Satellite Crowding and the Class of the Right to Observe

The Enclosure of the Sky: Satellite Crowding and the Class of the Right to Observe

When a company places thousands of satellites in orbit, who pays the bill?

Dear Young Comrades,

This piece is about one of the most important social debates astronomy is having today. First we will give a short summary, then go into detail.

In brief: what is happening?

• Orbit is filling up fast. Today there are more than 14,000 working satellites around the Earth; including dead satellites and debris, the number of tracked objects is approaching 32,000. This number has almost doubled in less than three years. SpaceX's Starlink network alone had passed 11,000 satellites in orbit as of 11 October 2026 (KeepTrack).
• The real wave has not arrived yet. In a study published on 1 July 2026, the European Southern Observatory (ESO) calculated that the applications submitted to regulators add up to more than 1.7 million new satellites. These are not launched satellites but the total scale of proposed projects. One million of that total comes from a single application: SpaceX's plan for an "orbital data centre" for artificial intelligence (ESO).
• The price is paid in three places. Bright streaks and a general brightening of the sky in telescope images; interference at radio telescopes; the risk of collisions and debris in orbit. Added to this is the pollution created by satellites burning up on re-entry into the atmosphere.
• ESO's proposal is a limit. That the total number of satellites, including existing ones, should not exceed 100,000 and that all of them should be too faint to be seen with the naked eye. The study's author, Olivier Hainaut, personally prefers 50,000.
• Who decides? The sky belongs to all humanity, but in practice the decisions on this are made by the FCC, the US communications regulator. On 9 July 2026, despite astronomers' objections, the FCC authorised Reflect Orbital's "space mirror" test satellite.

The basic question is this: When a company places thousands of satellites in orbit, who will bear the cost to astronomical research, to other satellites and to future generations? Our answer, in short: in today's order the company collects the profit, while science, the public and future generations pay the cost. This is called an externality. In Marxist terms, it is the enclosure of a common asset by private capital.

Now let us go into detail.

---

Why is the sky a "commons"?

The night sky is humanity's oldest shared heritage. We built the calendar, agriculture, navigation, religion, philosophy, poetry and of course science by looking at the sky. For this, no one asked anyone's permission, and no one paid anyone rent.

The law, at least on paper, recognises this. Article I of the 1967 Outer Space Treaty says that the exploration and use of outer space shall be "the province of all mankind". Article II states that outer space is not subject to national appropriation by claim of sovereignty, by means of use or occupation, or by any other means. Article VI makes states responsible for the activities of their companies in space.

This treaty is a product of the Cold War. Because each of the two poles feared the other would seize space, they agreed to say "no one can own it". As we described earlier in our Sputnik piece, the space race was a race between states. Today's race is different: on stage now there are companies more than states, and the treaty says "states cannot appropriate" but gives no clear answer to the question "can companies fill it up?"

The result is this: an orbit that legally belongs to no one in practice belongs to whoever gets there first. Satellite frequencies and orbital slots are allocated at the International Telecommunication Union (ITU) largely on a "first come, first served" basis. Ownership is established not by title deed but by occupancy rate.

Satellite crowding in numbers

Indicator	Value
Working satellites in orbit (July 2026)	More than 14,000
Tracked objects, including dead satellites and debris	About 32,000
Starlink satellites (11 October 2026)	11,137
Total of proposals submitted to regulators	More than 1.7 million
Of which, the SpaceX "orbital data centre" application	1 million
Upper limit proposed by ESO	100,000 (Hainaut's preference: 50,000)
New satellites launched per day (2025 average)	About 10
Satellites or rocket bodies re-entering the atmosphere per day	More than 3

Sources: ESO, KeepTrack, ESA Space Environment Report 2026.

The most striking row in the table is the single application for a million satellites. In January 2026 SpaceX applied to the FCC to place up to a million satellites in orbits between 500 and 2,000 kilometres. The justification: "unprecedented computing capacity to power advanced AI models" (DCD). In other words, a significant part of the sky's new load is being demanded not for internet access but for the computing needs of the AI monopolies. We will come back to this detail below.

What is happening at the telescope?

Optical observation: streaks and a glowing sky

Satellites reflect sunlight. Especially at evening and morning twilight, when the ground is dark, a satellite high above still sees the Sun and shines. In a long-exposure telescope image this shows up as a line cutting across the photograph from edge to edge.

The ESO study's findings, in summary:

• If SpaceX's planned network is completed, every image taken by the Very Large Telescope (VLT) in Chile in the second hour of the night will contain dozens of streaks; the loss of field of view could reach 28 percent.
• If the satellites are a little brighter, most of the Vera C. Rubin Observatory's images could become unusable for several hours every night.
• For most of the night hundreds, and at some hours several thousand, satellites will be visible in the sky.
• The diffuse light of faint satellites and the scattered light of bright ones will raise the overall brightness of the sky. This effect has been calculated numerically for the first time in this study.

Why do these streaks matter? Because satellites are far brighter than the objects astronomers are looking for. A streak can hide a distant galaxy, the faint light of an Earth-like exoplanet or an asteroid that could be dangerous to the Earth. The last point shows that the debate is not only about "scientists' curiosity". Planetary defence is a public good too.

ESO's Betty Kioko names the situation plainly: "For optical astronomy, this is an existential threat." Hainaut describes low orbit as "the coastline of the sky". The comparison is apt: coastlines are commons too, and in Turkey we know very well how coastlines are enclosed with concrete.

A mirror in the sky: Reflect Orbital

The most extreme example in the debate is the US start-up Reflect Orbital. The company wants to build satellites with giant mirrors that will reflect sunlight down onto the Earth at night. The target is 50,000 satellites by 2035. By ESO's calculation, these satellites will appear four times brighter than the full Moon to an observer inside the beam, and as bright as Venus to those outside it. The full fleet could brighten the sky overall three or four times over.

On 9 July 2026 the FCC granted a two-year test authorisation to the company's Earendil-1 test satellite, which has an 18×18-metre mirror. The American Astronomical Society and dozens of science and dark-sky organisations had asked for the application to be rejected. The society criticised the fact that not even a formal coordination agreement was required to protect publicly funded observatories (Astronomy).

Let us stop and think here. The darkness of night is a natural cycle thousands of years old. A company wants to turn this cycle into a product and sell "sunlight at night". This is the purest form of capitalism's relationship with nature: every natural cycle is seen as a gap in the market.

Radio astronomy: the noise no one hears

The problem is not only light. Satellites transmit communications, but beyond that they also unintentionally produce electromagnetic leakage. A study carried out with the LOFAR radio telescope in the Netherlands and published in 2023 in Astronomy & Astrophysics detected unintended emission between 110 and 188 MHz, including bands protected for radio astronomy, from 47 of the 68 Starlink satellites observed. This emission is a million times weaker than a mobile phone, but because the satellites are close it can be comparable to, or stronger than, signals from the distant universe. In the words of Benjamin Winkel of the Max Planck Institute: "This worries us not only for the existing networks but much more for the planned ones" (MPG).

For decades radio astronomers have built their observatories behind mountains, in deserts, in "radio quiet zones". Escaping noise from the ground was possible. There is nowhere to escape noise from the sky.

What is happening in orbit?

The collision clock

The satellites themselves are a threat to one another. The "CRASH clock" developed by Aaron Boley of the University of British Columbia and his colleagues calculates how long it would take for the first dangerous collision to occur if all satellites stopped making avoidance manoeuvres. In January 2018 this was about 164 days. In June 2025 it fell to 5.5 days (Scientific American).

What this means is that low orbit is no longer a safe place by itself. It is a system kept standing by constant manoeuvring, constant calculation and constant software. A disruption such as a solar storm, a software bug or a war could start a chain of collisions. Scientists call this, after the NASA scientist who first described it, the Kessler syndrome: an avalanche in which the debris produced by each collision leads to new collisions and the orbit becomes unusable for decades.

Debris multiplies by itself

The European Space Agency's Space Environment Report, published on 14 September 2026, makes an even more frightening finding: fragmentation events are producing debris faster than it falls into the atmosphere naturally. That is, even with no new launches at all, the debris population will continue to grow (ESA).

This is a debt that today's decisions leave to tomorrow's generations. The company filling the orbit today blocks the way of every country, every scientist and every not-yet-born engineer who will want to use that orbit tomorrow.

Pollution falling from the sky

Satellites in low orbit have short lives; new ones are launched every few years and the old ones are burned up in the atmosphere. According to ESA, on average more than three satellites or rocket bodies re-enter the atmosphere every day. A study published in Geophysical Research Letters in 2024 suggests that burning satellites deposit aluminium oxide in the upper atmosphere and that this could delay the recovery of the ozone layer (Scientific American). The question is still being researched, but the direction is clear: as the satellite becomes a "throwaway" product, the atmosphere becomes a rubbish dump.

A class reading: private profit, social cost

So far we have set out the facts. Now let us ask the questions we always ask: In whose hands? Under whose control? For whose benefit? And let us add a fourth: Who decides?

In whose hands: the enclosure of orbit

When Marx describes the birth of capitalism, he describes how the common pastures of England were fenced off and turned into private property. Land that peasants had used in common became the property of sheep-raising landowners. We call this primitive accumulation. In earlier pieces we carried this concept into the digital world and spoke of "digital enclosure": humanity's shared knowledge was seized as raw material for the models of the AI companies.

This is what is happening in orbit too. Low orbit, which belongs to no one, is turning into the de facto property of whoever arrives first and launches most. There is no title deed, but there is de facto occupation. The Outer Space Treaty forbids "appropriation by occupation"; yet filling an orbital shell with tens of thousands of satellites makes it practically impossible for anyone else to enter. This is appropriation without the name.

Marx's theory of rent is useful here too. Whoever monopolises a limited, non-reproducible natural condition, such as the most fertile land or a waterfall, draws a monopoly rent from it. The most favourable shells of low orbit and the most productive frequency bands are exactly such limited natural conditions. The company that arrives early both shuts off these conditions and raises the costs of those who come later.

For whose benefit: externality, or sending the bill to someone else

Economists call it an "externality" when others pay the price of a company's activity. Like a factory dumping chemicals into a river. Satellite crowding is a textbook externality:

Who gets the profit?	Who gets the bill?
Subscription revenue and AI computing power: the company	Telescope images made unusable: the observatories
De facto monopoly over orbit and frequency: the company	Loss of value of publicly funded observatories: the taxpayer
Low-cost "throwaway" satellite model: the company	Pollution from satellites burning in the atmosphere: everyone
Fast growth and share value: the shareholder	Risk of collision and debris: other satellites and future generations
The "sunlight at night" market: Reflect Orbital	Dark skies, ecosystems and biological clocks: nature and society

The most striking row of this table is the second. The Vera Rubin Observatory was built with US public funds, the ESO telescopes with the taxes of the peoples of Europe. This public science infrastructure, the product of decades and billions of euros, is losing value because of a private company's business model. This is the silent transfer of public property to private capital. Without any privatisation tender.

Who decides: one country's regulator, all humanity's sky

Perhaps the most striking point is this: the most important decisions about the sky of all humanity are made by the FCC, the national communications regulator of the US. Even ESO, the UK's Royal Astronomical Society and the International Astronomical Union had to submit their objections to the FCC as "public comments". The Reflect Orbital application received more than 1,800 comments, and the SpaceX application about 1,500. The FCC authorised Reflect Orbital anyway.

A shepherd looking up at the sky in Chile, where the telescopes are built, a highland herder in Anatolia finding the way by the stars, a villager in Africa using the sky as a calendar: none of them has any say in this decision. The sky is global, the decision national, the profit private.

The United Nations Committee on the Peaceful Uses of Outer Space (COPUOS) has kept "Dark and Quiet Skies" on its agenda for several years. But these discussions do not turn into binding rules. While international law moves slowly, rockets take off several times a week.

The AI connection: the sky becomes a data centre

We wrote above that the justification for SpaceX's million-satellite application is artificial intelligence. This detail completes a line we have long been following on our blog. The AI monopolies first made humanity's shared knowledge, that is, the general intellect, into raw material for their models. Then they built giant data centres to run these models; these centres consumed water, electricity and land. Now that the Earth has become too small, they are taking to the sky.

So the same logic of accumulation operates at three stops: the enclosure of the knowledge commons, the depletion of natural resources on the ground and now the occupation of the orbital commons. This is exactly what we mean when we say "the machine is new, the chain is old".

State and capital: civilian satellite, military orbit

Let us also note this: these satellites are not only commercial. Starlink's role in the war in Ukraine showed that a company can affect the course of a war. That Musk refused in 2022 to enable Starlink access around Crimea was recounted in Walter Isaacson's biography of Musk. SpaceX's military arm, Starshield, builds satellite networks for US intelligence agencies. China's Guowang (targeting more than 13,000 satellites) and Qianfan networks are also part of state strategy.

Earlier, in our Roman Telescope piece, we described a science telescope born from a spy satellite mirror. Today the direction is reversing: the sky of science is falling into the shadow of military and commercial infrastructure.

"But isn't internet access important too?"

This objection is serious and must be answered seriously. Yes, billions of people in the world still lack decent internet access. In remote villages, on islands, in disaster zones and on ships, satellite internet meets a real need. We are not against connectivity. In fact we defend internet access as a public right.

But there are three questions to be asked:

First, access on whose terms? Starlink is not a public service but a commercial subscription. A company decides the price of access, its coverage and in which region it is switched on and off. A country's communications infrastructure depending on the decision of a foreign billionaire is not independence but a new dependence.

Second, is this scale really necessary? The million-satellite application was made not for internet access but for AI computing. A rural student's need for the internet and an AI monopoly's appetite for computing are not the same thing. Presenting them as a package makes the needs of the poor into a shield for the expansion of the monopolies.

Third, is there no other way? Fibre infrastructure rolled out through public investment, local wireless networks in public or cooperative ownership, and a limited public satellite service bound by strict rules through international cooperation are all possible. The problem is not technology but ownership and planning.

Seen from Turkey

Turkey is not far from this debate. In September 2023 Elon Musk met President Erdoğan in New York and requested a licence for Starlink; the same month SpaceX officials held a meeting at BTK, Turkey's telecommunications regulator (Hürriyet Daily News). In November 2025, ministry sources stated that "the Starlink issue is not currently on our agenda" (Teknoblog, in Turkish).

Let us draw attention to two points in this picture. First, in Turkey the debate about satellite internet runs almost entirely along the axis of "security" and "control". That is, the question is asked not in terms of "public access" but of "state control". We know what this means in a country full of censorship and access bans. Second, the astronomy debate is entirely absent from public discussion. Yet Turkey too has public investments in the sky, such as the TÜBİTAK National Observatory in Antalya and the 4-metre telescope of the Eastern Anatolia Observatory in Erzurum. The value of these investments falling because of decisions made by another country's regulator is our problem too.

Astronomers and physics engineers in Turkey need to raise their voices in this debate, and professional organisations need to put it on their agenda. The problem of the sky is not a problem for a "Western astronomy club" but a problem of public science and the knowledge commons.

Two views, two skies

The market's sky	The sky of the commons
Orbit belongs to whoever gets there first	Orbit belongs to all humanity; its use is planned and limited
The decision lies with a national regulator and the company	The decision lies with binding international rules, together with the scientific community
The cost is externalised and billed to society	The polluter pays; a fee is charged for the use of orbit
The satellite is a short-lived "throwaway" product	Long-lived infrastructure that can be dismantled, with its return planned
Internet is a commercial subscription	Internet is a public right; infrastructure is in public or cooperative ownership
Night is a gap in the market ("sunlight at night")	The darkness of night is a natural and cultural heritage to be protected
Science works in whatever gaps are left	Science is a founding party to the decision-making process
An orbital data centre for AI	Computing needs are planned according to social priority

What do we want? Concrete demands

In this debate it is necessary to offer not only criticism but proposals. Combining the scientific community's proposals with a class perspective, we put forward the following demands:

• A binding upper limit. As ESO proposes, a science-based ceiling should be placed on the total number of satellites. This limit should rest not on national regulators but on a binding international agreement.
• Standards for brightness and radio emission. Satellites should be too faint to be seen with the naked eye (visual magnitude greater than 7, that is, fainter than 7th magnitude); unintended electromagnetic emissions should be limited. These standards should be a licence condition, not voluntary "good will".
• The polluter pays principle. A price should be placed on the use of orbit. Economists have been discussing this idea for years under the name "orbital use fee". The funds collected should go to debris removal, dark-sky protection and public science.
• A ban on mirrors in the sky. Reflective satellite projects aimed at lighting up the night should be banned internationally because of the harm they do to the natural night cycle and to science.
• The scientific community's right to decide. Astronomers should not be "stakeholders" writing public comments but a founding party to the decision-making process. Protected sky zones should be defined around observatories.
• Public communications infrastructure. Internet access should be recognised as a public right; access for rural and remote areas should be provided through public investment and cooperative models; a country's communications should not depend on a single company's on-off switch.
• Social control of AI computing. Plans to turn the sky into a data centre cannot be discussed separately from the question of what and whom AI computing serves.

Concrete tasks for young comrades

• Look at the sky and take notes. Go somewhere dark one night and watch the sky for ten minutes. How many satellites do you count? Use apps to find out which of them are Starlink. Making a scientific observation is the first step in making the problem stop being abstract.
• Take part in citizen science. There are volunteer networks around the world that measure sky brightness. Contributing to this data is taking part in the public monitoring of the sky.
• Open the agenda in physics and astronomy departments. Make this debate the subject of seminars, talks and papers at universities. Bring astronomy students together at the same table with computing and communications students.
• Take it to the professional organisations. The professional organisations of physics engineers, electrical and electronics engineers and computer engineers should put the sky commons and public communications infrastructure on their agendas.
• Reject the "internet or the sky?" dilemma. When you are told to choose between the two, explain that the problem is one of ownership and planning.
• Think about the knowledge commons together. Orbit, just like humanity's accumulated knowledge, is a commons. We cannot oppose the enclosure of one and stay silent about the other.

---

The right to look at the stars

Human beings learned to think by looking at the sky. Galileo's telescope showed not only the moons of Jupiter but also the cracks in the Church's world view. Throughout history astronomy has been a science that demolished the rulers' narrative that "the Earth stands still and the order is eternal".

Today another kind of rule is trying to close off the sky. This time not the clergy but capital. Not with a banned book but with thousands of bright dots. They are not forbidding us to look at the sky; they are leaving us no sky to look at.

Dear young comrades, seeing the stars is not a luxury but humanity's shared heritage. It is our task not to let this heritage be written into a company's balance sheet, and to leave future generations a dark and clean sky. The sky belongs to no one; it belongs to everyone.

Knowledge belongs to everyone. So does the sky.

---

Sources

• ESO, "Beyond the limit": one million satellites and mirrors in space pose grave threat to the night sky, press release eso2607, 1 July 2026 (study: O. Hainaut, Astronomy & Astrophysics)
• Gizmodo, ESO Study Finds That No More Than 100,000 Satellites Should Orbit Earth, July 2026
• Phys.org, Planned 1.7 million satellites 'devastating' for astronomy: Study, July 2026
• KeepTrack, How Many Starlink Satellites Are in Orbit?, accessed 11 October 2026
• DatacenterDynamics, SpaceX files for million-satellite orbital AI data center megaconstellation, 31 January 2026
• Astronomy, FCC approves Reflect Orbital filing for mirror in space, July 2026
• ESA, ESA Space Environment Report 2026, 14 September 2026
• Scientific American, Satellites Used to Have Months to Avoid Collisions—Now They Have Days
• Max Planck Gesellschaft, Starlink satellite electronics interfere with radio telescopes, 2023 (study: Di Vruno et al., A&A 676, A75)
• Scientific American, Satellite Mega Constellations Could Jeopardize Ozone-Hole Recovery, 2024
• Wikipedia, Guowang
• Hürriyet Daily News, Musk's Starlink seeks to offer service in Türkiye, 23 September 2023
• Teknoblog, Starlink Türkiye 2026: Ministry statement (in Turkish), 10 November 2025
• UN Office for Outer Space Affairs, The Outer Space Treaty (1967)

Related pieces from Knowledge Commons

• The Space Race from Sputnik to Starlink
• To Look Far Is to Look Into the Past
• From Futurism to Techno-Fascism

This piece is published under the CC BY-SA 4.0 licence. You may share, reproduce and build on it with attribution.

Tags: space, science, commons, artificial intelligence, Elon Musk, monopoly capitalism, ecology, law