A Really Cool Blog

… about science & space, people & politics, various musings & other cool things too.

Page 2 of 61

“Combatant Commands” & U.S. Defense Policy

As a key organizational and operational construct within the highest levels of the United States’ military, combatant commands – along with their commanders – play significant roles in U.S. defense policy. These commands, organized either on a geographical or functional basis, provide effective command and control of military forces, regardless of branch of service, in peace and war. By their nature, they offer U.S. defense policy and application of military force several benefits, but likewise face several challenges.

The combatant command structure emerged out of an evolution, prompted by historical circumstances, in the military’s approach to force unity and command and control. Prior to the Second World War, the services operated independently. Coordination between branches of the armed forces was frequently marred by bureaucratic distrust and service rivalry. The experience of the Second World War, particularly in the Pacific theater, demonstrated the operational pitfalls of this arrangement; differences between the Army and Navy precluded any sort of consistent unified command. While commanders were able to work together to defeat Japan, it wasn’t without considerable fiction and infighting as commanders within the independent, powerful commands competed for the same resources and the attention of senior leadership.

Given the dynamics of the post-war geopolitical environment, a global military presence was viewed as a guarantee against Communist expansion. Achieving such presence necessitated an effective, geographically-focused, long-term, joint-command arrangement. Between this and the lessons learned from the Second World War, legislation over the course of three decades established geographic unified commands, clarified lines of command for unified commanders, and, per the Goldwater-Nichols Department of Defense Reorganization Act of 1986, permanently assigned forces to unified combatant commands.

The premise and purpose of Combatant Commands remains essentially the same today. Combatant commanders integrate and utilize air, land, sea, and amphibious forces within the geographic or functional area of their control to achieve U.S. national security objectives. They are responsible for accomplishing missions assigned to their region, designating objectives, assigning tasks, and giving authoritative direction over all aspects of military operations as well as all aspects of joint-training and logistics. Likewise, they are responsible for establishing command relationships with subordinate commands and for organizing subordinate units as deemed necessary. They take direction from the Commander in Chief – the President – and are responsible to the Secretary of Defense.

Combatant commands offer several benefits. First, and perhaps most importantly, they provide unified and integrated command and control of cross-service forces in a theater. This is critical for the United States’ current military doctrines, which emphasis use of joint-service application of force. By minimizing confusion, competition, and conflict between the services in operations, the military can more effectively conduct its primary mission of keeping America secure and waging conflict. For functional combatant commands, integrating all elements of a particular warfighting domain – such as cyber, space, and strategic forces – under a single chain of command allows for easier oversight, coordination, and rapidity of action.

Second, as regional combatant commanders wield considerable influence over American foreign policy in their area of responsibility and have the resources and authority to act quickly, they can be disproportionately effective, compared to other American foreign policy actors, in shaping dialogue and events in theater so as to advance American security and foreign policy interests. Likewise, their relationships and integration with regional counterparts through military engagement programs provides them policy leverage that other elements of the U.S. foreign policy apparatus may not possess. Frequently, the power to “get something done” may lie with a foreign military rather than in a foreign ministry. As such, the United States’ government has increasingly deferred to and provided resources for combatant commanders to achieve regional objectives, regardless of whether those objectives are distinctly military in nature or not.

Combatant commands also come with challenges both inherent to their structure and manifest from their historical and contemporary execution. First is manifest from the policy influence that combatant commanders have. Commentators have suggested that combatant commanders have overstepped their bounds in executing American foreign policy. The resources of the DoD and combatant commands may provide a quick response to a foreign issue, but not necessarily the correct or most effective response. Once combatant commands are engaged in a foreign policy effort, they tend to continue, as it is difficult and time consuming to transfer responsibilities to civil organizations. Because of this, civilian agencies that are, by law and by tradition, supposed to manage U.S. foreign relations have tended to be relegated to an increasingly secondary role – and have faced budgetary pressures reflecting that. Similarly, the influence of combatant commanders and their role in executing U.S. foreign policy raises questions about the civilian oversight and control.

This issue ultimately comes down to a key point bout handling the issues of today’s world. Military engagement programs with other countries can be seen only as part of the overall engagement activity of the U. S. government. Modern security challenges such as terrorism, narcotics, smuggling, and international criminal networks can no longer be managed as single agency programs but must be integrated into “whole of government” programs. That combatant commands and their commanders have widened, not closed, the gap between the foreign policy agencies involved in a “whole of government” approach is a distinct challenge and concern.

Next, the geographically organized and regionally focused combatant command structure does not readily accommodate solutions to post-Cold War era challenges such as the emergence of transnational groups and regional powers, which operate outside traditional borders and in nontraditional domains. Strict geographic regionalism is no longer how the world operates. From this stems issues of and disagreement over responsibility over problems that extend across areas of command. Coordination on handling these new and emerging challenges will be increasingly difficult if the rigidity of the combatant command structure is maintained. Meanwhile, the lead for two major security concerns of priority today – international terrorism and the proliferation of weapons of mass destruction – falls squarely on the shoulders of two functional combatant commands – Special Operations command and Strategic Command, respectively. The geographic combatant commands are, by their nature, not designated as the lead organizations for managing two of the United States’ primary military challenges.

Considering these challenges, experts both within and outside the Department of Defense have offered several suggestions for reform of the combatant command structure – or fundamental change to the way the United States organizes command and control. Nonetheless, in the present day, combatant commands play a vital role within the United States’ military and defense policy, and their commanders are actors of considerable influence for American foreign policy. The combatant command structure has brought the United States considerable military advantage throughout the past decades – even if it comes with drawbacks and faces increasingly difficult questions and issues.

On “Strategy”

For leaders and decision-makers grappling with a diversity of complex, occasionally competing, national interests, as well as with challenges in every region of the world, the ability to hierarchize interest and threat vis-à-vis national goals enables them to effectively allocate attention and resources to priorities. Ideally, this is achieved through crafting and implementing “strategy.” While the process of formulating and employing strategy is rife with challenges and shortcomings, it is nonetheless one of significant meaning and importance for U.S. foreign and defense policy. In lack of a strategy – a firm grasp, prioritization, and articulation of national interests – there is a danger that policy will wander according to the crisis or fashion of the moment and that policymakers will be exposed to distraction, confusion, and be led to overreach.

A “good” strategy is a method, not an endpoint. It is a process of defining political ends, of identifying critical approaches for achieving those ends, and of problem-solving in circumstances where much lies outside decision-makers’ ability to control. Good strategy reduces uncertainty in times of ambiguity and clarifies which options for action and reaction are in line with national priorities, preparing leaders to act in integrated ways even when surprised. Likewise, in an international environment saturated with potential challenges, threats, and opportunities, a strategy delivers guidance on which should be deemed and treated as critical for national attention and which should not.

Beyond providing a methodology for thinking about and acting upon decisions, crafting strategy offers functional benefits for the organizations that engage in it. First, a strategy helps bridge the gap between the government’s resources and policy ambitions. Limitless policy and programmatic opportunities are bounded by real constraints in funding and manpower. Through the act of declaring top priorities and goals, policymakers can identify the programs that support them most closely, which warrant a greater share of the limited available budget, and appropriate accordingly. Next, the process itself of developing a strategy helps focus policymakers and bureaucracies. As with any thought-exercise, it provides policymakers an opportunity to consider more closely national needs, goals, and options to pursue them. It allows them to hone in on an integrated way of thinking and doing amidst the confusion of constantly changing events.  Articulating priorities provides guidance to implementers throughout the bureaucracy for devising executable steps toward achieving them.

Finally, strategy serves an important messaging function that conveys the intent of defense policy to audiences both domestic and international. In the domestic sphere, it informs the public and Congress of the Department of Defenses’ interests, concerns, and policy priorities. Deciding whether to abide by a strategy through authorizations and appropriations that align with its tenets is ultimately the prerogative of Congress; nonetheless, it is valuable for members and their staff to understand how the Department of Defense intends to approach issues – how it “thinks” – when developing policy. Internationally, it communicates to friends and foes alike how the United States sees global developments and how it will act to shape them and in response. This offers allies an opportunity to align their own strategies to support and supplement the United States, and serves as a form of warning or deterrent to potential adversaries.

The strategies that most effectively achieve these aims consist of certain key elements. First, and perhaps most important, a good strategy is realistic. One that fails to consider, or misconstrues, the real limits of national means and power defeats the purpose of providing decision-makers guidance on how best to employ available resources to advance the national interest. A realistic strategy acknowledges constraints; indeed, in doing so, it prepares decision-makers for making choices that account for these constraints and work around them. Similarly, good strategies are limited in objective and scope. A broad strategy makes it difficult for policymakers to prioritize top objectives, especially given the limited means available to execute, and runs the risk of leading to overextension and decisions that conflict with each other. With a broad strategy, hierarchies of interests and threats tend to become blurred or collapse altogether.  Instead of cataloguing every interest, strategy should consist of decision rules that allow for application to events as they unfold.

Good strategies are clear in direction, to allow for their consistent implementation through all levels of the bureaucracy and across multiple decisionmakers. Strategies that are unclear or ambiguous in tenet and direction lead to different interpretations among various levels of the bureaucracy, which can muddle execution and lead to non-integrated action. Finally, a good strategy is one that is conceptualized and crafted through an inclusive process that involves all stakeholders. Not only does this secure needed buy-in from the wide array of actors responsible for implementing the strategy, but it allows for a diversity of perspectives to be offered, taken into consideration, and included.

The process of crafting and implementing defense strategy is a challenging one. With multiple constituencies seeking to involve their particular interests or “pet projects” in a strategic framework, the strategy that emerges risks being burdened by excessive and specific detail. Detail can be constraining – a good strategy is limited in scope and breadth – as it minimizes the space for recalibration and adaptation to changing circumstances. The discipline to minimize detail in a strategy is not always forthcoming. Meanwhile, competition regarding priorities, programs, and approaches to be included in a strategy tends to emerge between career bureaucrats and political appointees.  By virtue of their different positions and prerogatives, both have different – often-contradictory – perspectives on the best course of action to take. Political appointees seek to advance the agenda of the Administration and correspondingly approach strategy with a short-term, politicized focus. Particularly within the Department of Defense, career bureaucrats with long institutional histories and knowledge tend to seek preservation of the status-quo and to protect or give priority to their own institutional interests. Striking a balance between these perspectives can be a difficult task, necessitating significant compromise and limiting the extent to which an Administration can pursue its defense policy aims and objectives.

A further challenge in executing strategy is linking the long-term goals set by planners and “strategists” with the short- and mid-term actions that operators can reasonably take to deal with a crisis of the moment. Imperatives do not always align on the short- and long-term perfectly, and planners and operators have different institutional incentives to address tensions. Even with good strategies that are clear in direction and guidance, implementation will suffer because of differing pressures and competencies throughout all levels of the bureaucracy. Finally, a continuing, indeed, inherent, challenge with crafting and implementing strategy is simply that of its nature. How does one know that a strategy is good? How can one be sure that a strategy is working? These are questions without clear answers – decision-makers can work only with their best assumptions and predictions about the world and alter those assumptions as needed. Strategies need to both offer a framework for dealing with security challenges and world events but also simultaneously allow for flexibility and adaptability to new situations.

Despite these challenges, the act of developing, articulating, and implementing strategy is and will remain an invaluable tool for the Department of Defense and American defense policy. An integrated way of thinking about and handling the world’s challenges, even if flawed, is always superior for advancing national interest than haphazardly stumbling into the crises of vogue.

Why Should We Go? Reevaluating the Rationales for Human Spaceflight in the 21st Century

In the 56 years since Yuri Gagarin became the first human to cross the Kármán line and slip into outer space, over 530 others have done the same. Between the present-day plans of Russia, China, NASA, and several private companies, along with the longer-term aspirations of others, human spaceflight appears poised to continue into the foreseeable future. Yet as the quinquagenary of the first human Moon landing quickly approaches, an important question remains without a definitive answer: why?

For many – if not most – who study, work on, or follow human spaceflight, the prevailing reason for its continuation intuitively exists beyond practical or material motivations: simply because space, to quote President Kennedy’s famous speech at Rice University in 1962, “is there.” To them (us), a meaningful rationale is not so much a justification of why human spaceflight could continue as it is a defense of why it should. Humanity’s expansion into space is taken as an ordained inevitability and our pursuit of it a compelled calling. It is understandable, then, the consternation felt when confronted with the hard reality that a majority views human spaceflight as a lesser priority than other projects, that humans have been essentially mired in low Earth orbit since the apex in exploration that was the lunar landings, and that most of the more audacious human spaceflight efforts have faced intense fiscal pressures, programmatic instability, or outright cancellation.

In this present era of national challenges which demand the attention of policymakers and the public – economic uncertainty, international turmoil and change, domestic political and social upheaval – it is more important than ever for the space community to reflect on the purpose of human space exploration. What value does it hold? Are the oft-repeated reasons that have sought to justify the enormous cost of human spaceflight applicable in the current day? Will advocates of a robust human presence in space be met with the same disappointments in the coming decades as they have in those that have passed?

This will all ultimately depend on how the question of human spaceflight’s efficacy as a tool for society is answered. Whether the justifications for human spaceflight are cohesive with national desires will be, as Dr. John Logsdon noted in Which Direction in Space, “key to decisions on the future of government space programs around the world.” If found, “the 21st century could see the full realization of both the practical and inspirational potentials of space.”[1] If not, human spaceflight may remain a far muted shadow of the grandiose visions (and expectations) put forth by the likes of von Braun and O’Neill.

As the United States works to develop a coherent and cohesive national space strategy, a reconsideration of the rationales behind human spaceflight and their relevance in the policy arena is increasingly warranted. Reevaluation and discussion of these rationales can, hopefully, enable the space community to better align its intent and aspirations with the needs of the nation. At the same time, the space “ecosystem” is rapidly and dramatically evolving. Private and commercial entrants with human spaceflight aspirations are becoming more extricated from the pressures and constraints of public policy and funding. Will rationales justifying their efforts even be necessary? Perhaps, for as long as they continue to interface with (and rely upon support from) government-run programs. But as spaceflight becomes more democratized with actors who can privately finance their efforts, the fundamental issue of “why” may simply turn into a question of markets and economics.

To begin a discussion on the rationales of spaceflight, it need be acknowledged that the space effort does not exist in a vacuum (at least metaphorically). Rather, for most of its history, space exploration – particularly that involving human flight – has been a matter of public policy. Especially in the United States, funding and programmatic decisions have been the purview of leaders in the executive and legislative branches. While granted leeway in strategic and practical implementation of missions, NASA as an agency is subordinated in goal-setting and resource allocation to the ideas, decisions, and whims of its political leaders. The character of the human spaceflight program, its successes, stumbles, and failures, are a result.

In public policymaking, rationales matter – persuasive ones appealing to the whole of or influential actors within society especially so. This is significant in a country such as the United States, which has a political system sharply characterized by competing groups – political parties, advocacy groups, industry organizations, scientific societies, to name a few – with distinct and active interest in shaping the nation’s direction, its allocation of resources and energy. Their goals and aspirations are often starkly different, at times contradictory. Their motivations range from the ideological to the practical and material. And they exist and operate in a resource-constrained environment. While the federal budget may grow and shrink, the United States’ government is limited to a finite amount of money it can throw toward its entire portfolio of projects and activities. Where the government chooses to allocate those funds is the product of policymaking – the process of judging and prioritizing the disparate needs and desires of stakeholders in the system.

The same holds true abroad, in countries with similarly representative political systems and those without. Even in authoritarian systems and command economies, limitless opportunities are bounded by limited resources. Where leaders decide to put their time and money are strategic decisions which cater to the interests of internal actors with political clout or which advance the standing – be it diplomatic, economic, or prestigious – of the state.

Those who advance the cause of publicly-funded human spaceflight find themselves operating in a larger political context and competing against equally worthy causes. To win support (and money), the rationale they put forth needs to be persuasive across a broad spectrum of political factions, appeal to potential supporters and opponents, and meet the perceived needs of large and diverse economic and political constituencies. In lack of a persuasive rationale, a proposed effort will be superseded by others seen by the broader polity as more realistically and immediately achievable or necessary.

This is a challenging task. That human spaceflight has, throughout its history, remained an ancillary part of public policy reflects the space community’s continuing struggle to arrive at a rationale compelling enough to heighten its stature on the policy agenda. Of course, this challenge is compounded further by the present-day practical circumstances of spaceflight: that “space is hard” – dangerous, costly, resource and time consuming, and technically difficult. Where there is overlap in purpose between human spaceflight and a cheaper terrestrial option, it is difficult to justify going with the former over the later.

Dr. Logsdon described this as the “potential liabilities associated with using space systems to carry out centrally important functions for society,”

“Such systems remain expensive to develop and launch. They have mixed records of reliability, and repair of problems or failures is at best very difficult… [w]hen these factors are taken into account, do space systems indeed compare favorably with terrestrial alternatives for carrying out the same function? Are there unique and valuable functions that only space systems can perform?”[2]

Dr. Logsdon’s last question is particularly key. Is there a function that only human spaceflight can perform, one which outweighs its costs? If there is, it has evidently not been properly articulated to policymakers or executed to its fullest potential in the past few decades. This notion is reflected in the Columbia Accident Investigation Board’s 2003 report, which made a condemning recognition of “the lack, over the past three decades, of any national mandate providing NASA a compelling mission requiring human presence in space.”[3]

Against this framing of the political environment’s dynamics, the most commonly advanced rationales for human spaceflight can be better addressed and understood. Academics, policymakers, industry leaders, and space enthusiasts have all weighed in with their justifications for why we – as a country, society, and species – should and will send humans into space. Many are deduced in retrospect, analyses informed by historical actions taken to meet past circumstances, challenges, and opportunities. Some are longer-term utopian prognoses, driven by ideological ideals, economic aspirations, and concepts of indefinite human survival. Others are more philosophical in nature, drawing on such notions as humanity’s inherently exploratory and adventurous character and “destiny.” Perspectives are diverse and occasionally disparate.

The oft-repeated rationales for human spaceflight are also reflective of the interests held by the various stakeholders of the space effort. For scientists and researchers, for example, it is to advance scientific research and knowledge. For commercial space companies, especially those that have emerged in the recent decade, it is to advance the sphere of economic activity beyond Earth. For policymakers, it is frequently cited as a means to advance the interests of their constituency and the nation – spaceflight creates high-skill, high-wage jobs, inspires the next generation of workers to fill those jobs, and is a tool for international prestige, cooperation, and leadership.

Roger Launius’s seminal Compelling Rationales for Spaceflight laid out five major themes used to justify efforts in space: geopolitics/national pride and prestige; national security and military applications; economic competitiveness; scientific discovery; and human destiny/survival of the species. Returning to Dr. Logsdon’s question on the unique value of spaceflight, he noted that, of these, “only the human destiny/survival of the species and geopolitics agendas require humans to fly in space.”[4] This largely holds true, at least in the present day. Much scientific discovery in space can be and is accomplished through robotic spacecraft. National security space systems are all automated. Indeed, early efforts for national security-related human spaceflight, such as the MOL Program, were cancelled in favor of non-human spacecraft. Meanwhile, most of the present-day economic value derived from space is done through satellites orbiting the Earth.

Recognizing the pressures involved in public policymaking, the geopolitical rationale appears, at least historically, the most significant and compelling. Underlying this is the fact that international events and circumstances, acting as forcing functions, can either heighten or lessen human spaceflight’s stature as an element of public policy and policymakers’ willingness to allocate resources toward it. Human spaceflight has, at least historically, been most valued as a part of the foreign policy “toolbox,” as a method to deal with emerging external challenges. As Professor Roger Handburg put in his Rationales of the Space Program,

“one needs an incentive, a compelling focusing event, strong enough to break through the existing political status quo and to place the issue of space on the policy agenda for political decision-making and policy formulation.”[5]

Closely related to the geopolitical rationale is that of prestige and national pride. Human spaceflight, as an enormously challenging yet rewarding task, reflects a country’s scientific, technological, and industrial strength. It is meant to appeal to audiences both domestic and international. The pride rational of human spaceflight, in the view of Harold Goodwin in Space: Frontier Unlimited,

“[S]hould be enough without all the other reasons and rationalizations that have been presented. It is the proper motivation of a prideful people with vitality, a sense of destiny, and confidence in their own ability.”[6]

The prestige and pride rationale is salient across the programs of the world’s space powers, and especially so the Chinese human spaceflight program. Goodwin’s assertion is reflected in the 2011 and 2016 white papers laying out China’s purpose for space exploration, China’s ambition for space achievement is driven by a belief that the prestige benefits that result increase China’s national power, thereby enhancing China’s overall influence and giving China more freedom of action in a region where it seeks heightened hegemony. Moreover, the human spaceflight program is intended to demonstrate the strength and validity of the Chinese leadership to domestic constituencies:

“[A]s a single-party undemocratic state built upon the Chinese Communist Party’s legacy, the leadership seeks to tangibly demonstrate progress that resonates with the Party’s narrative of continual economic prosperity, scientific achievement, and national pride and unity so as to legitimize continued one-party rule… spaceflight is conducted to demonstrate that the Chinese Communist Party is the best provider of material benefits to the Chinese people and the best organization to propel China to its rightful place in world affairs.”[7]

The prestige rationale can also be seen in nearly every human spaceflight effort the United States has undertaken. As noted by Launius,

“The United States went to Moon for prestige purposes, but it also built the Space Shuttle and embarked on the space station for prestige purposes as well… [p]restige will ensure that no matter how difficult the challenges and overbearing the obstacles, the United States will continue to fly humans in space indefinitely.” [8]

Prestige and pride are powerful motivators, but are they alone enough to justify a robust human spaceflight program? Apparently not in the minds of policymakers, who weigh it against other indicators of national prestige – such as a strong national defense, global humanitarian presence, or leadership in the arts, athletics, or terrestrial sciences. Rather, it seems that the prestige and pride rationales for human spaceflight are most compelling in two cases: first, when there is a development in space that threatens the prestige of the nation and the pride its citizens hold in it. The Soviets beating the United States in orbiting the first satellite and astronaut, for example. Second, when a country sees space prestige as a method to complement and buttress a broader and pressing geopolitical goal. Such is the case for China, which actively seeks hegemony in the Asia-Pacific and hopes that its space program will demonstrate superiority over neighbors. Notably, at present both a distinct threat in space that threatens national prestige and a specific strategic goal that’s actively supported by leadership in space are lacking for the United States.

This relates to Professor Handburg’s notion of “compelling focusing events,” to which we return. Demonstrative of their importance, it is around these events which most of the enthusiastic narratives of human spaceflight have been built. The Apollo project, the Shuttle program, the International Space Station – these successes have been the product of fortuitous alignment of rationales put forward by domestic interests, the existence of external challenges those rationales were cohesive with, and political will to expend the necessary funds to achieve them.

Let’s explore these in turn. The Apollo landings, as perhaps the seminal series of events in the history of human spaceflight, have been ascribed with a slew of reasons for why they occurred: to promote peace “for all mankind,” to advance the technological and industrial capacity of the nation, to conduct scientific research and discovery. Yet despite President Kennedy’s rhetoric laying out these rationales, a singular reason existed for Apollo’s conception and drove its continued funding and execution. The United States had to beat the Soviet Union to the Moon. Without the broader context of the Cold War, Sputnik and Gagarin, the failed Bay of Pigs invasion and the Cuban Missile Crisis, the effort of landing on the Moon would not have begun or, if it did, the government would not have dedicated over four percent of annual GDP to achieve it. And, of course, once the “race to the Moon” was solidly won, subsequent missions lost political appeal and were accordingly cancelled.

Rationales put forth for the International Space Station include scientific research and international cooperation. It need be remembered, though, that the project evolved out of Space Station Freedom – a Reagan-era proposal for a U.S. station that faced stiff Congressional skepticism for reasons of funding and purpose. With the collapse of the Soviet Union, the Russian economic crisis and deorbit of Mir, and concerns about a diaspora among the Russian space industrial base, the United States brought in Russia and other international partners to the station project. In effect, the ISS found political support where Space Station Freedom failed for the prestige the project could achieve in a new global order, for the cost-sharing of a partnered effort, and in that it could serve to coopt Russian talent lest they went abroad to build space systems – or ICBMs. Now, with the ISS’ operational costs consuming a significant portion of NASA’s budget and its R&D output being less than expected and preferred (especially in the much anticipated field of space-produced pharmaceuticals), it is understandable that the agency, policymakers, and partners are noncommittal to extending its lifetime beyond 2024 or flying a follow-on platform after it deorbits.

The Space Shuttle was intended for cheap and routine access to space but, equally important, as a vehicle to deliver national security-critical Department of Defense payloads, conduct classified missions, and (perhaps) retrieve and return sensitive satellites from orbit. Much of the will to fund it came from DoD’s interest in the vehicle, which manifested in a complex set of design requirements. When the Shuttle failed to live up to the former purpose and was significantly scaled back for the latter following the Challenger disaster, the program was arguably left adrift in search of a mission – and eventually transformed into what was essentially a construction and delivery service for the ISS. It is not surprising then, even if disconcerting, that the program was ended without a follow-on capability in place.

And for these successes, there are others where an alignment of rationale and need didn’t exist outright; where the rationales put forward fell short of addressing an immediate national challenge, where the resources required couldn’t be justified when put against alternate projects. The whole of the Space Transportation System, the Space Exploration Initiative, the Vision for Space Exploration, and (to some) the “Journey to Mars” come to mind.

How so? The Space Transportation System, of which the Space Shuttle was envisioned as just an element of the larger architecture, found itself struggling for political buy-in and resources in the wake of the “victory” of the space race and competing with the rising pressures of the Vietnam War and domestic social change. The Nixon Administration could only justify a part of the program, the vehicle that had won DoD buy-in and had distinct national security purposes. Moreover, the decision to move ahead with the Shuttle was as political as it was motivated by some space-related rationale – Nixon didn’t want to be seen as the President who “killed the space program.”  The Space Exploration Initiative, with a total price-tag of over $500 billion, was balked at by Congress for its cost. The Vision for Space Exploration was cancelled because, as suggested by President Obama in so many words, the United States had “already been” to the Moon. And today’s “Journey to Mars,” with its significant schedule slippage, aborted asteroid redirect element, and currently unfunded cis-lunar “proving ground” phase, seems to be faring little better.

Several general points can be derived from these successful and failed human spaceflight projects of past. Foremost is an affirmation of the importance of the “compelling focusing event,” as described by Dr. Handburg, in providing the political impetus and will for starting and continuing support for a program to fly humans in space. The Apollo program and the International Space Station may have fulfilled important purposes – fostering international cooperation, demonstrating the United States’ leadership, enabling scientific discovery – but their inceptions were catalyzed as distinct policy responses to meet specific circumstances. Political consensus for these programs coalesced around the perceived national need to address the external challenges of the “space race” and of the Soviet Union’s collapse. These spaceflight programs won the support of broad enough political constituencies to be executed not merely because they involved outer space, but because they were seen as better tools to accomplish a strategic goal than terrestrial alternatives. As such, the substantial public funding and continued programmatic stability necessary for their success was provided until such time as the national need was met – but not much further after that. Other failed proposals, such as the Space Exploration Initiative or Vision for Space Exploration, would’ve equally fulfilled scientific, exploratory, and prestige purposes, but lacked a forcing function significant enough to warrant the creation of strong political coalitions that could bring them to fruition.

This leads to a second important point – that many of the rationales used to justify a human spaceflight program are either ancillary to the politically compelling purpose of meeting a perceived national crisis and geopolitical challenge, or are applied after the fact. Science, exploration, inspiration – these are, as described above, often-cited rationales that are almost inherent elements of any effort in space. Yet they have rarely, if ever, been the explicit and primary purpose of human spaceflight. As noted earlier, stakeholders of the spaceflight effort seek to justify that effort by the interests they hold – scientists desire discoveries and research, and justify programs for their scientific benefits; educators and politicians see spaceflight’s inspirational value as a method to bring students into science and technology, and therefore talk of the jobs created by it. But these alone are not enough to catalyze a new program. The groups who advance them do so in the hope that policymakers, of whom they’re constituents, will continue to support the space effort to advance their vested interest. As perhaps best said by Robert Colborn, “most of the motives advanced for [human spaceflight] seem… more like by-products than like major purposes.”[9]  This is not to minimize the value of these rationales, but to underscore their apparent unimportance in the creation of public policy pertaining to space.

Third, the rationales ascribed to a human spaceflight program generally lose importance or relevance once the program’s main purpose is complete and, accordingly, political will to sustain that program wanes. This is especially evident in the Apollo Program, where follow-on missions which carried scientific benefit were nonetheless cancelled after the United States clearly demonstrated its technical and scientific prowess. This again suggests the ancillary value of most rationales to the space effort and the significance instead of singular goals which a human spaceflight program seeks to achieve.

From these points, an overall conclusion can be drawn about human spaceflight, at least as a government-run effort. It is a tool to achieve immediate national needs toward which political consensus and will exists. More often than not, that consensus and will emerges from factors and circumstances in the broader domestic and international environment. Rationales justifying a program are equally valid and invalid depending on how they align with the necessity that program sets out to address, and may be reflected in political rhetoric regarding that program, but are generally not alone the driving force for its inception or even its sustainment.

Members of the space community should draw their own conclusions about their rationales from these points. Several suggestions, however, can be offered:

  • While obviously having value, the conventional rationales for human spaceflight are evidently not compelling enough to win sustained support from broad political coalitions or raise the stature of human spaceflight in the policy agenda. Continuing to justify human spaceflight on the basis of these rationales is unlikely to result in dramatic shifts in the space program. The space community either needs to find new rationales to justify spaceflight, or redefine the scope and character of those they currently do.

  • Rationales that are compelling must appeal to groups across the political spectrum, constituencies which exist outside the space community, and to policymakers who have immediate vested interests elsewhere. They cannot advocate for space for space’s sake alone; rather, they must advocate for space as a means to support society’s goals back on Earth. And to be successful against groups competing for the same limited resources, they must address current pressing needs (i.e. beat the Soviets to the Moon) rather than long-term or utopian ideals (i.e explore the unknown, settle outer space). The space community’s rationales need to be more focused, short-term, and relevant to the needs of people on Earth.

  • Rationales need an external forcing function, a “compelling focusing event,” to have relevance. The space community needs to be alert for perceived crises in the international and domestic arenas against which they can align their justification for spaceflight to drive the policy discussion. They must act as policy “entrepreneurs” by selling their rationales to policymakers as solutions to these problems.

Up to this point, this piece has discussed the relevance of rationales as they pertain to government-run efforts and public policy. But, again, the character of spaceflight is quickly changing with commercial and private actors entering the fold. Similar to the public sector, private companies are bounded in their desires by available resources. In the private sector, available resources are determined by market returns and capital investment. Where investors choose to allocate their funds is the result of risk calculation, market forecasting and, occasionally, personal motivation. Likewise, where customers choose to spend their money is a decision on the perceived value of the service they will receive.

This warrants a brief look at the economic rationale for human spaceflight. Despite the growing optimism among the enthusiastic public and the private sector about the economic rationale for human spaceflight, it remains to be seen whether a sustainable and profitable economic endeavor in space requires a human presence. Tourism, be it on suborbital spacecraft, circumlunar flights, or orbital platforms, is approaching, but is unlikely to be a robust market or catalyze a dramatic growth in human presence in space. Unless the cost of spaceflight is dramatically reduced, tourism will remain the province of a niche community of the ultra-wealthy. And while space companies such as United Launch Alliance and Blue Origin talk about “a cis-lunar economy” and “millions of humans living and working in space,” there is still no clear answer as to what economic activity those humans would be doing. In-space assembly, manufacturing, and production can be automated, as can lunar and asteroid resource mining. If the ISS is to be trusted as a case-study, basic and applied research conducted by humans is not a “killer application” for making money in space either.

Nonetheless, these companies will strive to see whether human spaceflight can be made economical. At present and into the near future, a variant of the economic rationale for human spaceflight could be seen as compelling, at least for the short-term: to see if a sustainable economic activity exists in space. Unlike the policy arena, whether this rationale remains compelling will not be judged against the needs of various constituencies and public interests, but by the market and the wallet.

This last point is important: by eliminating the pressure felt in the public sector to balance resources among competing groups, market-based human spaceflight enables other rationales to become more prominent so long as they are profitable. This is particularly true for the “utopian” rationale of human spaceflight – colonizing other worlds, ensuring humanity’s indefinite survival, creating new civilizations in the space frontier. This utopian rationale has always been an underlying assumption of human spaceflight, even if it has not been applied as a distinct or even ancillary goal of programs to date. Taken against the analysis above, this is understandable: it does not address an immediate and distinct national need, there is no forcing function for it (nor, hopefully, will there be, considering that would entail some sort of extinction-level event occurring), and it is difficult to see how it would win support from powerful political groups who worry more about first improving the plight of Earth and those living upon it. Nonetheless, per Launius,

While “[t]he quest for utopia in space has been implicit rather than explicit, there has never been any question but that the long-term objective of spaceflight is human colonization of the cosmos. Virtually all models for the future of spaceflight have at their core human expansion beyond Earth.”[10]

This motivation underlies the plans of Jeff Bezos and Elon Musk, who indeed see the economic rationale of their companies’ plans as a means toward this end. While the latter rationale will necessarily rely upon the success of the former, it is easy to see an undeniable paradigm shift occurring with the embrace of the utopian rationale as a key purpose for human spaceflight.

Taken all together, what does this mean for human spaceflight in the coming decades? It is, of course, impossible to foresee the future, but some predictions may be made. Unless the United States faces a geopolitical crisis which warrants a space solution or develops a national grand strategy which cohesively integrates human spaceflight as a valued tool to achieve its aims, it is unlikely that the stature or funding of its human spaceflight program will increase. Large spaceflight programs with enormous costs will likely continue to face significant fiscal pressures and programmatic instability as space policy ascends and descends in political importance. Even if, as said by Launius, prestige and pride will ensure that the United States continues to fly humans in space, they will not guarantee a robust human presence or the success of an ambitious program; there is no reason to assume so, if they haven’t done so historically. Meanwhile, other “rising” countries with distinct geopolitical goals, such as China, will continue to exploit the prestige factor of human spaceflight until such time as their aspired position is obtained.

However, if the private sector succeeds in its economic aspirations, human spaceflight may become more prevalent and the rationales used to justify it more varied. There could emerge a successful synthesis of the private sector’s aspirations, justifications, and capabilities with the civil space program’s goals and needs – humans in space because of economic reasons working to support a government-directed program, for example. This synthesis of capabilities and rationales may be key to the 21st century’s “full realization of both the practical and inspirational potentials of space.”

Does all of this answer the question of “why?” No. Perhaps this is because it is a question without a single, compelling, definitive answer. The answer to “why” will change according to circumstances, politics, and economics. Ultimately, it comes down to spaceflight’s value as a tool – to be used when the time and circumstances are right. And still, attempting to find a definitive and singular answer for it will likely remain as popular an activity as it is an ultimately futile one. What would be more fruitful for the space community is not continuing to seek them out; instead, it would be find a way to adjust the answers which do exist in such a way to make them compelling when the country needs to address a crisis and is asking “how.”

[1] John M. Logsdon, “Which Direction in Space?” Space Policy, May 2005. Pg. 88.

[2] Ibid. Pg. 87.

[3] Ibid.

[4] Roger D. Launius, “Compelling Rationales for Spaceflight: History and the Search for Relevance” in Steven Dick and Roger Launius, Eds. Critical Issues in the History of Spaceflight (2006). Pg. 68.

[5] Roger Handberg, “Rationales of the Space Program” in Eligar Sadeh, Space Politics and Policy, 2002. Pg. 29.

[6] Harold Leland Goodwin, “Space: Frontier Unlimited” (1962). Pg. 111.

[7] Cody Knipfer, “The Asian Space Race and China’s Solar System Exploration, Domestic and International Rationales,” The Space Review, 2016. http://www.thespacereview.com/article/3007/1

[8] Roger D. Launius, “Compelling Rationales for Spaceflight: History and the Search for Relevance” in Steven Dick and Roger Launius, Eds. Critical Issues in the History of Spaceflight (2006). Pg. 50, 52.

[9] Robert Colborn, “In Our Opinion,” International Science and Technology, January 1963. pg. 19.

[10] Roger D. Launius, “Compelling Rationales for Spaceflight: History and the Search for Relevance” in Steven Dick and Roger Launius, Eds. Critical Issues in the History of Spaceflight (2006). Pg. 43.

Page 2 of 61

Powered by WordPress & Theme by Anders Norén