Assessing the Risks of Dual-Use SSSB Deflection

With thanks to Stephen Clare, Matthew Gentzel, Fin Moorhouse, Matt Lerner, Shaan Shaikh, and others for their thoughts and comments on previous versions of this document.

Summary

A small solar system body (SSSB)—a term used by the International Astronomical Union—is an object in the solar system that is not a planet, dwarf planet, or satellite.^[1] SSSB deflection technology is sometimes classified as a potentially weaponizable dual-use technology: if one can deflect an asteroid away from the Earth, one can also deflect it towards the Earth, and cause unprecedented destruction. This possibility was sometimes called “Ivan’s Hammer” in the Cold War, was popularized by Carl Sagan in the 1990s, and occasionally receives attention as a potential catastrophic risk. I argue here that:

1. In the near term, SSSBs are not an operationally useful class of weapons for most actors in the international system.
2. SSSBs are indiscriminate, single-use, and difficult to conceal, among other issues.
3. SSSB deflection capability does not necessarily entail SSSB targeting capability.
4. Even near-omnicidal actors have more accessible and cheaper alternatives when considering WMD, and therefore even certain terrorist groups and doomsday cults seem unlikely to turn to SSSBs.
5. If humanity or Earth-originating artificial intelligence spreads across and settles the Solar System, the probability of catastrophic SSSB weaponization may rise in the future.
6. Once again, however, SSSBs will remain relatively useless weapons, in spite of their high destructive potential. Moreover, many better weapons may be available to the kinds of advanced civilizations that can colonize a solar system.

Generalizing the last point, our uncertainty on specific military technologies of the future makes “picking winners” unlikely to be an effective prioritization strategy. Rather than try to predict specific paths of techno-military development, therefore, philanthropists should consider investing in more threat-agnostic interventions. General capacity building for arms control of novel WMD, improved international coordination mechanisms, and investment into better governance of space resources may be more cost effective options. I also recommend more work on risks from unknown future WMD and more work on space resource allocation, especially in a world with advanced artificial intelligence. In an appendix, I briefly discuss cybersecurity implications of SSSB deflection.

The SSSB Redirection Threat in Theory

Humans have long worried about the risks that asteroids pose to human survival, for good reasons — were a sufficiently large asteroid to collide with Earth, it could destroy humanity as it destroyed the dinosaurs (and about 75% of species alive at the time of end-Cretaceous mass extinction).^[3] Thanks to far-sighted actions by scientist-activists and policymakers, humanity has tracked the vast majority of dangerous asteroids. In the words of the existential risk scholar Toby Ord, “no other existential risk is as well-handled as that of asteroids and comets.”^[4] In 2022, NASA successfully conducted its Double Asteroid Redirection Test (DART), a further step towards successful planetary defense.^[5]

Nonetheless, as Ord acknowledges, there is a risk that as humanity develops dual-use SSSB deflection technologies: “methods for deflecting asteroids away from Earth also make it possible to deflect asteroids towards earth.”^[6] This is not a new concern. The earliest mention I am aware of is not —as is sometimes incorrectly claimed^[7] — by Carl Sagan in 1992, but in the early 1960s by the American aerospace thought leader Dandridge Cole, outlined in his and Donald William Cox’s 1964 Islands in Space: The Challenge of the Planetoids.^[8] It is also mentioned in various science fiction stories. Sagan further popularized the idea in various articles in the 1990s and his book Pale Blue Dot, and is usually associated with it (hence the title of this post).^[9] Since then, the fear periodically surfaces in popular science articles around news of asteroids or planetary defense. Daniel Deudney also discusses the threat of planetary bombardment in his 2020 book Dark Skies, which I cite throughout this document.

The concern about weaponizing asteroid deflection technology has also appeared in other work on space security. Fin Moorhouse’s 80,000 Hours profile on space governance included a section on “Avoiding weaponised asteroid deflection” at the time of this writing.^[10] Citing Sagan, Moorhouse wrote, “the possibility of deliberate deflection of near-Earth objects could pose the greatest threat to Earth from space this century.”^[11] (N.B. this was accurate as of November 2024, but the problem profile has been updated to reflect arguments in this document as of 4 December 2024.)

At a basic level, the reason to worry is best summarized by tables 7.4.A. and 7.4.B. in Deudney’s Dark Skies, which compare the violence that nuclear weapons can inflict with the violence that asteroids can inflict. As point estimates, these are of limited use, but they provide a sense of different orders of magnitude of destructive power. I summarize this in a shortened table below:

In other words, if SSSBs could be weaponized, they could dwarf the destructive power of nuclear weapons by several orders of magnitude. We should at least examine this possibility in greater detail.

SSSBs are not operationally useful weapons for most actors

Despite the destructive potential of asteroids, however, there are several strong reasons to believe that the threat is smaller than Sagan et al. imply. In the following sections, I summarize the factors that mitigate concern about weaponized asteroids before arguing that these mitigating factors may deteriorate over longer timescales.

First, for most actors in the international system (especially major military states), intentional bombardment using SSSBs is a suboptimal weapon. Such a weapon would be:

1. Indiscriminate, potentially destroying friendly polities on the target planet, and possibly destroying valuable resources and territory that could be captured rather than destroyed.
2. Difficult to procure, relying on the availability of suitable objects in the vicinity at the time of attack.^[12] 
3. Procurement may become easier in a future with widespread movement of asteroids for commercial reasons, but SSSBs would remain single use…
4. Single-use. Unlike, for example, directed energy weapons, an asteroid can only be used once, and each use depletes the natural stock of available “munitions,” making it nonrenewable.
5. This would make them less credible and have a high cost-per-strike, especially if success is not guaranteed, as indicated in the next bullet.
6. Difficult to conceal, facilitating early detection. A suitable SSSB large enough for destructive attack would be known to both attacker and defender (and an attempt to artificially construct such an object would also be conspicuous).
7. Possible to re-deflect away from the target with similar technology.
8. Slow at the extreme distances of space, especially when compared with other hypothetical space weapons, like powerful directed energy weapons.^[13]
9. As a quick illustration, even if attackers used immense amounts of energy to accelerate objects to the speed of very fast asteroids like 2021 PH27, traveling at 240,000 mph, such objects would take over 24 days to travel the average distance from Mars to Earth (it would likely travel much further, from the asteroid belt).

Points 5 and 6 are especially important. The development of asteroid weapons pre-supposes the development of deflection technology, such that the defenders may be able to re-deflect an object away from its target. As discussed below, there is a defensive advantage; the defender needs to only move the object anywhere but the target.^[14] Most importantly, however, naturally-occurring SSSBs are slow at the extreme distances of space. With sufficient situational awareness, this takes away the element of surprise, facilitating defense. Slowness also makes negotiations exceedingly difficult; by the time the weapon reaches its target, the political situation may have changed significantly.^[15] This also makes negotiations and changing minds to obtain political objectives — the ultimate aim of wars — more difficult.

SSSBs could theoretically be accelerated to immense speed with the right technology and large amounts of energy (e.g. by harnessing the power of the Sun with a Dyson sphere/swarm), which would make defense more difficult, but the kinds of advanced civilizations that are able to expend massive amounts of energy to redirect large asteroids at their whim likely have access to other technologies that would be far more useful as weapons. I discuss this below ("Advanced civilizations will likely have developed better weapons").

Moreover, most actors in the international system don’t optimize for maximum damage. As Stephen Clare and I wrote in our report on transformative technologies, militaries generally have specific objectives, such as capturing territory, securing waterways, or targeted killings of insurgent group leaders, and seek weapon systems that can achieve these objectives. I think this is likely the weakest objection — sprawling military bureaucracies occasionally pursue mass-destruction capabilities with questionable military utility (e.g., the Soviet Union’s work on “strategic” biological weapons). Moreover, some terrorist groups (like Aum Shinrikyo) and disturbed individuals appear to have near-omnicidal ideologies. I explore this further below (“near-omnicidal actors have more accessible WMD options.”) Nonetheless, it is not clear what marginal military benefits an asteroid bombardment capability would give to a powerful nuclear-armed state — the kind of actor most likely to have the immense resources for targeted asteroid deflection.

Deflection capability does not necessarily entail targeting capability

First, one important point is that targeting is more difficult than deflection — i.e. asteroid bombardment favors the defense.^[16] The intuition behind this consideration is straightforward. We are comparing two targets: Earth and ¬Earth, i.e. literally anywhere else. The size of the target ¬Earth is far larger than Earth, as it is the rest of the possible deflection area. This possible deflection area obviously depends on factors like the size of the object, its speed, its distance from Earth, etc., but the basic spatial intuition looks like this:

It is therefore not necessarily the case that improving deflection technology improves targeting technology. Deflection technology that can hit anywhere but Earth may be significantly less sophisticated than a technology that can hit a target as small as Earth is in (the targetable area of) space. Hitting ¬Earth when an object is on trajectory near Earth might require a simple push away from Earth. Hitting the small target of Earth (or even more difficult, a specific city or country on Earth), on the other hand, may require the use of sophisticated targeting systems, strong space situational awareness (SSA)^[17], and course-correcting technologies.

This problem for the attacker is especially pronounced if defenders also deploy re-direction technology. Large resources spent on a one-shot, non-renewable SSSB would be wasted if the defender could subsequently redirect the SSSB away from the target. Course-correcting technologies might allow an attacker to again direct the (now twice-redirected) SSSB towards the target, but the structural advantage remains with the defender — the non-target area is larger than the target area. The further away the attacker is from the target, and the better the defenders’ SSA (i.e. the more warning time the defender can obtain) the more the defender has this advantage. Even in an arms race between attackers and defenders, the defense would likely continue to dominate.

Earth’s relatively larger gravity well does, however, make it easier to target than smaller targets in the Solar System (e.g. the Moon, Mars, Lagrange-point settlements), a consideration I discuss below, in The Longer-Term Threat.^[18] First, however, I argue that the kinds of near-omnicidal actors who might realistically be most likely to deploy SSSB weapons are unlikely to use their resources in this way.

(Near-)Omnicidal Actors Have More Accessible WMD Options

Again, large SSSBs are too indiscriminate to be useful to all but certain near-omnicidal actors, at least in the near term. The kind of damage and climate change a massive impact would cause would have unclear deterrent effects, with huge backfire problems.^[19] Powerful states have other options in conventional and nuclear strikes. Thus, RAND scholars argued in one of the few public assessments of the utility of asteroid weapons, “For nations that already have nuclear arsenals, asteroid weapons might be of only academic interest.”^[20] Kinetic orbital bombardment — “rods from god” stationed in Earth orbit — would seem to be far more reliable (because always available) and easier to target, while achieving a similar psychological and physical effect as asteroids. The RAND analysis Space Weapons, Earth Wars (especially Appendix C), presents one of the most complete treatments of SSSB weapons. After examining problems of size, frequency, and availability, they conclude:

“With some patience, waiting perhaps a month or two, suitable asteroids could be routinely found that would produce weapon effects equivalent to nuclear weapons with yields ranging from tens of kilotons to many megatons. With some effort, they could be diverted to weapon [sic] using technology (and extensive supporting infrastructure) similar to that for exploiting lunar materials, generating solar power with satellites, or defending against asteroids. However, at best, it would take months after a decision to use one as a weapon to reach the desired conclusion.”

The slowness of these weapons is a point I return to later. The kinds of actors interested in such a weapon of mass destruction, moreover, have many cheaper alternatives for WMD. Asteroid deflection remains extremely costly and difficult; any actor who can unilaterally deflect a large asteroid likely has the resources and know-how to create an arsenal of simpler WMD: “Because much cheaper, more responsive weapons of mass destruction are readily available, [SSSB weaponization] is likely to remain safely in the realm of science fiction.”^[21] To make this more concrete: unless humanity gets its act together, mass destruction will be simpler with contagious biological weapons — and much cheaper than redirecting SSSBs.

The Risk of SSSB Deflection Technology May Rise in the Future

In short, I think that Sagan et al. overstate the risk of weaponized SSSB deflection. Over the longer-term, however, the character of this threat may change if humanity spreads into Earth orbit, Solar orbit, and beyond. Several analysts have suggested that Earth-originating civilization may naturally settle beyond Earth, simply because of the immense resources available in the Solar System and beyond, and to de-risk the situation of having all our civilizational eggs in one basket.^[22] In his book-length effort to argue against such space expansionism, Daniel Deudney argues that fears of asteroidal bombardment are “more plausible after successful colonization than among Earth-bound states.”^[23]

In Deudney’s argument, the main factors contributing to this changing risk profile are:^[24]

• Multiple factions of intelligent civilization exist in the solar system, potentially with radically different aims or value systems.
• This may include non-human intelligence, e.g. AI-enabled space settlements.
• In a world where widespread space settlement is possible, SSSB deflection may be both possible and necessary, and the sophistication of space technology makes targeting easier.
• E.g., because this technology is a pre-requisite for the effective use of immense resources on asteroids and comets.
• Mundicide^[25] becomes possible as the problem of backfire effects (e.g., climate effects on the same world one inhabits) decreases.
• Different gravity wells make some civilizations more vulnerable to kinetic bombardment than others, and may thus make the pursuit of SSSB weapons appear rational to the less-vulnerable civilizations.^[26]
• Earth civilization sits “at the bottom of a much deeper gravity well. Because of its closer proximity to the main asteroid belt and its almost complete reliance on space technology, a Martian colony would probably achieve military superiority over Terra long before its population reached even a sizeable fraction of Earth’s.”^[27]
• Differences in gravity wells also make deterrence and a balance of power much more difficult — Earth is at a disadvantage to Mars; Mars at a disadvantage to civilizations that are not planet-bound.

In my understanding, this boils down to two problems:

1. In some future scenarios, with advanced space technology, the SSSB bombardment balance shifts in favor of some non-Earth settlements (it’s easier for Mars to attack Earth than the other way around).
2. With the settlement of the solar system, weaponizing SSSB deflection technology may become militarily advantageous for some off-Earth political entities.
   1. Simultaneously, these entities may have greater technological know-how than Earth because their economy is built on manipulating space objects.

This suggests that in a future where intelligent agents settle the solar system, SSSB bombardment may become more attractive for some settlements than for others. Unlike other potentially dangerous weapons with fewer legitimate civilian uses, SSSB-deflection/movement technologies may have legitimate and widespread uses in the future (for economic development in maneuvering asteroids for mining and other uses). This threat does, however, rely on the existence of mundicidal intent among the groups who are able to move asteroids; others seem to find this obvious, positing that e.g. Mars civilization would naturally diverge from Earth civilization and view all of Earth as a potential adversary. But I see it equally plausible that political divisions will remain on Earth, that adversaries would want to target some groups/territories and not others, and that they may not want to destroy Earth itself, but rather preserve its natural resources for conquest. The point stands, however, that the existence of extraterrestrial settlements may increase the probability of mundicidal intent.

SSSBs will likely remain ineffective weapons in the future

Although technological advances and settlement of the Solar system may make SSSB weaponization relatively more appealing for some actors than for others, however, I think most of the arguments above still stand.

First, deflection is still easier than targeting. In a purely offensive arms race, where e.g. Mars civilization and Earth civilization are both developing SSSB-bombardment technology, Mars civilization does have an advantage of better access to asteroids and the difference in gravity wells.^[28] But there would also be investment in defensive technologies, including early-warning systems and re-deflection technology. Once again, the non-target area is much smaller than the target area, so defense would be easier than offense.

Second, SSSBs remain clunky weapons. Many of the factors against SSSB use listed above — procurement difficulties, non-renewability, long warning times, etc. — would still be true in the future imagined here. Attackers could increase the velocity of the object with great energy expenditure, which might make defense more difficult, but I argue below that this energy could be much more effectively used in other kinds of weapons, such as directed energy weapons.

Third, the kind of civilization that is able to create SSSB-bombardment technology is likely able to develop much better weapons. The next section briefly discusses this.

Advanced civilizations will likely have developed better weapons

Other viable space weapons have many advantages over SSSBs, and the extreme energy required to redirect SSSBs could be much more fruitfully used for other weapons. For example, directed energy weapons would be extremely fast (because they use electromagnetic radiation), and therefore impossible to evade, difficult to defend against, reusable, easy to target (no complicated ballistics/trajectory calculations) and potentially have variable destructive power, allowing “dial-a-yield” calibration to different targets and missions. An advanced civilization with large amounts of power (e.g. by monopolizing Solar energy) would have a renewable access to these kinds of weapons, which also could be developed with plausible civilian energy applications.

Why would a highly advanced civilization choose to use large rocks as its weapons? Trends in warfare have been towards precision targeting and increased speed in recent years. The threat “if you don’t do what I want, you will get hit by a large rock in several months” (at which point the negotiating positions and political situation may have changed completely) is much less compelling than “if you don’t do what I want, you will be instantly obliterated.” Other weapons that an advanced civilization might pursue instead might include:

• Advanced bioweapons (e.g. targeting specific populations)
• Advanced cyber operations
• Advanced information weapons/cognitive warfare^[29]
• Directed energy weapons
• Improved nuclear weapons

Importantly, some of these weapons would leave valuable infrastructure intact.

The broader point I am trying to make is not necessarily that we certainly won’t see the development of asteroid weapons, though I estimate a low likelihood of their development. Rather, I am trying to make a point about the many different possible futures, and the difficulty of predicting specific future weapon systems:

1. There are many possible military futures. SSSB weaponization is only one highly specific scenario among many that we can imagine, and even more that we cannot imagine.
2. We struggle to predict technological progress, and have no strong reasons to believe that future civilizations might preferentially pursue SSSB weaponization.
   Thus, it seems highly unlikely to me that investing now in averting a possible future weapon is going to be a competitive use of philanthropic funds, especially given the arguments against SSSB weapons.

What Philanthropists Could Fund

I don’t recommend funding SSSB-specific interventions, but instead recommend investing in general arms control capacity and improving international coordination mechanisms. Nonetheless, in the next section I briefly discuss what philanthropists could fund if they believed that SSSB weaponization is a threat, before outlining more general funding opportunities.

SSSB-specific interventions

If one were convinced by the importance of mitigating these risks, one could seek to beneficially leverage near-term international cooperation. This might take the form of promoting the idea that SSSB deflection should be performed only by international consortia.^[30] It might also take the form of international agreements on SSSB deflection, which may include:

• “Rules of the road” and confidence-building measures (CBM) ensuring transparency on SSSB deflection (e.g. sharing exact deflection paths with other spacefaring countries)
• Specific binding agreements, e.g.:
• Banning highly maneuverable (and therefore weaponizable) SSSB deflection technologies.^[31]
• Setting strict limits on the distance that asteroid and comet mining activities must keep from Earth when moving valuable space resources closer for economic activity.
• Creating incident-sharing mechanisms for accidents and miscalculations related to SSSB deflection and maybe space traffic management in general. (“Incidents at Sea Agreement, but for Space”)

Philanthropists could fund a variety of efforts to support this work:

• Multi-year think tank research on SSSB weaponization risks.
• Track II dialogues between major space-faring nations and companies to discuss these issues.
• A cybersecurity-focused project on SSSB deflection similar to RAND’s Securing Model Weights project.
• Think tank research and policy advocacy on dual-use space technologies.
• This also has benefits for more near-term Great Power conflict problems.

More cost-effective interventions that also help with the SSSB threat

Again, however, it seems highly unlikely to me that investing now in averting a possible future weapon is going to be a competitive use of philanthropic funds. Instead, we might want to invest in general arms control capacity and improving international coordination mechanisms. This may be especially important if we are concerned about a future of explosive growth (possibly AI-driven).^[32] This may include preventing extreme concentrations of power (e.g. control of the sun) that could be instrumental to many different kinds of weapons. For example, we could invest in equitable governance of space resources, and amend the outer space legal regime to make the emergence of Solar System hegemons (and therefore long-term value lock-in) less likely. I don’t think “picking winners” from different possible techno-military futures is likely to be a cost-effective approach, even though SSSB weaponization is interesting.

Philanthropists could fund various threat-agnostic problems:

• Risk-general arms control fellowships (e.g. at ACONA) targeted at catastrophic risks to increase the talent pool for regulating future weapons.
• Other capacity-building on arms control and existential risks (training and Intergovernmental Personnel Act placements for talented scientists and technologists to work on emerging tech policy).
• Information security for high-consequence systems.
• Technology forecasting initiatives.

Conclusion

To summarize, I argued that three things are true about the dual-use risks of asteroid deflection technologies:

1. This risk appears over-hyped in the near-term, contra Sagan et al.
2. Therefore, deflection-focused projects appear unlikely to be competitive with other cost-effective philanthropic interventions.
3. Over the long-term, this risk may grow, but this still seems unlikely.
4. Given our uncertainties about the character of future space weapons, however, broad interventions on arms control and resource allocation may be better uses of philanthropic funds.

I think there is a lot more research to be done on space security and the governance of space resources, and think this could be a major area for further research in longtermism/global catastrophic and existential risks.

Appendix: Opportunistic Threats and Cybersecurity

I believe that few states and non-state groups would turn to SSSB deflection as their WMD of choice. Nonetheless, if SSSB deflection (or a large-scale test of SSSB deflection) were imminent, we might expect opportunistic actors to take advantage of this situation. A terrorist group or rogue state with powerful cyber capabilities, therefore, might attempt to hijack the control system of an SSSB deflection program, threatening mass destruction.

Given the targeting difficulties discussed above, I would expect the main threat model to be:

1. A large object is on track to collide with Earth;
2. A civilian space agency organizes a large scale effort to redirect the object;
3. A terrorist group (e.g. a doomsday cult inspired by the “celestial object” à la Heaven’s Gate) launches a cyber operation targeting the redirection effort, disabling it.
4. The object remains on track to collide with Earth (either until the group’s demands are met, or until impact).

This problem, however, would only arise in the exceedingly rare instances when large asteroids are actually on track to collide with Earth. A scenario that may arise more frequently if humanity continues to invest in planetary defense is deflection testing:

1. A large deflection test effort is organized to alter the orbit of an SSSB that is not already on track to collide with earth.
2. A group launches a cyber operation targeting the effort, directing the object towards Earth.

Yet again, however, this scenario runs into the targeting problem described above: a system designed to merely deflect away from one trajectory cannot necessarily be hijacked to strike a relatively small target — it is a more sophisticated kind of problem.

Moreover, these are highly specific scenarios, and — unlike other global catastrophic risks — philanthropists could address them on an ad hoc basis; if a large deflection effort were organized, then philanthropists may advocate for the cybersecurity of the deflection effort. Funding such work now appears unlikely to be competitive with other Founders Pledge-recommended funding opportunities, and may instead be a strong use case for an opportunistic future grant from the Patient Philanthropy Fund.

If the need arose, interventions to reduce this specific threat would be concrete and tractable:

• Fund work analogous to RAND’s Securing Model Weights project to better understand the cyber vulnerabilities of SSSB deflection programs.
• Fund work that promotes shared understandings, best practices, and “cyber norms” for various space programs.

• Fund work highlighting this threat to decision-makers in national governments and international organizations.

  Rather than focus on this particular threat, however, philanthropists could invest more in other projects designed to bolster the security of high-consequence technologies. For example, projects on securing AI model weights or on information security around nucleic acid synthesis may yield valuable insights for SSSB deflection system security. More broadly, a threat-agnostic “cybersecurity for high-consequence systems” project could provide a threat-agnostic way to address a range of related problems.

About Founders Pledge

Founders Pledge is a global nonprofit empowering entrepreneurs to do the most good possible with their charitable giving. We equip members with everything needed to maximize their impact, from evidence-led research and advice on the world’s most pressing problems, to comprehensive infrastructure for global grant-making, alongside opportunities to learn and connect. To date, they have pledged over $10 billion to charity and donated more than $950 million. We’re grateful to be funded by our members and other generous donors. founderspledge.com  

If you are interested in supporting work to mitigate global catastrophic risks, consider donating to the Global Catastrophic Risks Fund. 

1. ^{^}

    This is not a perfect term, as we are not necessarily limited to the solar system, and could potentially weaponize even planets and dwarf planets. However, I think it is more useful than other less inclusive or more confusing terms that are usually used in this context, including “asteroid,” “meteoroid”, “planetoid,” and “near-Earth object.”  I did consider “naturally-occurring kinetic energy weapons” (NOKEW), but decided that “knock-you” is too funny to be taken seriously.

2. ^{^}

    Really, high kinetic energy, rather than just high mass.

3. ^{^}

    For an in-depth overview, see “Asteroids and Comets” (67-74) in chapter 3, “Natural Risks,” of Toby Ord’s The Precipice: Existential Risks and the Future of Humanity.

4. ^{^}

    Ord, The Precipice, 72.

5. ^{^}

   “Double Asteroid Redirection Test (DART),” accessed March 12, 2024, https://science.nasa.gov/mission/dart/. 

6. ^{^}

    Ord, The Precipice, 73. Moorehouse writes, similarly: “any technology capable of deflecting an asteroid away from a collision course with Earth will make it easier to divert it toward Earth.”

7. ^{^}

    “Possibly the earliest mention of what is now known as the ‘deflection

   dilemma’ can be found in Carl Sagan’s 1992 article in the Bulletin of the Atomic Scientists.” Jakub Drmola and Miroslav Mareš, “Revisiting the Deflection Dilemma,” Astronomy & Geophysics 56, no. 5 (October 1, 2015): 5.15-5.18, https://doi.org/10.1093/astrogeo/atv167.

8. ^{^}

    I became aware of Cole and Cox’s work through its detailed discussion in Daniel Deudney, Dark Skies: Space Expansionism, Planetary Geopolitics, and the Ends of Humanity (Oxford, New York: Oxford University Press, 2020).

9. ^{^}

   Carl Sagan, Pale Blue Dot: A Vision of the Human Future in Space, Reprint edition (New York: Ballantine Books, 1997).

10. ^{^}

    Fin Moorehouse, “Space Governance - Problem Profile,” 80,000 Hours, accessed March 12, 2024, https://80000hours.org/problem-profiles/space-governance/.https://80000hours.org/problem-profiles/space-governance/. 

11. ^{^}

    Fin Moorehouse, “Space Governance - Problem Profile,” 80,000 Hours, accessed March 12, 2024, https://80000hours.org/problem-profiles/space-governance/. 

12. ^{^}

     The required “search” for naturally-occurring “munitions” would again add to the early warning time of the defender — suitable SSSB candidates at the time of attack would likely be known to both the attacker and defender.

13. ^{^}

     “at best, it would take months after a decision to use one as a weapon to reach the desired conclusion.” Robert Preston et al., “Space Weapons Earth Wars” (RAND Corporation, January 1, 2002), https://www.rand.org/pubs/monograph_reports/MR1209.html 

14. ^{^}

     Re-deflection may be more difficult if the attacker decided to explode the large object into many smaller objects still traveling towards the target.

15. ^{^}

     In special cases, this would be an advantage for credibly signaling commitment, but this does not

16. ^{^}

     “Advocates of planetary defense, most notably James Oberg, reply that Sagan’s scenarios are not plausible because altering an orbit to strike a specific target is more difficult than deflection.” Deudney, Dark Skies.

17. ^{^}

     The Aerospace Corporation defines SSA as “​​the knowledge, characterization, and practice of tracking space objects and their operational environment.” https://aerospace.org/ssi-space-situational-awareness 

18. ^{^}

     See Deudney, Dark Skies for an in-depth discussion of gravity wells and “planetary geopolitics.”

19. ^{^}

     In the words of the RAND analysts, the “effects [of large asteroids] are too great to be useful for strategic deterrence.” Robert Preston et al., “Space Weapons Earth Wars” (RAND Corporation, January 1, 2002), https://www.rand.org/pubs/monograph_reports/MR1209.html. And Deudney, in Dark Skies, writes, “... like the ‘nuclear winter’ blowback from nuclear war, the effects of a large asteroidal bombardment might also afflict the territory of the attacker.”

20. ^{^}

    Robert Preston et al., “Space Weapons Earth Wars” (RAND Corporation, January 1, 2002), https://www.rand.org/pubs/monograph_reports/MR1209.html.

21. ^{^}

    Space Weapons, Earth Wars, 183.

22. ^{^}

     “If we survive the next century, we are likely to build self-sufficient colonies in space. We would be motivated by self-interest to do so, as asteroids, moons, and planets have valuable resources to mine, and the technological requirements for colonization are not beyond imagination [...] Colonizing space sooner, rather than later, could reduce extinction risk [...] as a species’ survivability is closely related to the extent of its range” Jason G. Matheny, “Reducing the Risk of Human Extinction,” Risk Analysis 27, no. 5 (2007): 1335–44, https://doi.org/10.1111/j.1539-6924.2007.00960.x. For a criticism of this line of thinking, see Deudney, Dark Skies.

23. ^{^}

     Deudney, Dark Skies.

24. ^{^}

     Most of this is a summary of my understanding of various parts of Deudney’s Dark Skies. Some of it is my addition.

25. ^{^}

     I.e. the killing of an entire world. Deudney uses the word “cosmicide” to refer to the destruction of Earth via space-originating forces, even though conventional usage (e.g. “homicide” “patricide,” etc.) would suggest that these events are usually named for the attack’s target, not for its origin.

26. ^{^}

     “Because the Terran gravity well is so much steeper than Martian or asteroidal wells, space colonies have an advantage.” Deudney, Dark Skies.

27. ^{^}

     Deudney, Dark Skies.

28. ^{^}

     Though it’s not at all clear that space settlement would necessarily involve Mars in a central way. Thanks to Fin Moorhouse for this point.

29. ^{^}

    https://www.rand.org/content/dam/rand/pubs/testimonies/CT400/CT473/RAND_CT473.pdf, https://www.rand.org/pubs/research_reports/RRA853-1.html 

30. ^{^}

     “An alternative approach to the collision threat, advanced by this author and others, proposes that an international consortium of spacefaring states undertake such efforts.” (Deudney, Dark Skies).

31. ^{^}

     NB this would have consequences for certain missile-related technologies, and would need to be framed very carefully.

32. ^{^}

     Thanks to Will MacAskill for pointing me to this.