Demokratía After the Vote

Who Really Exercises Krátos in a Society of Infrastructure, Capital, and Gatekeepers?

An investigation into modern democracy, the concentration of power, and the institutional genealogy of innovation

Summary

The Greek word δημοκρατία (dēmokratía) is usually translated as "democracy," but its composition is quite direct: dêmos + krátos, that is, power/government of the people. Dêmos did not simply mean "any individual"; it could designate the sovereign body of citizens, while krátos referred to power or dominion.

Modern democracy, however, operates within a social architecture radically different from that of the Greek city-states. The contemporary citizen does not participate only in a political community: they depend on financial, technological, energy, logistical, informational, legal, and administrative systems of enormous scale.

This produces a question that the existence of voting does not resolve:

If the dêmos possesses formal political sovereignty but depends on structures to which it has no direct access and over which it exercises only indirect control, where is effective krátos located?

This article proposes investigating this question through the concept of gatekeeping.

The hypothesis is not simply that powerful organizations exist. That is obvious. The deeper hypothesis is that effective social capacity may be concentrated in structures that control access to the conditions necessary to transform individual action into capacity at scale.

And a second, even stronger hypothesis emerges:

The very innovations that appear to decentralize power may be born, financed, scaled, absorbed, or later controlled by networks that already occupy central positions in the economic, technological, governmental, and financial architecture.

In that case, the question is no longer simply whether innovation is decentralizing.

It becomes:

Does innovation actually displace centers of power, or does it merely renew the instruments through which centers of power exercise that power?

The Problem Begins with the Word Demokratía Itself

There is a contemporary temptation to treat "democracy" as though its meaning were contained within a set of procedures: elections, representation, separation of powers, individual rights, the rule of law, and so on.

These elements are fundamental to describing modern democratic regimes. But they do not exhaust the philosophical question posed by the word itself.

The very formation dêmos + krátos points to something more elementary:

Who possesses the power?

Athenian democracy should not be idealized either. It was a direct democracy restricted to the citizen body and excluded a large part of the population. But precisely for that reason, it helps make visible a question that can become hidden in contemporary democracy:

The relationship between the people and the concrete places where decisions are made.

In Athens, the relationship between citizen, assembly, decision, and authority was relatively visible.

In modern society, the chain can be:

citizen
↓
company
↓
platform
↓
infrastructure
↓
financial system
↓
intellectual property
↓
regulation
↓
state
↓
international institutions.

And each layer has its own rules, specialists, organizations, and mechanisms of access.

The philosophical problem then becomes:

Is being formally sovereign enough to possess krátos?

The Difference Between Formal Power and Effective Capacity

A useful distinction can be made.

Formal Power

This is institutionally recognized authority.

The citizen:

Votes;
Elects representatives;
Can run for office;
Can protest;
Can create a company;
Can publish;
Can organize;
Can invest;
Can found an organization.

None of this should be denied.

Effective Power

This is the capacity to make something happen at a given scale, especially when doing so depends on resources controlled by third parties.

Imagine a person who invents a technology.

They may possess:

the idea + the knowledge + the prototype.

But to transform that into a meaningful social capability, they may need:

capital + a factory + computing power + patents + certification + suppliers + logistics + distribution + contracts + customers + infrastructure + regulatory approval.

The question then changes:

Who controls the conditions that transform potential capacity into effective capacity?

This is where the gatekeeper emerges.

The Gatekeeper

For the purposes of this investigation, we can define a gatekeeper as:

An agent or structure capable of granting, denying, conditioning, or redirecting access to a resource, network, infrastructure, or institution necessary for other agents to transform potential capacity into effective capacity.

This is different from simply being "a powerful person."

The power lies in the gatekeeping position.

Let us consider the following:

A bank can control access to capital.

A platform can control access to consumers.

A cloud company can control access to computing capacity.

A payment network can control access to transaction infrastructure.

A regulatory authority can control legal access to a particular market.

An intellectual property owner can control the use of a particular technology.

A large logistics infrastructure can control physical access to particular markets.

And these functions can overlap.

The Real Architecture Is Much Larger Than a Company

One limitation of thinking about power only as "company X controls sector Y" is that contemporary architecture consists of multiple layers.

We can provisionally represent it this way:

text

                          CULTURE / ZEITGEIST
                                │
                                ▼

CAPITAL ↔ COMPANIES ↔ TECHNOLOGY ↔ INFRASTRUCTURE
    │          │           │             │
    └──────────┴───────────┴─────────────┘
                       │
                       ▼
                   DISTRIBUTION
                       │
                       ▼
                     SCALE
                       │
                       ▼
                    SOCIETY
                       ▲
                       │
                 REGULATION / STATE
                       ▲
                       │
                  POLITICS / GOVERNMENT

But there is another layer that permeates this dynamic:

Human relationships.

Families, friendships, universities, boards of directors, business associations, foundations, law firms, banks, funds, academic institutions, and elite social networks can connect organizations that are legally independent.

This does not prove conspiracy.

But it makes it inadequate to treat each organization as an island.

Social science has for decades used interlocking directorates—individuals who occupy positions on multiple boards—to study the relational structure of corporate power. Recent research continues to reconstruct corporate elites precisely through these networks of positions, rather than through wealth or prestige alone.

The Stronger Hypothesis: The Institutional Genealogy of Innovation

Here the investigation takes a step further.

It is relatively easy to demonstrate:

A technology was born decentralizing and later became concentrated.

But that leaves open a much more interesting question:

Who was present before the concentration?

We need to reconstruct the institutional genealogy of innovation.

For each technology, investigate:

Origin

Who developed it?

Research

Who funded it?

Infrastructure

Who provided computers, laboratories, networks, energy, manufacturing, and data?

Capital

Who financed the company?

Relationships

Who knew whom?

State

Were there public contracts?

Regulation

Who participated in shaping the rules?

Scale

Who enabled the technology to reach millions of people?

Acquisition

Who bought the innovators?

Later Infrastructure

Who came to control the means necessary for its operation?

Outcome

Who came to own or control the capacity that was created?

The most important question may be:

Do the same centers that controlled important parts of the earlier infrastructure reappear in the later infrastructure?

The Hypothesis Does Not Have to Be "A Conspiracy"

This distinction is fundamental.

There are at least four possibilities:

A. Deliberate Coordination

Agents consciously coordinate their actions to achieve a particular outcome.

B. Strategic Convergence

Independent agents discover that their interests coincide and act in similar ways.

C. Structural Continuity

The same networks of capital, people, and institutions continue to occupy central positions, even without central coordination.

D. Emergence

No agent plans the outcome, but network effects, economies of scale, and institutional mechanisms produce concentration.

A serious investigation must be capable of distinguishing these possibilities.

We do not need to demonstrate A to demonstrate B, C, or D.

And we should not automatically use D to explain everything when evidence exists for A or B.

The Internet: A First Extraordinary Demonstration

The history of the Internet is particularly useful because its origins already cross the state, military, universities, and corporations.

ARPANET emerged within ARPA/DARPA. The agency contracted BBN to build the first routers, and the initial network began with four nodes in 1969.

Then came NSFNET.

The National Science Foundation built infrastructure connecting supercomputing centers and universities; in 1987, NSF awarded the contract to build the new high-speed infrastructure to a consortium involving Merit, IBM, and MCI.

NSFNET became the backbone of the North American Internet and was decommissioned in 1995, when commercial infrastructure had grown sufficiently to replace the public backbone.

This allows an important observation:

The commercial Internet did not simply appear spontaneously from the market.

It was preceded by decades of public, military, academic, and corporate research.

This does not mean that "the government invented the Internet alone." Nor does it mean that its commercialization was planned to produce today's giants.

It means something more precise:

The genealogy of the infrastructure that enabled the digital economy passes through highly concentrated institutions long before the emergence of contemporary platforms.

And then another phenomenon occurs.

The open architecture of the Internet does not prevent concentration in infrastructure.

An Internet measurement study published by ACM found significant concentration in DNS and computing: more than 30% of queries to the two ccTLDs studied originated from five major cloud providers.

Therefore:

Distributed protocol ≠ distributed infrastructure.

And:

Individual ability to publish ≠ distributed control of the infrastructure that enables publication at scale.

Social Networks: Decentralized Voice, Centralized Distribution

The promise of social networks appeared even more decentralizing.

Anyone could:

Publish;
Build a community;
Create an audience;
Distribute information;
Produce entertainment;
Organize movements;
Find people.

And that really happened.

But a new layer emerged:

Who decides what will be seen?

Production and distribution are different things.

A person can produce content.

But a platform can control:

Ranking;
Recommendation;
Advertising;
Monetization;
Discovery;
Data;
APIs;
Identity;
Access to an audience.

European legislation began explicitly using the category gatekeeper for large platforms that function as important gateways between businesses and users. The Digital Markets Act includes marketplaces, app stores, search engines, social networks, cloud services, and advertising among the services covered.

And we have a concrete example of the second hypothesis we are investigating: acquisition.

The FTC sued Meta Platforms, alleging that the company maintained its market power through a strategy that included the acquisitions of Instagram in 2012 and WhatsApp in 2014. The case remains the subject of judicial dispute; the FTC appealed in 2026 after a decision favorable to Meta in district court in 2025.

This does not prove the stronger thesis that every acquisition of innovation is an operation of suppression.

But it demonstrates that:

An innovation that emerges outside a dominant platform can later become the property of the dominant structure itself.

That is exactly the kind of transition that institutional genealogy must follow.

Credit Cards: The Infrastructure Behind Transactional Freedom

Credit cards appear to directly increase individual autonomy.

The consumer does not need to carry physical cash.

They can buy immediately.

They can obtain credit.

They can conduct transactions internationally.

But there is a highly concentrated infrastructure behind this apparent simplicity.

In 2024, Visa and Mastercard together represented approximately 85% of general-purpose credit cards in the United States, according to the Federal Reserve.

Another Federal Reserve document records that Visa, Mastercard, and American Express accounted for approximately 96% of U.S. credit-card purchase volume in 2023.

And the U.S. Department of Justice had previously sued Visa and Mastercard, alleging that the common governance structure among member financial institutions limited competition between networks and hindered the development of products and technologies.

More recently, the DOJ again sued Visa for alleged monopolization in debit-network markets.

The philosophical question is not simply:

"Is Visa too large?"

It is:

Does the citizen who possesses the card possess the infrastructure that makes their transactional capacity possible?

The answer is clearly not the same thing.

Cryptocurrencies: The Most Interesting Experiment

Crypto may be the most intellectually interesting case because its decentralizing intent is explicit.

The project is, in part:

To remove certain intermediaries.

But the economic infrastructure built around it once again displays concentration.

Exchanges.

Custody.

Stablecoins.

Funds.

Liquidity.

Mining.

Computational infrastructure.

Bridges between crypto and the traditional financial system.

The system itself begins to create new intermediaries.

And this produces a paradox:

A technology can decentralize the protocol and centralize the economy built on top of the protocol.

The question, therefore, is not:

"Is Bitcoin decentralized?"

That question is simplistic.

The better questions are:

Which layer is decentralized?

Which layer is centralized?

Who controls the interfaces between the two?

Innovation Can Simultaneously Increase Freedom and Concentration

Here we encounter perhaps one of the fundamental paradoxes of technological modernity.

An innovation can increase an individual's capacity:

And increase the center's capacity even more.

There is no contradiction.

Imagine:

Before

Individual: low capacity.

Institution: high capacity.

After the innovation

Individual: high capacity.

Institution: even higher capacity.

The mistake would be to assume:

"If the individual became more powerful, the system became more decentralized."

Not necessarily.

A platform can allow billions of people to publish while simultaneously becoming the world's primary center for distributing that publication.

A payment network can allow billions of people to transact while simultaneously concentrating transaction infrastructure.

The Internet can allow anyone to publish a website while a small number of providers concentrate large portions of computing, hosting, and DNS.

Crypto can allow direct transfer of value while exchanges, stablecoins, and other intermediaries concentrate important parts of the economic system.

Thus:

Freedom at the periphery does not imply decentralization at the center.

The Funnel Problem

A simplistic explanation would say:

there are 100 good ideas;
20 obtain funding;
8 obtain infrastructure;
5 obtain distribution;
3 achieve scale;
1 wins.

But this description hides the fundamental question:

Who built the funnel?

If the same structures have influence over:

Capital + infrastructure + distribution + intellectual property + standards + regulation + scale,

they are not simply observing a process of natural selection.

They participate in constructing the conditions of selection.

This does not automatically mean that they are coordinated to eliminate a particular innovation.

But it does mean that "the market" alone may be an insufficient explanation.

The Most Interesting Power May Be the Power to Modify the Rule Itself

There is a difference between:

"This rule prevents you"

and:

"I possess enough resources to alter how this rule operates."

An individual may encounter a regulatory barrier.

A large organization can:

Hire specialists;
Produce studies;
Lobby;
Participate in consultations;
Challenge decisions in court;
Create alternative structures;
Buy companies;
Change standards;
Fund research;
Wait;
Push for legislative change.

This does not mean that large organizations have unlimited power.

It means that the ability to negotiate with one's own constraints is unequally distributed.

And that inequality is central to our question.

The High-Level Gatekeeper

This is where we need to abandon a naive conception of the gatekeeper.

We are not talking about an employee who simply follows a protocol.

We are interested in the agent or network that has the capacity to decide on strategic access.

For example:

Who receives funding?

Who receives infrastructure?

Who gets an audience?

Who can hire particular talent?

Who can buy a particular competitor?

Who can enter a particular network?

Who can obtain an exception?

Who can participate in creating a standard?

Who can turn an experimental technology into global infrastructure?

This gatekeeper may fully understand the power they possess.

And perhaps it is precisely their ability to operate across different layers that makes them relevant.

The Problem of Apparent Independence

There is another hypothesis here that needs to be investigated empirically.

A power network does not need to appear as a single organization.

It may appear as:

Company A
Fund B
Foundation C
Bank D
University E
Firm F
Agency G
Company H.

Legally, these are distinct entities.

That does not mean they are secretly a single organization.

But relationships between them may exist through:

Ownership;
Financing;
Boards;
Contracts;
Holding companies;
Subsidiaries;
Trusts;
Foundations;
Firms;
Family relationships;
Professional relationships;
Investments;
Debt;
Joint ventures;
Executive circulation.

The methodological question then becomes:

Where does legal independence end and structural interdependence begin?

And this can be studied.

That Is Why We Need to Reconstruct Networks, Not Just Companies

The unit of analysis should not be:

"Company X."

It should be:

Network X.

We can map:

text

           FAMILY
             │
             ▼
         INDIVIDUAL A
         /     │              ▼      ▼       ▼
    COMPANY   FUND   FOUNDATION
       │        │        │
       ▼        ▼        ▼
 INFRASTRUCTURE CAPITAL UNIVERSITY
       │        │        │
       └────────┼────────┘
                ▼
             INNOVATION
                │
                ▼
              SCALE
                │
                ▼
          NEW GATEKEEPER

The question ceases to be:

"Is there a conspiracy?"

And becomes:

"What is the topology of the network that produces the capacity for action at scale?"

That is a much stronger empirical question.

Interlocking: Continuity Can Survive Changes in Names

An important discovery in the literature on corporate elites is that power can be analyzed by following structural positions, rather than individuals alone.

Researchers specifically study networks of interlocking directorates: when the same person sits on the boards of different companies, they create a connection between legally separate organizations. The literature treats these networks as one of the classic tools for studying cohesion, coordination, and the structure of corporate power.

A recent study of the Spanish corporate elite between 1920 and 2020, for example, uses the structural centrality of directors in interconnected networks to investigate continuity and renewal among elites.

This provides a concrete methodology for our hypothesis:

We do not need to find the same individuals for a hundred years.

We can find:

Continuity in positions and connections.

Individuals can change.

Companies can change.

Technologies can change.

Governments can change.

But certain positions within the network may remain extraordinarily stable.

And What About Acquisitions?

Here another fundamental mechanism appears.

A truly threatening innovation may:

Compete;

Be ignored;

Fail;

Be regulated;

Be absorbed;

Be licensed;

Be acquired;

Transform the structure.

Acquisition is particularly interesting because it does not need to destroy the technology.

It can do something simpler:

Transfer control of the technology to another center.

A paper published by NBER in September 2026 analyzed 1,200 startup acquisitions by eight large technology companies. The authors found complex patterns: acquired startups tended to display certain patterns of innovation before acquisition; acquired patents received more citations after acquisition; and the effects cannot simply be reduced to the idea that "acquisition = destruction of innovation."

This is methodologically excellent.

Because it prevents our investigation from becoming a search for examples that confirm a thesis.

We need to ask:

Who controls the innovation afterward?

Not only:

Did the innovation survive?

Bell System: A Historical Laboratory

The AT&T case is particularly important because it allows us to observe a gatekeeper on a historical scale.

The old Bell System combined telecommunications, equipment, infrastructure, and research within a highly integrated structure.

The Department of Justice sued AT&T in 1949. The 1956 agreement did not immediately dismantle the company, but imposed important restrictions and required Bell Labs to license patents to third parties. In 1982, an agreement was reached that led to the breakup of the local operations, implemented in 1984.

Economic research on the 1956 decree found that compulsory licensing produced a lasting increase in innovation built on Bell patents outside the telecommunications sector.

This is extremely important for our hypothesis because it demonstrates that:

A concentration of infrastructure, capital, intellectual property, and network access can effectively condition who is able to innovate.

But it also demonstrates something else:

A gatekeeper can be institutionally broken apart.

And another thing:

Breaking up an organization does not necessarily eliminate the structural function of gatekeeping.

That last question requires following what happened afterward.

The "Same Center" Hypothesis Needs to Be Treated Carefully

It is tempting to say:

"The same players are behind everything."

But that would be a much stronger hypothesis than the evidence currently presented.

What we can test is something more rigorous:

H1

The same individuals appear repeatedly.

H2

The same families appear repeatedly.

H3

The same funds finance successive technological waves.

H4

The same organizations provide infrastructure for successive waves.

H5

The same professional networks connect apparently independent companies.

H6

Dominant companies systematically acquire potentially competing technologies.

H7

Innovation changes the instruments of power without substantially changing its distribution.

H8

Genuinely decentralizing innovations constitute identifiable exceptions.

That last hypothesis is particularly important.

Exceptions May Be More Informative Than the Rule

Instead of trying to demonstrate:

"All major innovations are controlled,"

we can look for:

Which major innovations managed to escape?

And then ask:

Why?

Perhaps they:

Built their own infrastructure;
Found an independent source of capital;
Used open protocols;
Did not require concentrated intellectual property;
Obtained distributed funding;
Found parallel markets;
Created network effects that were initially impossible to capture;
Emerged under exceptional geopolitical conditions.

If we find several robust exceptions, the structural-capture hypothesis weakens.

If the exceptions are extremely rare, that becomes a finding.

And Then We Reach the Problem of Democracy

Modern democracy usually asks:

Who chooses the government?

Our investigation adds:

Who controls the conditions under which the government can act?

And:

Who controls the infrastructures on which government depends?

And:

Who controls the capital necessary for particular policies?

And:

Who controls the technology necessary to implement them?

And:

Who controls the information necessary to form public opinion?

And:

Who controls the platforms through which opinion circulates?

And finally:

Who controls access to the structures that produce scale?

These questions do not invalidate voting.

They make the question of sovereignty much more demanding.

The Paradox of the Modern Citizen

The modern citizen may simultaneously be:

More individually free

and

More structurally dependent.

They can:

publish instantly;
travel;
create a company;
access knowledge from around the world;
create a community;
transfer money;
create software;
invest;
stream video;
build an audience.

All of this is historically extraordinary.

But they also depend on:

energy
→ telecommunications
→ Internet
→ cloud
→ operating systems
→ platforms
→ payments
→ banks
→ logistics
→ intellectual property
→ legislation
→ state infrastructure.

The expansion of individual capacity has not eliminated dependence.

In some cases, it has increased the number of activities that can be carried out through centralized infrastructures.

Efficiency Produces Invisibility

There is an idea here that I consider especially important for this article:

The more efficient an infrastructure becomes, the less visible its architecture tends to be to the user.

When you tap your phone and a message arrives instantly, you do not see:

Antenna → carrier → fiber → routing → DNS → data center → operating system → application → database → CDN → authentication → energy → undersea cables → synchronization systems.

You see:

"I sent a message."

The infrastructure disappears.

The interface produces an experience of simplicity.

But perceived simplicity can conceal an extraordinarily complex architecture.

And this has a political consequence:

A citizen may possess enormous freedom at the interface and little understanding of the structures that make that freedom possible.

The Real Test of a Decentralizing Innovation

An innovation should not be called structurally decentralizing merely because it allows more people to do something.

We need to ask:

Did it decentralize production?

Did it decentralize distribution?

Did it decentralize capital?

Did it decentralize infrastructure?

Did it decentralize information?

Did it decentralize coordination?

Did it decentralize the ability to define rules?

Did it decentralize the ability to achieve scale?

Or did it simply make the periphery more powerful while making the center even more indispensable?

This last possibility is particularly important.

A More Demanding Definition of Disruption

We can now reformulate the proposition we developed earlier:

An innovation constitutes a structural rupture when it changes not only what agents can do, but the distribution of the capabilities necessary to determine who can do it, at what scale, under what conditions, and with access to which resources.

And we can add:

If an innovation increases individual capacity while maintaining—or increasing—the concentration of coordination, infrastructure, capital, and scale capabilities, it may be technologically revolutionary without being structurally decentralizing.

This distinction resolves a series of apparent contradictions.

The Central Problem of Krátos

We then arrive at the question that motivated the entire investigation.

If dêmos-krátos means power of the people, we must ask:

Where is the power that actually produces results at scale?

It is not enough to know:

who votes.

We need to know:

who finances;

who builds;

who connects;

who distributes;

who owns;

who regulates;

who defines standards;

who controls infrastructure;

who can acquire;

who can wait;

who can open or close a gate;

who can alter the conditions of that gate.

Krátos may be formally attributed to the people while materially distributed across a much more complex architecture.

That does not automatically mean democracy is a fraud.

It means that sovereignty needs to be analyzed on more than one layer.

The Question That Remains

The deepest question in this study may not be:

"Do we or do we not live in a democracy?"

That formulation is too poor.

The better question, in my view, is:

"What are the material, technological, financial, institutional, and relational conditions necessary for the dêmos to transform its will into effective capacity for action?"

And then:

"Who controls those conditions?"

And finally:

"Do innovation and economic transformation displace that control, or simply create new forms through which it can be exercised?"

The answer may vary radically across sectors.

Some areas may be genuinely decentralized.

Others may be highly concentrated.

Some structures may be consciously coordinated.

Others may result from incentives and network effects.

And others may combine both.

This differentiation is precisely what historical investigation can reveal.

The Research Program

The next step, therefore, should not be to construct a total theory.

It should be to test it against different historical cases.

An initial program could include:

Cards and Payments

1950–2026

Capital → banks → networks → infrastructure → regulation → concentration.

Internet

1960–2026

DARPA → universities → NSF → telecommunications → privatization → cloud → platforms.

Social Networks

1995–2026

Social web → platforms → venture capital → network effects → acquisitions → concentration of attention.

Crypto

1980–2026

Cypherpunks → Bitcoin → exchanges → stablecoins → funds → institutionalization.

Open Software

1980–2026

GNU/Linux → companies → cloud → corporate adoption → absorption.

Telecommunications

Bell System → AT&T → breakup → reconcentration

A historical laboratory for studying the destruction and reconstruction of gatekeepers.

Distributed Solar Energy

A potentially decentralizing technology confronting power grids, utilities, financing, regulation, and infrastructure.

3D Printing

A potentially distributed production technology confronting intellectual property, materials, industrial scale, and supply chains.

Smartphones and App Stores

Mobile personal computing → Apple/Google → software distribution → identity/payment infrastructure.

AI

Public research + universities + chips + cloud + capital → foundation models → computational infrastructure → contemporary concentration.

The Decisive Criteria

For each case, we need to answer the same questions:

1. Who invented it?

2. Who funded it?

3. Who provided the infrastructure?

4. Who had access before expansion?

5. Who opened access?

6. Who regulated it?

7. Who established the standards?

8. Who controlled distribution?

9. Who acquired potential competitors?

10. Who controlled the technology after it reached scale?

11. Did the innovation reduce concentration?

12. Or did it increase concentration?

13. Who gained new decision-making capacity?

14. Were the actors appearing in the next cycle already present in the previous one?

15. If not, who connected them to the existing centers?

And one cross-cutting question:

Are the observed relationships purely economic, or are they also familial, institutional, educational, professional, and social?

What Could Falsify the Hypothesis?

This is indispensable.

An investigation of this kind is intellectually interesting only if it can produce a result that contradicts its own hypothesis.

We would encounter a serious problem for the hypothesis if we repeatedly discovered that:

major innovations emerge without meaningful connections to existing centers;
remain independent;
build their own infrastructure;
achieve scale without depending on gatekeepers;
effectively displace centers of power;
disperse capital and capacity;
and existing centers are unable to absorb them.

In that case, we would have evidence of genuine structural decentralization.

If, on the other hand, we repeatedly discover:

innovation → centralized capital → centralized infrastructure → acquisition/integration → new concentration,

then the hypothesis gains strength.

And if we discover that:

the networks that financed or facilitated the innovation were already connected to the previous centers of power,

the hypothesis becomes even more interesting.

Conclusion

Perhaps the most peculiar characteristic of modern democracy is that the citizen has never had so much individual capacity while simultaneously never having depended on such an extraordinary quantity of infrastructures they do not directly control.

They can speak to the entire world.

But they depend on networks.

They can sell to the entire world.

But they depend on payments.

They can create software.

But they depend on computers, chips, cloud, and distribution.

They can create a company.

But they depend on capital, infrastructure, legislation, and markets.

They can create a new technology.

But the decisive question becomes:

Who allows that technology to become a social-scale capability?

And then the question of krátos emerges.

Perhaps the problem of modern democracy is not simply to discover who formally possesses political power.

Perhaps it is necessary to map where the effective capacity to transform decisions, knowledge, capital, technology, and organization into social reality is located.

Democracy can be understood as an institutional structure in which the dêmos possesses formal mechanisms of sovereignty.

But the investigation must continue:

Do those mechanisms reach the places where power is actually concentrated?

If they do not, there is a gap between formal sovereignty and material capacity for action.

And that gap is precisely the space this investigation seeks to study.

The final question is therefore not:

"Who governs?"

It is deeper:

"Who can make things happen?"

And deeper still:

"Who controls access to the conditions that make it possible to make things happen?"

Because perhaps that is where krátos really resides.

Sources and Starting Points for the Investigation

Methodological note: the sources above do not demonstrate a single theory of "control." They document different pieces of the problem: concentration, infrastructure, acquisition, corporate networks, regulation, public financing, and technological transformation. The thesis of this article is precisely that these pieces need to be placed within the same investigation without presuming in advance that they form a single coordinated structure. The empirical question is to discover when they do, when they do not, and which mechanisms produce continuity or rupture.