom with the freedom of capital. The self-image needs no reality; it needs confirmation. And criticism that challenges the self-image is not examined but pigeonholed: those are left-wing polemical terms, and with that the matter is settled without any engagement having taken place. Classification replaces argument. The critic has not been refuted — he has been located, and what has been located no longer needs attention.

This settling in is not confined to the leadership. It takes place most stably among ordinary members, for whom the party has long ceased to be a political choice and has become a home. Homes are not chosen because they are optimal but inhabited because they are familiar. The party becomes a congregation, the programme a liturgy, dissent a heresy. Under these conditions renewal of substance is almost impossible, for whoever asks new questions disturbs not only the leadership but the community — and communities respond to disturbance with exclusion, not with discussion. A new liberal party would settle in like every one before it. It is only a matter of time.

V. The false bottom

Michels looked at parties in isolation. What he did not sufficiently take into view is that political oligarchy rests on an economic one. The economy itself is not a democracy; it is organised hierarchically, the owner rules, the employees obey, a few control the means of production. The political apparatuses are embedded in this structure and dependent on it: through large donations, through the revolving door between politics and corporate boards, through lobbyists who write draft legislation, through media corporations that decide who becomes visible. As long as economic power is distributed in this way, a party that is merely one element in this distribution cannot change the distribution.

To this comes a second finding, which aggravates the situation. The state that would have to implement such a change has surrendered its own competence over thirty years. What is described in The Hollowed-Out State applies here as a precondition: a state that no longer knows how to plan a bridge, build software, manage procurement is also in no condition to design and implement a new model of society. It depends on the consultants who live off its incapacity. A party that entered this state would find an apparatus that is no longer capable of action — and would become part of its administration, not its correction.

From this it follows: the problem lies deeper than with the individual parties. It is not that the wrong parties govern, but the form in which politics is organised — the form of the party, embedded in an economic oligarchy, carried by a hollowed-out state. Another party, of whatever quality, changes nothing about this form, because it is itself a part of it.

VI. Shifting the question

If no party in the familiar mode can carry the liberal idea, then the right question is no longer which party, but what takes the place of the party. This is the point at which the reflection leaves the familiar. It is not about a better party, but about something that fulfils the function of the party — bringing people into parliament — without taking on its form and thus its fate.

A party is an organisation with purposes of its own: a programme, an identity, an apparatus, an interest in self-preservation. It is precisely these purposes of its own that trigger the iron law and make settling in possible. What if one created a structure that does not have these purposes of its own — that administers no programme, defends no identity, feeds no apparatus, but solely places a tool in the hands of the citizen? A structure that does not represent but empowers?

This is not a rhetorical question. On beyond-decay.org several interconnected concepts have been developed that attempt precisely this shift. They are not finished solutions but blueprints — working documents open to discussion. But they show in which direction the answer lies once the party drops out as a form. They can be arranged along the two structural ties that still shackle the representative system today: the tie to a party and the tie to a place.

VII. The blueprints

The first blueprint is called Representatives Without Party. The idea is a voters' association with a single purpose: to bring non-party citizens into parliaments. It has no political programme, no position on pensions, climate or taxes. It is a mere vehicle that exists only because electoral law requires a party in order to get onto the ballot paper. It gives those elected no instructions, no party whip, no party line. Each person elected is bound solely by his conscience and to his voters — as Article 38 of the Basic Law provides in any case, but as the practice of the parties systematically undermines.

This blueprint solves the Michels problem at the root. An association without a programme of its own and without an interest in self-preservation cannot become an oligarchy, because it accumulates no power that a leadership stratum could draw to itself. It cannot settle in, because it has no identity that could become a home. The candidates are not selected by an executive board but by the citizens themselves, in an open procedure. Artificial intelligence has a single function in it: to inform the citizen — transparency profiles, consistency checks of public statements, flagging of conflicts of interest, bundled citizens' questions. It does not judge, it does not assess, it makes visible. The decision remains with the citizen.

This first blueprint dissolves the tie to a party, but it leaves a second tie in place. The non-party constituency candidate, too, remains tied to a place: to his constituency, to the interests of his region, to re-election by the local voters. Whoever has a constituency thinks in terms of a constituency — not out of weakness of character but out of the logic of the system. Parliament is therefore not the place where the whole is thought through, but the sum of local interests, administered through party discipline.

The second blueprint is therefore called Mandate Without Place. It proposes a category of office-holders who are not tied to a region: legitimised by a nationwide vote, bound to no party and no constituency, accountable to the electorate as a whole. They do not replace the regional members of parliament but stand alongside them — a small, qualified minority positioned not locally but at the level of the whole state. Within this category two profiles are conceivable, chosen by the candidate himself: the national generalist, who is bound to no particular interest but to the whole — the free mandate of Article 38 in its pure form; and the subject-mandate holder, who publicly commits himself to a policy field in which he brings competence, and who can be measured precisely by that field.

It is the subject-mandate holder in particular who matters for a programme such as Prosperity and Freedom for All. Today's parliament has no specialist competence; it has parliamentary-group discipline. The health politician who votes along coalition lines because his re-election depends on it is not an expert but a loyalist. The subject-mandate holder who has publicly committed himself to a field is more measurable than any other member of parliament in the existing system, because he cannot hide behind the complexity of the whole. A nurse with twenty years' experience has a different but no lesser qualification for health policy than a professor of medical law — and which qualification convinces is decided by the citizens, not by a committee, not by a party. The subject mandate thus anticipates an idea of Johannes Heinrichs, who wanted to divide the democratic system into independent subject parliaments: it realises the same structural principle at the level of the individual office-holder, without presupposing a restructuring of the system.

The third blueprint is called The Transparency Council. It does not start from the individual mandate but from the environment in which all office-holders act. What is proposed is a fifth constitutional body whose task is neither legislation nor jurisdiction nor government, but making things visible: it documents how office-holders vote, compared against their promises; it makes lobbying contacts visible; it produces regulatory impact assessments; it discloses financial flows. It does not sanction, it does not accuse, it establishes facts. What the citizens make of the findings is democracy. Here too AI is a tool, not a judge, and here too it must be a public good — open source, verifiable, not privatised.

The three blueprints are complementary. The representative without party dissolves the tie to a party — he creates an office-holder freed from party discipline. The mandate without place dissolves the tie to a region — it creates an office-holder bound to the whole or to a subject field rather than to a region. The Transparency Council changes the conditions under which everyone acts — it creates the light in which deviation becomes visible. The first frees the individual from the party, the second from the region, the third changes the field on which everyone stands. Together they are the attempt to fulfil the function of the party — to bring people into parliaments and to think the whole — without taking on its form and thus its fate.

VIII. What follows

With this the finding of this second part is complete. Germany needs a liberal force, but it cannot be a party in the familiar sense, because the party as a form is subject to the iron law of oligarchy and to settling in, and because it does not reach the deeper problem — the economic oligarchy and the hollowed-out state. The answer does not lie in a better party, but in a different structure: one that empowers the citizen instead of representing him, and that has no interest of its own that it could turn against him.

What this does not yet answer is the question of substance. A structure that brings non-party citizens into parliament and makes power visible is a form — but it is not yet a programme. The citizens who came into parliament by this route would have to stand for something. And what they would stand for has been named by the title of this reflection from the outset: prosperity and freedom for all. What that means in concrete terms — how freedom relates to the social, what prosperity for all means today, which economic order corresponds to it — is the subject of the third part. This second part had only the task of showing that the form in which this programme has been pursued until now — the party — is no longer fit for purpose. Only once that is settled can one talk meaningfully about the programme.