Innovation in the launching market:
Challenges to the liability regime of
outer space
Master’s Thesis Spring 2016
Table of Contents
1. Introduction ... 3
1.1. Scientific background to the dangers of damage caused by space objects ... 4
1.2. Innovation in the launching market ... 7
1.3. Methodology ... 8
1.3.1. Examination of liability for damage caused by space objects ... 8
1.3.2. Framework for the comparison between outer space and the high seas ... 9
2. The who: coverage of innovative means to gain access to outer space ... 11
2.1. Coverage under outer space law ... 11
2.1.1. Ordinary meaning of ‘launching State’ ... 12
2.1.2. Context, and object and purpose of the definition of ‘launching State’ .... 15
2.2. Coverage in outer space compared to coverage on the high seas ... 17
2.3. Interim conclusions ... 19
3. The what: definition of damage ... 20
3.1. Coverage under outer space law ... 20
3.1.1 Scope and definition of damage under outer space law ... 20
3.1.2. Obstacles posed to ADR operations by the outer space liability regime ... 21
3.2. Coverage in outer space compared to coverage on the high seas ... 22
3.3. Interim conclusions ... 24
4. The how: nature of liability ... 24
4.1. Coverage under outer space law ... 24
4.2. Coverage in outer space compared to coverage on the high seas ... 26
4.3. Interim conclusions ... 28
5. Conclusion and propositions ... 28
6. List of Abbreviations ... 32
1. Introduction
Liability for damage caused by space objects, so far, has been invoked once.1 This thesis argues that the absence of more claims is due to the nature of the liability regime for damage caused by space objects rather than to the absence of damage caused by space objects. The risk of such damage occurring is on the rise.2 With innovative launching technology granting access to an ever-bigger crowd of space operators, this risk can be expected to continue to grow. Any shortcomings in the liability regime for damage caused by space objects – established in the 1972 Convention on International Liability for Damage Caused by Space Objects (hereinafter Liability Convention)3 – thus become more pressing in the future. We will identify three such shortcomings concerning the who, the what and the how of liability: first, the notion of ‘launching State’, which attaches liability for damage caused by a space object to a State, is unable to take into consideration innovative methods of gaining access to space (section 2). Second, the notion of ‘damage’ does not include increases in risk due to the pollution of the outer space environment (section 3). Third, the requirement of ‘fault’ in the establishment of liability for damage caused by space objects to other space objects is exceedingly difficult to prove (section 4). Concluding, we will present an updated framework that would allow mitigating the effects of these three shortcomings (section 5).
In sections 2 to 4, we will start with a doctrinal analysis of the state of the law: the shortcomings are identified, and possible interpretations mitigating them are highlighted. Then, we draw a comparison with international liability for activities on the high seas, another internationalised area (a notion further defined in section 1.3.). The purpose of these comparisons is to find propositions for outer space law on how to accommodate the increased activity in terms of liability, formulated in section 5. The analysis is limited to international liability, the obligation to repair damage
1 In 1978, the Soviet satellite Cosmos 954 disintegrated over the Canadian Northern Territories,
scattering its component parts – including a nuclear power source – over a surface the size of Austria. Canada claimed 12 million USD for the damage caused. See E. Galloway, ‘Nuclear powered satellites: The USSR Cosmos 954 and the Canadian claim’ (1979) 12(3) Akron Law Review, 401, 407.
2 The International Space Station (ISS) conducted four manoeuvres to avoid collision with space
objects in 2015, compared to one per year before 2012. See National Aeronautics and Space Administration (NASA), ‘Two More Collision Avoidance Maneuvers for the International Space Station’ (2015) 19(1) Orbital Debris Quarterly News, pp 1-2; NASA, ‘Increase in ISS Debris Avoidance Maneuvers’ (2012) 16(2) Orbitral Debris Quarterly News, pp 1-2.
3 Convention on International Liability for Damage Caused by Space Objects (Liability Convention,
caused. Responsibility – answerability of the author State for its internationally wrongful acts –4 is only examined when it is pertinent for the analysis of international liability. The key difference between the two is that liability is independent of a violation of a norm of international law – the determining factor is that there is damage.
In short, this thesis evaluates, concerning liability for damage caused by space objects, what the regime’s shortcomings are in relation to non-traditional launch technologies and the increasingly easy access to outer space they provide, and how the liability regime of the high seas deals with similar issues. Preliminarily, we will introduce the science required to grasp the risks increased activity in outer space entails (section 1.1.), examine innovations in the launching market that result in increased space activity (section 1.2.), and line out the methodology used in assessing the legal issues related to this increased activity (section 1.3.).
1.1. Scientific background to the dangers of damage caused by space objects
In this section, we will examine the physics to which space objects are subjected. This scientific background will then be used to identify the dangers space objects pose to other space objects, to the environment of outer space, and to the surface of the Earth. Space objects can attain stable orbits if their velocity balances the gravitational pull of planet Earth.5 The stability of these orbits is limited due to atmospheric drag, especially in lower orbits. While the air density in regions of up to 2000 km above sea level is low, air is still present and thus pulls space objects towards Earth. Atmospheric drag is dependant on multiple factors, including solar activity, and cannot be predicted reliably. Space objects in Low-Earth Orbits (LEO) with an intended long lifespan – such as the International Space Station (ISS) or the Hubble Telescope – need a boost from time to time to stay in a stable orbit.6
4 B. Cheng, Studies in International Space Law (Clarendon Press, Oxford 1997), p 604-605.
5 J. Clay Moltz, Crowded Orbits: Conflict and Cooperation in Space (Columbia University Press, New
York 2014), p 22.
6 Generally, drag is associated with the loss of velocity. However, in outer space this is not so; any
velocity lost due to atmospheric drag is more than compensated by the increased gravitational pull. Thus, drag pulls space objects towards Earth rather than slowing them down. See Space Weather Prediction Center, ‘Satellite Drag’ <http://www.swpc.noaa.gov/impacts/satellite-drag> (accessed 31 May 2016); R. Fitzpatrick, ‘Effect of atmospheric drag on artificial satellite orbits’ <https://farside.ph.utexas.edu/teaching/celestial/Celestial/node94.html> (accessed 31 May 2016).
manoeuvrable space objects placed in LEO, on the other hand,will inevitably re-enter the atmosphere at one point.
Orbital decay – space objects’ gradual rapprochement to planet Earth – can take a long time. At an altitude of 1000 km above sea level, atmospheric drag is minimal. For an object placed there to descend to an altitude of 900 km, it takes 1200 years. With the increasing air density closer to Earth, atmospheric drag becomes more powerful, thus accelerating the process: to get from 400 km to 300 km, it merely takes a year and a half.7 Thus, while orbital decay is assured, the timeframes involved, especially for objects in higher orbits, can be considerable.
A problematic aspect related to this thesis’ focus is the speed at which space objects need to travel in order to counter Earth’s gravitational pull. In LEO, velocities of up to 28’000 km/h are required for space objects to maintain orbit.8 Due to these extreme speeds, space objects carry immense amounts of kinetic energy and relating thereto a great potential for causing damage.
Most immediately, space objects pose a danger to other space objects. With impact velocities of up to 50’000 km/h, a collision between two such objects may result in anything between incapacitation and complete destruction of the objects, depending on the force of their impact.9 The remains of the objects involved in the collision stay in orbit however long it takes for them to decay.10 The pieces of formerly functioning space objects then become part of the ever-growing population of space debris, an issue of concern for all space-faring nations. Pieces of space debris – any non-functional man-made object in Earth orbit –11 are uncontrollable, difficult to track when the pieces are small, and highly dangerous because of their speed.
Space debris, if it remains unmanaged, will result in collisional cascading, often referred to as the ‘Kessler-syndrome’. Non-manoeuvrable space objects collide with
7 L. Perek, ‘Space Debris’ in K. H. Böckstiegel [ed.], Environmental Aspects of Activities in Outer
Space (Carl Heymanns Verlag, Cologne 1990) p 12.
8 Space Academy, ‘Orbital Parameters’ <http://www.spaceacademy.net.au/watch/track/leopars.htm>
(accessed 31 May 2016).
9 L. Viikari, ‘Environmental aspects of space activities’ in F. von der Dunk (ed.) Handbook of Space
Law (Edward Elgar Publishing, Northampton, MA 2015), pp 721-722.
10 The 2009 collision between the two satellites Cosmos-2251 and Iridium-33 created 2000 pieces of
trackable (i.e. larger than 10 cm) debris, which is bound to stay in orbit for decades before it decays. See Secure World Foundation, ‘2009 Iridium-Cosmos Collision Fact Sheet’, <http://swfound.org/media/6575/swf_iridium_cosmos_collision_fact_sheet_updated_2012.pdf> (accessed 2 May 2016).
11 Inter-Agency Space Debris Coordination Committee, ‘Space Debris Mitigation Guidelines’
each other in LEO, creating clouds of orbital debris. These pieces of debris, in turn, collide with other space objects, resulting in a cascade of debris-producing events.12 Countering that debris production is orbital decay – pieces burning up when they re-enter the atmosphere. If the rate of production of debris due to collisional cascading is higher than the rate of reduction of debris due to gravitational pull and atmospheric drag, the debris population is expected to grow exponentially.13 This is an on-going process, and it has already started.14 With collisions becoming more prevalent, space-based aspects of our daily lives become increasingly risky to the point of being economically non-viable. We stand to lose weather forecasting,15 certain forms of telecommunication,16 and other LEO-based services. The Inter-Agency Space Debris Coordination Committee (IADC) – a forum including all major space agencies of the world –17 submitted a report to the United Nations Committee on the Peaceful Uses of Outer Space (UN COPUOS) stating that even at the current population of space objects, LEO will soon destabilise due to the Kessler-syndrome; a ban on the introduction of more space objects would not be enough to avert collisional cascading. If the Kessler-syndrome is to be averted, we need to engage in Active Debris Removal (ADR).18 This implies the controlled deorbiting of large pieces of space debris or their moving to less populated graveyard orbits.19
Deorbiting space objects may also pose a danger to Earth. Normally, space debris burns up once it reaches 80 km above sea level due to the friction of the increasing air density. However, big objects made from materials with high melting points can remain intact during re-entry and crash on the surface of the Earth.20 The idea of a space object hurling down unto Earth hitting unsuspecting citizens literally out of thin
12 B. Weeden, ‘Overview of the legal and policy challenges of orbital debris removal’ (2011) 27 Space
Policy, 38, 38.
13 D. J. Kessler, B. G. Cour-Palais, ‘Collision Frequency of Artificial Satellites: The Creation of a
Debris Belt’ (1978) 83(A6) Journal of Geophysical Research, 2637, 2643.
14 Space Safety Magazine, ‘Don Kessler on ENVISAT and the Kessler Syndrome’
<http://www.spacesafetymagazine.com/space-debris/kessler-syndrome/don-kessler-envisat-kessler-syndrome/> (accessed 13 July 2016).
15 NASA Earth Observatory, ‘Cyclone Global Navigation Satellite System (CYGNSS)’
<http://earthobservatory.nasa.gov/blogs/fromthefield/category/cygnss/> (accessed 1 June 2016).
16 Iridium, ‘The Global Network: Satellite Constellation’
<https://iridium.com/Download?attachmentID=1197> (accessed 1 June 2016).
17 See member list on < http://www.iadc-online.org/> (accessed 1 June 2016).
18 UN Office for Outer Space Affairs (OOSA), ‘Stability of the Future LEO Environment: Report of an
IADC Study’ <http://www.unoosa.org/pdf/pres/stsc2013/tech-12E.pdf> (accessed 1 June 2016).
19 B. Weeden (n 12), 38-39.
20 NASA Orbital Debris Program Office, ‘Orbital Debris Reentry’
air is one of the reasons space activities are labelled as ultra-hazardous. However, while the potential for damage when a space object crashes into an inhabited area is great, the probability of such an event taking place is exceedingly small.21 This thesis is concerned with the sustainability of the outer space environment. Liability for damage caused on Earth is mentioned to contrast liability for damage caused in outer space, but is not examined for its own sake.
1.2. Innovation in the launching market
After a dip following the 2003 Columbia disaster, the number of launches per year has been steadily climbing. In 2014 the highest number of launches in two decades was registered.22 Concurrently to the increased launch activity, a disruptive force has entered the market: the silicon valley-based start-up SpaceX reduced the price of getting into orbit from formerly 10’000 USD per kilogram23 to 1600 USD per kilogram.24 The established market leaders – Arianespace and United Launch Alliance – are trying to keep up with the price pressure in order to remain competitive.25 The quest for cost reduction results in two phenomena: first, with access to space becoming cheaper, the satellite market opens up for new ventures, resulting in increased activity in outer space.26 Second, means of getting into space alternative to the relatively inefficient vertical rocket lift-off will become more attractive.27 Such cost-reducing methods include the already existing (although still vertical) launching from a semi-submergible platform placed in the ocean at the
21 B. A. Hurwitz, State Liability for Outer Space Activities in Accordance with the 1972 Convention on
International Liability for Damage Caused by Space Objects (Kluwer Academic Publishers, Dordrecht
1992), p 29; European Space Agency (ESA), ‘Space Debris’
<http://www.esa.int/Our_Activities/Operations/Space_Debris/Re-entry_and_collision_avoidance> (accessed 1 June 2016).
22 Space Launch Report, ‘Worldwide Orbital Launch Summary by Year’
<http://www.spacelaunchreport.com/logyear.html> (accessed 1 June 2016).
23 J. Coopersmith, ‘Affordable Access to Space’ (2012) XXIX(1) Issues in Science and Technology,
available at <http://issues.org/29-1/jonathan/> (accessed 1 June 2016).
24 Werner’s Blog, ‘The Space Cost Race’ <http://wernerantweiler.ca/blog.php?item=2014-09-21>
(accessed 1 June 2016).
25 Phys.org, ‘European satellite chief says industry faces challenges’
<http://phys.org/news/2014-06-european-satellite-chief-industry.html> (accessed 1 June 2016); Denver Business Journal, ‘ULA plans
new rocket, restructuring to cut launch costs in half’
<http://www.bizjournals.com/denver/blog/boosters_bits/2014/10/exclusive-ula-plans-a-new-rocket-restructuring-to.html> (accessed 1 June 2016).
26 Silicon Valley Business Journal, ‘DFJ’s Steve Jurvetson on why he invested in SpaceX, Planet Labs’
<http://www.bizjournals.com/sanjose/blog/techflash/2015/06/dfjssteve-jurvetson-on-why-he-invested-in-spacex.html> (accessed 1 June 2016).
27 90% of a vertical lift-off rocket’s weight is in the fuel it carries to get the payload into orbit, meaning
that most of the fuel is spent on getting the remaining fuel off the ground rather than the payload. See J. Coopersmith (n 23).
equator, using the Earth’s rotation as additional boost,28 air-based launching from 12 km above sea-level with the airplane carrying the spacecraft acting as a reusable first stage,29 and the space elevator, currently in the research phase.30
The technological changes that are taking place and are about to take place in the launching market open the door for ventures previously deterred by the high costs of getting to orbit. In contrast to this enthusiasm for facilitated access to space are the developments examined in the previous section: increased activity also means increased risk of collision, to the point of engendering collisional cascading. Ideally, preventive measures should be taken so as to diminish the probability of collisions occurring. We will highlight measures that are capable of doing so. The focus, however, lies on liability for damage caused by space objects – the legal consequences when preventive measures fail to ward off collisions.
1.3. Methodology
This section introduces the liability regime of outer space law in order to line out how it will be examined. It will be presented, firstly, what methods for examining the norms that establish liability are used (1.3.1.), and secondly, what framework for comparing the liability regime of outer space to the corresponding regime of the high seas will be adopted (1.3.2.).
1.3.1. Examination of liability for damage caused by space objects
In the sections 2 to 4 the scope of the definitions and of the articles establishing liability set forth in the Liability Convention is examined, to evaluate whether the challenges identified in the sections 1.2. and 1.3. are adequately addressed.
Article I Liability Convention defines ‘damage’ inter alia as ‘loss of or damage to property of States or of persons, natural or juridical,’ the term ‘launching’ as including attempted launching, the term ‘launching State’ as either a ‘State which launches or procures the launching of a space object’ or as a ‘State from whose territory or facility a space object is launched’, and finally, the term ‘space object’ as including ‘component parts of a space object as well as its launch vehicle and parts
28 BBC, ‘Sea Launch rocket company returns to service’
<http://www.bbc.com/news/science-environment-15034079> (accessed 1 June 2016).
29 The Space Review, ‘Air launch, big and small’, <http://www.thespacereview.com/article/2543/1>
(accessed 1 June 2016).
30 P. Swan, Space Elevators: An Assessment of the Technological Feasibility and the Way Forward
thereof.’ Section 2 examines whether the scope of ‘launching State’ encompasses the three innovative means of getting into outer space listed in section 1.2.: sea-based launches, air-based launches, and space elevators. Section 3 examines in how far increases in risk due to space debris are covered under the definition of damage set forth in Article I.
Article III Liability Convention establishes liability for damage to a space object of one launching State caused by the space object another launching State. The latter is liable only if the ‘damage is due to its fault or the fault of persons for whom it is responsible.’ Article II Liability Convention, on the other hand, provides for the launching State's absolute liability for ‘damage caused by space objects on the surface of the Earth or to aircraft in flight.’ Hurwitz argues that the differentiation between the nature of liability for damage caused on Earth and in space – between absolute and fault liability – is due to the difference in the assumption of risk: un unsuspecting citizen of the Earth did not agree to the risk of a deorbiting space object demolishing his or her house, thus justifying absolute liability.31 A space operator, on the other hand, is conscious of the risks his or her space object incurs, thus justifying the higher standard of fault liability.32 Section 4 examines the notion of ‘fault’ required to establish liability under Article III Liability Convention.
The examination of these articles follows the rules of interpretation established in the 1969 Vienna Convention on the Law of Treaties (VCLT), the ordinary meaning of the terms, their context, and their object and purpose.33
1.3.2. Framework for the comparison between outer space and the high seas
The norms under examination are compared to corresponding norms pertaining to the high seas, another internationalised area. First, this comparison is justified by the inherent economic, historic, and legal similarities between the two regimes. Then, the method for comparing the regimes is introduced, and finally, the parameters against which the comparison is undertaken are established.
Economically speaking, outer space and the high seas constitute common goods, as referred to in Hardin’s classic article published in 1968. The tragedy of such common goods is that individuals (or States) have an incentive to overuse them, since they reap
31 B. A. Hurwitz (n 21), p 31. 32 ibid. p 33.
33 Vienna Convention on the Law of Treaties (VCLT, adopted 23 May 1969), UNTS 1155, 331, art
the full gains, whereas the risks of such use are shared.34 Thus, a lack of compliance with safety standards lowers costs of operation in shipping on the high seas, while the results of an accident are potentially devastating to the environment of nearby States – as illustrated by the ExxonValdez oil spill.35 Similarly, as already mentioned in section 1.1., placing more and more man-made objects in Earth orbits may be financially viable for the individual operator, but in aggregate, this trend leads to overcrowded orbits and collisional cascading. A space environment riddled with an exponentially growing debris population loses its appeal when the risks one’s space assets are exposed to become too large to bear. The two areas are thus subject to the tragedy of the commons.
Historically speaking, the economic similarities examined above have shaped the drafting of outer space law. When the need for a new area of law arises, lawmakers rarely invent rules from scratch; rather, they borrow from similar regimes, adopting what has worked well therein, and adapting it to the specificities of the new regime. In the inception of international space law, the law of the seas has played a pivotal role:36 the high seas seemed to be an appropriate analogy due to the vastness of both space and the high seas.37
Legally speaking, the borrowing from the law of the sea resulted in similarities between the regimes: in the 1982 United Nations Convention on the Law of the Sea (LOS Convention) and the 1967 Outer Space Treaty (the ‘Magna Charta’38 of outer space law on which the Liability Convention expands39), the contracting parties have agreed to keep these areas outside the reach of claims of sovereignty.40
For these reasons, the high seas are an appropriate source from which to draw inspiration for propositions of an improved framework for outer space.
34 G. Hardin, ‘The Tragedy of the Commons’ (1968) Vol. 162 No. 3859 Science 1243, 1244.
35 National Transportation Safety Board, ‘Safety Recommendation’
<https://web.archive.org/web/20100611194527/http://www.ntsb.gov/Recs/letters/1990/M90_26_31A.p df> (accessed 3 May 2016).
36 M. Lachs, The Law of Outer Space: An Experience in Contemporary Law-Making (Sijthoff, Leiden
1972), pp 20-23.
37 ‘Space is to celestial bodies as seas are to islands’. See M. J. Peterson, International Regimes for the
Final Frontier (SUNY, Albany 2005), p 49.
38 H. L. van Traa-Engelman, Commercial utilization of outer space: law and practice (Nijhoff,
Dordrecht 1993), p 279.
39 Liability Convention (n 3), preamble para 2.
40 For the high seas, see United Nations Convention on the Law of the Sea (LOS Convention, adopted
10 December 1982) 1833 UNTS 3, art 89. For outer space, see Treaty on Principles Governing the Activities of States in the Exploration and Use of Outer Space, including the Moon and Other Celestial Bodies (Outer Space Treaty, adopted 27 January 1967), 610 UNTS 205, art II.
The three shortcomings mentioned in the beginning of section 1 are compared to the corresponding provisions regarding the high seas according to Kamba’s three-phase method. Kamba asserts that a comparison between legal regimes entails a descriptive phase, an identification phase, and an explanatory phase. The first phase introduces the concepts which are to be compared. The second phase identifies differences and similarities, and the third phase tries to account for them.41
The objective of the comparison is to find inspiration as to how to deal with the dual trend of increased activity through cheaper access to outer space, and of increased risk for damage caused by space objects due to the increased activity. As space debris is a form of pollution, parameters against which the regimes are compared relate to the pollution of the respective environment. More specific parameters depend on the content of the shortcoming that is examined, and are identified at the beginning of the section.
2. The who: coverage of innovative means to gain access to
outer space
Price pressure forces launch operators to consider alternative launching methods to the relatively inefficient vertical lift-off rocket. It is not clear whether these new means are covered under the Liability Convention’s definition of ‘launching State’, which attaches liability to a State. This section discusses whether different cost-reduction means are covered under the liability regime (2.1.). Then, the findings are compared to how liability attaches in the high seas (2.2.). The parameters through which the outer space regime is compared to the regime of the high seas and the seabed are the comprehensiveness of the criterion of attachment and to which type of actors liability attaches.
2.1. Coverage under outer space law
Article I(c) Liability Convention has two subsections to define the ‘launching State’. Cheng notes that these two subsections contain four alternative criteria in total: the State that launches the object, the State that procures the launch of the object, the State from whose territory the object is launched, and the State from whose facility
41 W. J. Kamba, ‘Comparative Law: A Theoretical Framework’ (1974) 23 International Comparative
the object is launched.42 If multiple States fulfil different aspects of the definition of ‘launching State’ for a single launch, they are all considered ‘launching States’ and are jointly and severally liable in case an event resulting in liability occurs.43 This broad system is an expression of the victim-orientedness of the Liability Convention;44 its purpose is to assure that if damage occurs, there is a State that is liable. 45 Developments in the launching market are putting this intended comprehensiveness to the test. So long as a State is involved in the launching, there is a ‘launching State’.46 However, when private persons act as launch operators, the launch needs to take place on the territory or the facility of a State for liability to attach.47 In the UN General Assembly resolution on the concept of the ‘launching State’, some countries have expressed the view that the already commonplace sea-based and air-sea-based launches might not be covered. The absence of coverage would also extend to means of getting into orbit that are still in the research phase, such as the space elevator. In the following, it is examined whether the criteria can be interpreted to cover these three means of getting to space, first, according to the ordinary meaning of the provisions (2.1.1.),48 and then according to their context, and their object and purpose (2.1.2.).49
2.1.1. Ordinary meaning of ‘launching State’
The semi-submergible platform Odyssey operated by the company Sea Launch allows for launches to take place at the equator on the high seas.50 The positioning at the equator allows rockets to benefit from an additional boost given by the Earth’s rotation, thus increasing payload capacity.51 The launch takes place on the high seas, i.e. on no State’s territory. If it is a private actor that launches or procures the launch, the only criterion left for there to be a ‘launching State’ is that of the ‘State facility’. Van Fenema notes that a sea-based launch platform could be construed as a ‘State
42 B. Cheng (n 4), p 613.
43 Liability Convention (n 3), art IV(1). 44 ibid. preamble para 4.
45 M. A. Kerrest, ‘Actualités du droit de l’espace : le responsabilité des États du fait de la destruction de
satellites dans l'espace’ (2009) 55 Annuaire français de droit international 615, 617.
46 Liability Convention (n 3), art I(c)(i). 47 ibid. art I(c)(ii).
48 VCLT (n 33), art 31(1). 49 ibid. art 31(1).
50 Sea Launch, ‘Launch Platform Odyssey’ <http://www.sea-launch.com/launch/11142> (accessed 10
May 2016).
facility’ of the State whose flag it flies – Liberia, in the case of Odyssey – but at the same time concedes that this approach is subject to debate.52 The flag State may exercise jurisdiction over the vessel,53 but to conclude that as a consequence of registration, a ship becomes a ‘State facility’ seems to be a stretch considering the ordinary meaning of the criterion. The exact wording of that criterion is ‘a State from whose […] facility a space object is launched’. The use of the possessive ‘whose’ implies that the facility needs to be owned or at least operated by a State. For Odyssey, however, this is not the case, as it is privately owned and operated. According to the ordinary meaning of the definitions of ‘launching State’, liability thus does not attach to privately operated sea-based launches.
In air-based launches, a conventional airplane carries the spacecraft to a high altitude from where the latter launches. This allows saving the fuel needed to get out of the densest parts of the atmosphere, with the airplane acting as a reusable first rocket stage.54 The private company Orbital ATK carries its spacecraft Pegasus to international waters, to drop it at an altitude of 12’000 meters above sea-level, from where it ignites its engines to reach low Earth orbits.55 Here, again, it is a private actor that launches. If furthermore the launch is procured by another private actor, a State’s territory or the facility of a State are the two criteria of attachment that remain. Concerning the question whether the airplane upon which the spacecraft ‘piggybacks’ to higher altitudes constitutes a ‘State facility’, the discussion on ‘State facility’ in relation to sea-based launches applies mutatis mutandis.
The criterion of the ‘State from whose territory […] the space object is launched’ on the other hand may provide for attachment. While the spacecraft’s engines are only ignited in high altitudes – above the high seas, in Pegasus’ case –, it is not a foregone conclusion that only the ignition of the engines is determinant for a spacecraft to be considered as ‘launched’. Hurwitz, drawing analogies from air law’s definition of ‘in flight’, argues that a space object can be considered ‘launched’ even before engine ignition. The 1963 Tokyo Convention considers an airplane to be ‘in flight from the
52 P. van Fenema, ‘Legal aspects of launch services and space transportation’ in F. von der Dunk (ed.)
Handbook of Space Law (Edward Elgar Publishing, Northampton, MA 2015), p 401.
53 LOS Convention (n 40), art 92(1). 54 The Space Review (n 29)
55 Orbital ATK, ‘Pegasus’
moment when power is applied for the purpose of take-off’.56 If ‘launched’ analogically is interpreted as ‘power applied for the purpose of lift-off’, then arguably the engine ignition of the airplane carrying the spacecraft could be determinant. The airplane replacing the spacecraft’s first stage, take-off contributes to the spacecraft’s reaching of outer space. Thus, the State from whose territory the airplane takes off would be considered a ‘launching State’, even if the spacecraft’s engines were to be ignited only above the high seas. This interpretation might provide for a ‘launching State’ when otherwise there would be none. However, there is no consensus on this interpretation and by extension neither on whether air-based launches have a ‘launching State’.57
Space elevators employ climbers that are sent up a tether to reach Geosynchronous Earth Orbit 36’000 km above sea level. A study of the International Academy of Astronautics (IAA) predicts that within the 2020’s we will have the technology to build such an elevator, allowing for routine, safe, and environmentally friendly access to outer space. Furthermore, the cost of access would be reduced to a estimated 500 USD per kilogram, effectively creating a highway between Earth and outer space.58 Such inexpensive and safe access would allow for a fundamental transformation of how outer space is used.59
According to the IIA study, the base node of a space elevator would most likely be placed on the high seas.60 The considerations concerning the ‘launching State’ of sea-based launches thus apply to the space elevator as well. While sea-sea-based launches are relatively rare,61 a space elevator with its transformative impact would bring the absence of coverage of privately owned and operated, sea-based means of access to space to the forefront.
As we have demonstrated, the ordinary meaning of ‘launching State’ might contain lacunae in the light of innovative means of accessing outer space. The source of the
56 Convention on Offences and Certain Other Acts Committed on Board Aircraft (adopted 14
September 1969) 704 UNTS 220, art 1(3)
57 UN General Assembly (UNGA) Res A/AC.105/768 Review of the concept of “launching State” (21
January 2002), paras 43-46.
58 P. A. Swan (n 30), p 9.
59 For instance, space-based solar plants could produce energy a lot more efficiently than their
ground-based counterparts. Also, interplanetary travel would become more feasible since spacecraft could be assembled in orbit rather than on planet Earth. See P. A. Swan (n 30), p 10-12.
60 ibid. p 260.
61 Sea Launch has conducted 35 launches so far. Sea Launch, ‘Past Launches’,
lacunae examined is above all the advent of private launch operators. Outer space law is State-centred; private space activities were but a minor concern to the drafters.62 This absence of foresight might become a problem in light of the innovative means of accessing outer space discussed here, as liability might not attach to a State even when all other conditions are fulfilled.
2.1.2. Context, and object and purpose of the definition of ‘launching State’
Together with the ordinary meaning of terms, their context and their object and purpose help with the interpretation of treaty provisions.63 The purpose of the Liability Convention is found in its preamble. Due to the ultra-hazardous character of space activities, victims suffering damage as a result thereof should be compensated promptly, fully, and equitably.64 This is often referred to as the victim-orientedness of the Liability Convention. Smith and Kerrest argue that broad interpretations of the terms of the convention can be followed if that allows for the compensation of victims.65 However, a broad interpretation is not the same as an interpretation contra legem. Broad interpretations stay within the margin of appreciation that is left by the indetermination of the norm, while interpretations contra legem operate outside of the scope of the norm.66 The drafters inserted the definition of the ‘launching State’ as it exists now for a reason. Other, more comprehensive proposals were rejected.67 Another aspect of outer space law, responsibility, has adopted the broad term ‘national activities […] carried on by governmental agencies or by non-governmental entities’ as the criterion of attribution.68 As Gerhard notes, most authors agree that ‘national activities’ cover all activities over which the State has jurisdiction.69 In 2015, States expressed the view within the Legal Subcommittee of UN COPUOS that they considered ‘national activities’ and the criterion of ‘launching State’ to be
62 F. von der Dunk, ‘International space law’ in F. von der Dunk (ed.) Handbook of Space Law
(Edward Elgar Publishing, Northampton, MA 2015), pp 30-32.
63 VCLT (n 33), art 31(1).
64 Liability Convention (n 3), preamble paras 3 and 4.
65 L. Smith, M. A. Kerrest, ‘Article I (Definitions) LIAB’ in Hobe [et al.] (ed.), Cologne Commentary
on Space Law: Volume 2 (Wolters Kluwer, Cologne 2009), p 111 para 47.
66 J. P. Cot, ‘Margin of Appreciation’ MPEPIL 1438 (Oxford University Press, Max Planck
Encyclopaedia of Public International Law), para 4.
67 The United States of America proposed States would be liable for any of their national’s space
activities. See L. Smith, A. Kerrest, ‘Article I (Definitions) LIAB’ (n 65), p 108 para 38.
68 Contracting parties are responsible for ‘national activities’, including governmental and
non-governmental activities. See Outer Space Treaty (n 40), art VI.
69 M. Gerhard, ‘Article VI’ in Hobe [et al.] (ed.), Cologne Commentary on Space Law: Volume 1,
different; that responsibility for ‘national activities’ did not necessarily imply liability as a ‘launching State’.70 This suggests that States Parties consider the criterion of ‘launching State’ to be more restricted than that of ‘national activities’.
This leads to the conclusion that the drafters, having had more comprehensive options for attachment of liability, but despite the victim-orientedness of the Liability Convention having chosen the definition at hand, wanted to have a proximity between the activity that gives rise to liability and the State to which said liability attaches. This proximity needs to be closer than mere jurisdiction over the entity that launches or procures the launch of the space object, otherwise a looser definition would have been adopted.
Interpreting the State from which an airplane carrying a spacecraft takes off as a ‘launching State’ by virtue of having offered its territory for the reusable first stage to take off from seems legitimate; the purpose of take-off, if not obvious by taking a single look at it, also needs to be documented in the journey log book, accessible to the ‘launching State’.71 The airplane’s take off is an integral part of the launch. There is proximity between the role of the State – consciously lending its territory for take off – and the launching of the spacecraft. It can thus be argued that air-based launches fall within the scope of the definition of ‘launching State’ as intended by the drafters. However, sea-based launches and a future sea-based space elevator take place from structures placed in the high seas. In the case of Sea Launch, the rocket is assembled on the control ship Sea Launch Commander on its way to the launch location on the high seas. The rocket is then transferred to the platform Odyssey from where it is launched.72 Even less proximate, a sea-based space elevator might receive cargo destined for space from ships from all over the world.
Here, the proximity is far more tenuous than with air-based launches. The jurisdiction of the flag State of the vessels from which the launch takes place comprises above all the control of seaworthiness of the vessel, the prevention of damage to the marine environment, and the assurance of compliance with labour standards.73 What
70 UNGA Res A/AC.105/1090, Report of the Legal Subcommittee on its fifty-fourth session (30 April
2015), para 68.
71 Convention on International Civil Aviation (adopted 7 December 1944) 15 UNTS 295, arts 29(d)
cum 34.
72 Sea Launch, ‘Sea Launch Commander’ <http://www.sea-launch.com/launch/11138> (accessed 6
June 2016); Yuzhnoe, ‘Sea Launch: The thirty-sixth launch of Zenith-3SL Rocket, <http://www.yuzhnoye.com/en/press-center/pressrelises/sea-launch-36.html> (accessed 6 June 2016).
activities take place on board of the vessel are irrelevant to the flag State so long as these maritime law regulations are complied with. It is a far stretch to conclude that this regulatory oversight relating to maritime law leads to sufficient proximity for the launch facility to be qualified as a ‘State facility’ in the sense of outer space law. Absent the required level of proximity, arguably neither sea-based launches nor objects carried to outer space by sea-based space elevators would have a ‘launching State’ as defined by the drafters.
In light of the context and the object and purpose of the definition of ‘launching State’, it is thus possible to construe air-based launches as launches from the territory of a State, the latter then qualifying as ‘launching State’. However, sea-based launches and sea-based space elevators seem not to fall within the scope of ‘launching State’ as defined by the drafters.
2.2. Coverage in outer space compared to coverage on the high seas
We will now examine how and to whom liability attaches on the high seas. After describing the liability regime, we will identify differences with the liability regime of outer space law, and then examine how these differences could inform an updated liability framework for outer space.74
This section examines the LOS Convention as the law generally applicable to the high seas, and the Convention on Civil Liability for Oil Pollution Damage (CLC),75 for oil pollution in particular.
On the high seas, States have a due diligence obligation76 to ensure that vessels under their jurisdiction – i.e. vessels flying their flag –77 do not cause damage by pollution to other States and their environment.78 Failure to observe said obligation results in the liability of the flag State.79 Liability attaches through the ‘flag State’. The criterion for a ship to fly the flag of a State is that there exists a ‘genuine link’ between the two.80 Ideally, this ‘genuine link’ requires the flag State to exercise regulatory
74 W. J. Kamba (n 41), p 511-512.
75 Protocol of 1992 to amend the International Convention on Civil Liability for Oil Pollution Damage
of 29 November 1969 (CLC, adopted 27 November 1992) 1956 UNTS 255.
76 T. Koivurova, ‘Due Diligence’ MPEPIL 1034 (Oxford University Press, Max Planck Encyclopaedia
of Public International Law), para 29.
77 LOS Convention (n 40), art 92(1). 78 ibid. arts 194(2) and (3)(b). 79 ibid. art 194(2).
oversight in administrative, technical and social matters,81 however, in practice the International Tribunal for the Law of the Sea (ITLOS) seems to suggest that a ‘genuine link’ is established as soon as the vessel is registered in the national registry of the flag State.82 Thus, under the LOS Convention, if liability can be established, it attaches to the State which registered the ship.
Pertaining to oil pollution specifically, liability may also attach to the owner of the ship at the origin of oil pollution within the Exclusive Economic Zone or the territorial waters of the injured State.83 Because private persons, natural or juridical, do not have the same deep pockets as States, there is furthermore a requirement for the conclusion of third-party liability insurance where the potential for damage is great.84 If the damage exceeds the insured amount, or the owner is exonerated, the International Fund for Compensation for Oil Pollution Damage (IOPC) covers the difference.85 The fund is composed of contributions from all major oil importers.86 This ensures that the victims are compensated while expanding the onus to do so from merely the oil shipping industry to the people that provide the incentive for the existence of said shipping industry.
Regarding pollution on the high seas causing damage to the marine environment of States, liability may attach to the ‘flag State’ in general, and to the owner of the ship in particular in the case of oil pollution.
The differences between the liability regime of outer space and the high seas regarding attachment of liability are two: first, the LOS Convention uses a criterion requiring a formal act – registration – for liability to attach, whereas in the Liability Convention, liability attaches according to factual criteria independent of any formal act on the part of the ‘launching State’. Second, the Liability Convention only holds States liable. In the law of the sea, in the case of oil pollution damage, the ship owner can be liable directly.
81 ibid. art 94.
82 ITLOS, 20 April 2001, Grand Prince (Belize v France), Prompt Release, Judgment, ITLOS Reports
2001, p 17, paras 82-84; D. König, ‘Flag of Ships’ MPEPIL 1166 (Oxford University Press, Max Planck Encyclopaedia of Public International Law), para 15.
83 CLC (n 75), arts II(a) and III(1).
84 Specifically, when the ship is carrying more than 2000 tons of oil. ibid. art VII (1).
85 ibid. art XIIbis(b), Protocol to amend the 1971 International Convention on the Establishment of an
International Fund for Compensation for Oil Pollution Damage (Fund Protocol, adopted 27 November 1992), 1953 UNTS 330, art 4(1).
The first difference does not provide any inspiration for how to make attachment of liability in outer space more comprehensive. Space objects also require registration, and the ‘State of registry’ is always also a ‘launching State’.87 In cases where the existence of a ‘launching State’ is dubious, but a State has registered the object, that State can be presumed to consider itself the ‘launching State’. The national registry of space objects thus already provides for a strong indication that liability may attach to the State of said registry.88
The second difference denotes a move away from State-centredness in the liability regime of the high seas. Seeing that the increasingly cheap access to space described in section 1.2. will stimulate more and more private activity in outer space, such a development could be envisaged in outer space as well. The establishment of a space debris fund backed by persons who control the space object – the launching provider for launching stages, the satellite operator for deployed payloads – would allow to incorporate important space actors other than States into the liability regime as well. The modalities of their backing is discussed further in section 4 on the nature of liability.
2.3. Interim conclusions
We have seen that outer space law attaches liability only to States. Furthermore, two unforeseen developments – private launching operators and innovative launching techniques – have resulted in certain means of access to space presumably not being covered (sea-based launches and space elevators), and others, even though potentially covered, to be surrounded by confusion as to whether they effectively are (air-based launches). The liability regime of the high seas, on the other hand, attaches liability to States and private actors alike. The high seas provide for a fail-safe mechanism when the damage is not or cannot be covered by the non-governmental actor: the International Fund for Oil Pollution Damage. These two elements, the extension of liability to non-governmental actors and the establishment of a fund, will be incorporated into our proposal for an updated liability framework for outer space law.
87 Convention on the Registration of Objects Launched into Outer Space (opened for signature 14
January 1975), 1023 UNTS 15, art I(c).
88 F. von der Dunk, ‘Towards ‘Flags of Convenience’ in Outer Space ?’ (2012) Space and
3. The what: definition of damage
The scope of the definition of damage in outer space is examined in this section, identifying two issues (3.1.): First, in the liability regime of outer space, damage to the environment and the related increase in risk of operating within it are not covered. Second, not only is damage to the environment not covered, the liability regime also poses an obstacle to Active Debris Removal (ADR). Then, the regime of outer space is compared to the regime of the high seas to find inspiration in how to deal with these issues (3.2.). The parameter for comparison is how events leading to increase in risks of operation are covered in the two liability regimes.
3.1. Coverage under outer space law
The definition of damage in the Liability Convention covers ‘loss of life, personal injury or other impairment of health; or loss of or damage to property of States or of persons, natural or juridical’.89 Whereas on the surface of the planet damage to the environment of a State can be covered under ‘property of a State’,90 damage to the outer space environment – not being under the sovereignty of any State – is not covered.91 This suggests that the debris produced by any activity falls outside the scope of the liability regime. In light of the increased activity that launching market innovations promise to bring to outer space, and the resulting increase in risk of collisions producing debris, this absence of coverage might prove problematic. In the following, we analyse the scope of the definition of damage under the rules of interpretation as set out in the VCLT (3.1.1.). Further, ADR operations are examined under the light of the liability regime (3.1.2.).
3.1.1 Scope and definition of damage under outer space law
Without damage, there is no liability. Damages to the environment per se are not included in the definition of damage in the Liability Convention, but indirect damages to property might be. Considering the victim-orientedness of the Liability Convention, Hurwitz proposes a broad interpretation of the term ‘damage’ that includes indirect damage.92 He further notes that there is precedent in general international law indicating that ‘damage’ includes indirect damage, inter alia the
89 Liability Convention (n 3), art I(a).
90 L. Smith, A. Kerrest, ‘Article I (Definitions) LIAB’ (n 65), p 113 para 55. 91 ibid. p 111 para 48.
Lusitania award rendered by the US-Germany Claims Commission in 1923.93 Another indication towards the conclusion that indirect damages are covered is that other international conventions specifically exclude them, thus suggesting that indirect damages are the norm from which explicit abrogation is needed.94
Applied to debris-producing events, the possibility for indirect damages means that there might not be liability for all of the debris produced, but that there is liability for the debris that later causes damage to another space object. The outer space environment is therefore only protected insofar as damages to it cause damage to space objects that are found in it. This is problematic because damage from space debris can be difficult to prove. Space debris comes in many different sizes; pieces smaller than 10 cm are not traceable, yet because of their velocity they still have the capacity to seriously endanger other space objects. Liability arises not out of the production of debris, but out of subsequent damage to property as a result of debris production. Damage can be caused by space debris without the injured party being aware what the source of the damage was. If the piece of debris cannot be traced to a launching State, no one is liable.
As mentioned in section 1.1., collisions become more prevalent and the debris population more important. Non-traceable and non-identifiable pieces of debris thus pose an ever-growing risk. All damage arising out of collisions with such pieces being exceedingly difficult to trace back to a launching State, the narrow definition of damage in the outer space liability regime hence poses an obstacle to the effectiveness of the regime.
3.1.2. Obstacles posed to ADR operations by the outer space liability regime
In von der Dunk’s reading of the Liability Convention’s definition of ‘space object’, all man-made objects in Earth orbits or beyond are covered.95 Whether or not an object in space is functional does not matter in the liability regime.96 Thus, all forms
93 Umpire Parker stated in the Lusitania award that proximate cause (i.e. damage with a clear, unbroken
connection between the act and the damage) was the rule, from which the parties had no intention of abrogating. US-Germany Mixed Claims Commission, 1 November 1923, Lusitania, Administrative
Decision No 1, Reports of International Arbitral Awards, 7, 29; B. A. Hurwitz (n 21), p 13.
94 Convention on Damage Caused by Foreign Aircraft to Third Parties on the Surface (adopted 7
October 1952), 310 UNTS 182, art 1(1); B. A. Hurwitz (n 21), p 16.
95 F. von der Dunk, ‘Legal aspects of private manned spaceflight’, in F. von der Dunk (ed.) Handbook
of Space Law (Edward Elgar Publishing, Northampton, MA 2015), p 679.
of space debris – be they paint specks or defunct satellites the size of a bus –97 are considered space objects, and any damage caused by them could give rise to liability. What is more, the State on whose registry the space object was launched retains jurisdiction over it in perpetuity.98 A space object cannot be removed without the consent of the State of registry.99
ADR operations have not yet been carried out, and already a series of problems relating to the aforementioned aspects of the liability regime are discernible. Most importantly, ADR operations are inherently risky. The purpose is to get the biggest non-functional space objects away from orbits where they might be hit by debris and break into thousands of pieces, ideally by causing them to burn up upon re-entry into the atmosphere.100 In order to do so, the vehicle engaging in ADR must interact with a non-functional space object that might be destabilised or tumbling. Attachment to such a space object might cause it to break up or explode, thus creating debris – the very thing the operation sought to prevent. Even if the attachment is successful and the piece of debris is moved to a lower orbit to quickly re-enter the atmosphere, there is no guarantee that the entire object will burn up. Pieces surviving re-entry might then cause damage on Earth, for which the launching State is absolutely liable.101 Hence, launching States might be reluctant to allow ADR operations to be carried out on their space objects. Seeing that their consent is needed in order to do so, the risk for them of incurring liability might prevent ADR operations to take place.102
3.2. Coverage in outer space compared to coverage on the high seas
The liability regime of the high seas attaches liability to damage caused by pollution of the marine environment.103 In the LOS Convention, pollution of the marine environment is defined as ‘the introduction by man […] of substances [resulting] or likely to result in such deleterious effects as harm to living resources and marine life, hazards to human health, hindrance to marine activities, including fishing and other legitimate uses of the sea, impairment of quality for use of sea water and reduction of
97 Space.com, ‘Huge Dead Satellite May Be Space Junk for 150 Years’,
<http://www.space.com/15640-envisat-satellite-space-junk-150years.html> (accessed 2 May 2016).
98 Outer Space Treaty (n 40), art XIII. 99 B. Weeden (n 12), p 41.
100 ibid. p 40.
101 Liability Convention (n 3), art II.
102 L. Viikari, ‘Environmental aspects …’ (n 9), p 734.
amenities’ (emphasis added).104 For the purposes of liability, the damage needs to be quantifiable in monetary terms.105 This includes interference with human activities, such as fisheries, but also – seeing the obligation to reduce and control pollution,106 and following the rule of proximate causation – the costs of the reinstatement of the status quo ante.107 For the high seas, only damage that is caused to the environment within the jurisdiction of the injured State is relevant.108 However, the source of the damage is not specified; thus, in an oil spill or another kind of vessel-based pollution occurring on the high seas, resulting damage suffered in the zones within the territorial jurisdiction of States is covered.
There are two differences between the liability regime of outer space on the one hand, and the liability regime of the high seas (when damage occurs within the jurisdiction of a State) on the other: firstly the environment per se is protected on the high seas, whereas in outer space, the liability regime not only fails to cover the environment per se or increases in risks of operation, but also poses an obstacle to the protection of the environment through the reduction of the abovementioned risks. Seeing that outer space – as opposed to the high seas – neither harbours any life forms nor is part of the ecosystem that sustains life on Earth in general, calling for the protection of the outer space environment per se might be too audacious. However, including increased risk as a hindrance to space activities in the definition of damage, like the liability regime of the high seas does, would allow establishing liability for the debris-producing event directly, and not only for any subsequent damage the debris causes. Liability for increased risk seems appropriate given that if we continue to use outer space as we do now, we will soon not be able to use it at all (section 1.1.) Thus, liability would cease to exist solely between two launching States (for the purposes of this thesis, inter partes liability), but encompass also damages to the common good as well (erga omnes liability). Secondly, from the protection of the high seas environment flows an obligation to reduce and control pollution; this activity can be assessed monetarily, and can thus be qualified as damage for which the polluter may be liable. Damage as increased risk of operation in outer space could be assessed by the cost of reduction of
104 LOS Convention (n 40), art 1(4).
105 M. H. Nordquist [et al.] (eds.), United Nations Convention on the Law of the Sea 1982: A
Commentary, Volume IV (Kluwer, The Hague 2002), p 66 para 194.10(f).
106 LOS Convention (n 40), 194(2) cum 194(2). 107 ibid. art 194(2), CLC (n 75), art I(6)(a).
the same amount of risk. A proposition concerning the metrics for assessing said increases and decreases is drafted in section 4 on the nature of liability.
3.3. Interim conclusions
Space debris polluting the environment of outer space does not in and of itself give rise to liability because the Liability Convention does not consider the environment as a potential object of damage caused. However, indirect damages to property are covered; thus, a debris-producing event may give rise to liability if the debris causes damage to another space object. Compared to the high seas, this narrow definition of damage is flagrant. There is no obligation to reduce or control pollution in outer space, whereas the same obligation exists for the high seas. Furthermore, the liability regime of outer space poses an obstacle to the reduction of pollution through ADR operations. This is problematic in light of the increase in space activities that is to be expected by the ever-cheaper access to outer space.
4. The how: nature of liability
In case of damage caused to space objects by other space objects, the launching State of the latter is liable if the damage is due to its fault or to the fault of persons under its responsibility.109 Fault liability – opposed to the absolute liability for damage caused on Earth or in air –110 is subject to heavy criticism, most notably that it precludes the establishment of liability altogether.111
4.1. Coverage under outer space law
The notion of ‘fault’ is not further defined in outer space law. Baker notes that fault could be construed both as subjective – blameworthiness as in the law of negligence – , or objective – as in non-compliance with a rule of international law.112 The objective approach is inoperable; in international law there are no rules that prescribe a certain conduct in outer space. Hence, the conduct of a State leading up to a collision cannot be tested against the conduct that is prescribed by law. Even for the subjective approach, a comparison with an ideal conduct is necessary. Hence, in order to
109 Liability Convention (n 3), art III. 110 ibid. art II.
111 H. A. Baker, ‘Liability for Damage Caused in Outer Space by Space Refuse’ (1988) XIII Annals of
Air and Space Law, 183, 220-221.
establish blame or negligence, some guidelines as to how to conduct space activities responsibly are required.113 To that end, Smith and Kerrest suggest that we look to codes of conduct adopted by space agencies or by international bodies, such as the IADC and its Space Debris Mitigation Guidelines (IADC Guidelines).114
However, the existence of such codes of conduct is not enough to provide for an effective liability regime, especially in light of collisions including non-navigable space debris. Assuming, as was the case in the first-ever collision between two satellites, one is functional and the other one is not.115 Is the State of the non-functional satellite liable because it left the satellite in a congested orbit, as the IADC Guidelines advise not to do?116 While that may be construed as faulty conduct, the conduct of the other party to the collision needs to be examined as well: if the other party’s space object is operational could there not have been an evasion manoeuvre?117 Was the other party at fault for not evading the non-navigable satellite?
Fault, i.e. deviance from responsible conduct, furthermore presupposes having a choice of conduct and choosing a non-responsible course of action. In order to make responsible choices, adequate data is required. Pieces of debris smaller than 10 cm are not traceable and do not appear in space object inventories. How can one choose the responsible course of action when the risk is not known?118 The ability to correctly assess risks of collision is therefore constrained by the inherent limitations of tracking technology.
Combining the uncertainty of the notion of fault with the limited reliability of the means to prove it, the establishment of fault of one party is nearly impossible. Lyall and Larson conclude that the regime of fault liability in outer space in its current form will not result in the actual establishment of liability in practice.119
113 ibid. p 220.
114 IADC (n 11); L. Smith, M. A. Kerrest, ‘Article III (Fault Liability) LIAB’ in Hobe [et al.] (ed.),
Cologne Commentary on Space Law: Volume 2 (Wolters Kluwer, Cologne 2009), p 133 para 131.
115 Secure World Foundation (n 10), p 1; F. von der Dunk, ‘Too-close encounters of the third-party
kind: will the Liability Convention stand the test of the Cosmos 2251-Iridium 33 Collision ?‘ (2009) 52 Proceedings of the International Institute of Space Law 199, 202-205.
116 IADC (n 11), art 5(3)(2).
117 F. von der Dunk, ‘Too-close encounters …’ (n 115), p 203.
118 L. Viikari, The Environmental Element in Space Law: Assessing the Present and Charting the
Future (Martinus Nijhoff, Leiden 2008), p 71.
119 F. Lyall, P. Larson, Space Law: a Treatise (Ashgate, Burlington 2009), p 108; L. Smith, M. A.
4.2. Coverage in outer space compared to coverage on the high seas
The liability regime of the high seas does not refer to fault. Rather, it contains rules of conduct against which the conduct of the entity causing damage can be tested. This mechanism of establishing liability is reminiscent of Baker’s aforementioned notion of ‘objective fault’ – damage as a result of non-compliance with a rule of international law. The regime establishes liability of the State if it violated its due diligence obligations in ensuring that damage will not be caused by operators under its jurisdiction. If damage occurs as a result of these rules’ violation, the State or private actor at the origin of the violation is liable. Furthermore, in case of oil pollution, the owner of the vessel at the source of the pollution incurs strict liability.
On the high seas, the due diligence obligation of the flag State in relation to the protection of the marine environment contains measures designed to minimize ‘pollution from vessels, in particular measures for preventing accidents and dealing with emergencies, ensuring the safety of operations at sea, preventing intentional and unintentional discharges, and regulating the design, construction, equipment, operation and manning of vessels.’120 In taking these measures, the flag State discharges its obligations;121 the due diligence character of the liability regime on the high seas results in the occurrence of liability not for failing to achieve the goal of the norm – no damage caused to other States and their environment – but for failing to take the necessary, diligent steps towards that end.122 Therefore, not every damage to a State and its environment caused by a ship flying the flag of another State necessarily results in the liability of the latter.
In the specific case of oil pollution, the owner of the ship at the origin of the pollution incurs strict liability.123 No fault is required; as soon as there is damage, the owner is liable. To mitigate the severity of strict liability, the owner can limit liability to a certain amount,124 and even be absolved in extraordinary circumstances.125 In cases
120 LOS Convention (n 40), art 194(3)(b).
121 M. H. Nordquist (IV, n 105), p 66 para 194.10(h). 122 T. Koivurova (n 76), para 3.
123 CLC (n 75), art III(1); A. Douhan, ‘Liability for Environmental Damage’ MPEPIL 1580 (Oxford
University Press, Max Planck Encyclopaedia of Public International Law), paras 20-21.
124 CLC (n 75), art V.
125 Inter alia when the damage was due to war, insurrection, or a natural phenomenon of exceptional,
