Writing Policy Recommendations for Academic Journals: A Guide for the Perplexed

Climate change. The rise of China and the U.S. response. The dangers posed by ISIS and other terrorist groups. The ability of the United Nations, the IAEA, the NPT, and other institutions to manage nuclear proliferation. Civil unrest and the potential for peaceful change. Globalization's benefits and perils. These are among the most important security issues facing the world today—and they are issues that scholars can, and do, speak to regularly. Thomas Homer-Dixon's “Environmental Scarcities and Violent Conflict,” John Mearsheimer's “The False Promise of International Institutions,” Andrew Kydd and Barbara Walter's “The Strategies of Terrorism,” Maria Stephan and Erica Chenoweth's “Why Civil Resistance Works,” and Henry Farrell and Abraham Newman's “Weaponized Interdependence” are among the most-cited articles ever published in International Security, and their scholarly impact is considerable.¹ Beyond their theoretical contributions, these articles illustrate the potential that International Security and other academic journals have for speaking to policy issues. All of them contain important ideas proposing what decision-makers should consider, do, and not do to make the world a better place.

Despite the potential importance of academic work to the policy debate, new scholars receive little training on why and how to make policy recommendations. Some academic journals only pay lip service to policymaking or ignore it altogether. There are many insightful works on the gap between policymakers and academics and why it should be bridged, and a host of training workshops, government funding programs, and other efforts push in this direction, often with valuable results.² Other articles stress how to craft relevant recommendations in general.³ Yet almost all these efforts focus on activities outside publishing in top academic journals, ignoring the important role that scholarly journals should play in shaping thinking on policy. This article seeks to fill this gap, advising contributors on how to write policy recommendations for articles in both International Security and, I hope, other high-quality academic publications that seek to inform the policy debate.

Providing helpful commentary on policy is challenging—as difficult as the academic research that leads to publication in a top journal—and it is doubly difficult when doing so for an academic journal. Publication time is measured in months or even years, in contrast to blogs and current affairs journals that offer more immediate turnaround. Policymakers rarely have time to read long articles, which are the staples of journals like International Security, and indeed “rarely have time to read what's not urgent in their inbox,” as one senior policymaker noted.⁴ Perhaps most important, it is often difficult for academics to understand a policy, let alone the pressures that policymakers face and the conflicting objectives that they try to juggle.

Yet it is vital for scholars to learn the dilemmas that policymakers face and to be relevant to public and private policy debates so that their research can help make the world a better place. Engaging policy also makes for better research. By focusing on questions important to the policy world, scholars avoid the trap of scholasticism—that is, when they concentrate on internal debates rather than on the original problems that first inspired academic research. But writing serious policy recommendations requires considerable modesty: crafting effective policy is hard, and academics should recognize the limits of their findings and the difficulties of moving the policy needle.

When crafting recommendations, scholars should take advantage of their objectivity and ability to challenge the prevailing wisdom. They are also well-placed to use history to learn lessons and to draw insights from large datasets. Academic journals, for their part, endure: they have long shelf lives and allow deep dives, providing space for scholars to present heavily researched empirical evidence, theories, and analyses. Long, well-researched articles can, over time, shape the broader narrative for how to think about a complex issue such as the potentially peaceful nature of democracy or how to make deterrence more robust.⁵ They may also help provide context when unexpected events occur and the policy community has little to draw on but theory and analogy.

As scholars begin crafting their articles, they should try to determine their policy audience—including leaders outside government in industry and civil society—to identify who might read their work and who might act on it. They should consider their variables, identifying ways to maximize better outcomes or minimize worse ones. To help academics think of ways to use their findings to influence policy, one strategy is to create or draw on existing menus of policy instruments (“What can diplomats do?” “How might financial tools assist with coercion or another goal?” and so on). At the same time, scholars should consider the costs and limits of their arguments, conveying enabling conditions and the level of certainty of their findings as well as their overall recommendations. Finally, they should use their academic journal work as a springboard for writing shorter pieces in policy journals, blogs, and opinion pages.

The remainder of this article has five sections. The first section explains why academic authors should speak to policy issues and why offering policy recommendations contributes to better scholarship. It also highlights the unique role of academic publications like International Security. The second section describes the dilemmas that policymakers themselves face—difficulties that scholars must recognize even if their ultimate advice criticizes or transcends these dilemmas. In section three, the heart of the article, I explain how an academic author might begin thinking about policy relevance. For some scholars, this process may involve identifying steps to take from the start of their research project. For others, it may involve considering how work undertaken with an academic audience in mind might also speak to policymakers. Section four poses a set of questions that scholars should consider as they craft their recommendations. It also illustrates how to apply the article's framework. The article concludes with a brief exhortation for scholars to engage the policy world in their academic research.

Policy-relevant scholarship is intended to produce findings that feature in the deliberations of government officials and others involved in policy decisions. This section makes three arguments. First, journals like International Security value policy relevance, as do most of the scholars who work on international security, and policy-relevant research can improve scholarship as well as inform policy. Second, academics have much to contribute to the broader policy debate. Third, academic journals have their own niche in the broader policy analysis ecosystem, complementing blogs, the opinion pages of major newspapers, and policy journals like Foreign Affairs.

Not every International Security article should be policy relevant, but most should. Scholarship seeks to expand human knowledge, but for international security, much of that knowledge is interwoven with policy challenges.⁶ The vast majority of articles that appear in International Security speak to some aspect of policy, ranging from avoiding international and civil wars to improving alliances to the nature of the international system.⁷

The editors of International Security consider the “so what” hurdle when they evaluate a submission—Why should a reader bother with your article when there are so many other ones to read instead?⁸ This hurdle is much easier to clear if the author makes the policy connection unambiguous. If an article's central question matters to those responsible for waging war, preserving the peace, maintaining fiscal stability, improving governance, and otherwise trying to promote security, then that article—and its recommendations for avoiding dangerous outcomes and increasing the chances of positive ones—is worth considering for inclusion in a journal.

Aside from this important question regarding publication, many scholars seek to do relevant research.⁹ It is likely that a policy question or world event piqued a scholar's interest in international relations. Many scholars believe (rightly!) that they can contribute to both internal government debates and broader, more public discussions of complex policy issues.

Scholars have much to offer the policy debate. Although scholars may be removed from the policymaking fray, that distance gives them a chance to present new ways of thinking about a problem and to take the long view. Unlike some policymakers, scholars are not driven by their inbox. This flexibility allows scholars to set long-term agendas. As the policy community celebrated the fall of the Soviet Union, for example, Graham Allison, Owen Cote, Richard Falkenrath, and Steven Miller presciently identified the threat of loose nuclear weapons and material from the former Soviet Union. Their evidence spurred policy attention and eventually action.¹⁰

In the near term, scholars can challenge prevailing wisdoms. In 2002, John Mueller questioned the post-9/11 consensus that Al Qaeda would continue to conduct numerous mass-casualty attacks like 9/11 or even more destructive ones. As time went on, Mueller built on his research and proved his initial argument that 9/11 was an outlier for U.S. casualties.¹¹

Scholars have time to dig deep: research for an article can take many years, a luxury the policy community lacks. Scholars can also create large datasets and survey significant amounts of open-source material. Some of this analysis occurs within the intelligence community, but scholars often create better-designed datasets and have more time to read and digest publicly available material.

In addition, scholars are bureaucratically (though not politically) neutral.¹² They do not have a vested interest in whether the State Department or the Defense Department oversees a peacekeeping operation, for example. Policymakers often reason by analogy, whereas scholars are trained to thoroughly research particular cases in an unbiased way and draw conclusions across cases, both of which add more insight than using a simple analogy.¹³ Perhaps most important, International Security and other leading journals publish rigorous, peer-reviewed articles that use precise and careful research methods and analyses to answer questions, challenge the conventional wisdom, validate empirical findings, and advance understanding about complex topics.

This reasoning may seem Pollyannaish, but imagine if scholars rejected policy contributions in their writing. There would be less work that is deeply informed, methodologically rigorous, and carefully reviewed. Daniel Drezner points out that many nonacademic public intellectuals are more partisan and less open to criticism than their academic counterparts. Such partisanship decreases the quality of public intellectuals’ work though not their influence.¹⁴

More focus on policy can also lead to better scholarship. Making policy is difficult, and making good policy is even harder. By speaking to these challenges, scholars are forced to ask themselves knotty questions and to better understand what they study and the data on which they rely for their analyses. For example, an academic who engages policy seriously may recognize that the paper trail of memoranda and strategy documents is often more spin than substance.¹⁵ With policy concerns in mind, scholars are less likely to emphasize elegant scholarship that elides real-world difficulties. As Rebecca Adler-Nissen argues, “Part of the reason why ‘bad ideas’ are allowed to develop in the first place is that we have established a hierarchy of prestige that values ‘clean’ and ‘elegant’ scholarly ideas over the ‘messy’ ones of practitioners.”¹⁶ She adds that policy engagement forces scholars to “begin to develop ideas that acknowledge the complexities, paradoxes, and hidden politics of ‘policy.’ We become curious about what practitioners find appropriate, shameful, or important.”¹⁷

Policy recommendations can be dangerous things. Policymakers might take them out of context to “prove” that their desired outcome is the best option. Work on the democratic peace, for example, was used to justify the 2003 invasion of Iraq. But this justification overlooked important context: many findings that extol the benefits of the peace do not focus on forced regime change.¹⁸ Paul Musgrave warns of “lab leaks” in political science, whereby magazines like the Atlantic and Foreign Policy present concepts with little context and fewer caveats, making them more likely to be misused.¹⁹ Scholars may also fear being labeled as activists if they promote particular policies. They are also taught to avoid normative language—a luxury that policymakers do not share.

The cures for these potential ills, however, are not to avoid policy engagement but to take it more seriously. Scholars cannot control the conclusions that people draw from reading their work. But clear writing and speaking directly to policy concerns make it more difficult for others to misuse a scholar's ideas for their misguided policies. As Erica De Bruin argues, “If irresponsible public scholarship is the issue, then developing a more rigorous ethic of public engagement is vital.”²⁰ Similarly, activism is a good thing when it reduces the risk of war, improves human rights, and otherwise makes the world a better place. In addition, as Charli Carpenter points out, engaging with advocacy organizations often helps researchers overcome academic biases because it requires them to consider different issues and learn about new problems.²¹ The key for scholars is to ensure that their work remains rigorous and objective, which at times means recognizing that their preferred approach has flaws and limits.

Academic journals like International Security are part of a vast ecosystem of outlets that all claim to speak to policy issues. These include foreign-policy-oriented long-form journals like Foreign Affairs and Survival, general media outlets such as the New York Times, Yomiuri Shimbun, and the Economist, podcasts like those featured at War on the Rocks, specialized outlets like Arms Control Wonk and Lawfare, and numerous others. As articles on policy relevance argue, scholars should publish in these outlets—and many do! These platforms publish articles more quickly than academic ones, enabling scholars to speak directly to the issues of the day.

Academic journals have their niche as well. In some cases, particularly when the danger is possible but not imminent (e.g., if China were to invade Taiwan), an academic journal might be an ideal outlet for in-depth work. Michael O'Hanlon wrote such a piece in 2000, and its findings remained relevant for years even as the international environment and China's military capabilities changed.²² Scholars can try to provide general guidance on a more specific problem. A 2020 piece explaining the sources of Russian bellicosity, for example, would still be helpful to policymakers in 2022, though it would need to be updated with insights on the invasion of Ukraine.

The longer length of an academic article also has trade-offs. A typical International Security article has 10,000–15,000 words, and some are even longer. For Foreign Affairs, the recommended length is 2,000–5,000 words, and for the New York Times it is 800–1,200 words.²³ Shorter pieces are more likely to be read, especially by more senior policymakers with crammed schedules. Conversely, it is harder to go into depth in a short piece.

The flip side of a long review process and longer length is a long shelf life.²⁴ Articles for International Security deliberately speak to broader issues and long-term problems. Consequently, they may remain relevant for many years. Twenty years after publication, Andrew Kydd and Barbara Walter's work on the strategies of terrorism still speaks to terrorist groups’ goals and methods.²⁵ Similarly, Caitlin Talmadge's focused study on Iran's possible blocking of the Strait of Hormuz remains highly relevant almost fifteen years later, identifying the many challenges to Iran, the possible responses for the United States, and so on.²⁶ Additional scholarly work, technological advances, and geopolitical change may affect the issues discussed in both articles, but the authors provide a set of concepts that create a valuable foundation on which to build policy.

In addition, the long shelf life changes the readership and value of the piece, allowing scholars to shape worldviews that inform a host of decisions. An International Security article may be on a syllabus for years or even decades after publication, framing how an issue is thought of for years to come, as RAND research on nuclear strategy did in the 1950s and 1960s.²⁷ It is plausible that a twenty-year-old student who reads an academic piece may retain its concepts and arguments as a mid-level policymaker two decades later. In the words of John Maynard Keynes, “practical men, who believe themselves quite exempt from any intellectual influence, are usually the slaves of some defunct economist.”²⁸ Be that defunct economist.

Journals like International Security are also more open to historical deep dives and reinterpretations. Because analogical reasoning is so prevalent, changing an understanding of an important historical event, such as the outbreak of World War I or the Cuban missile crisis, can inform how to think about what to do in the present. Keir Lieber points out that historians and policymakers like Henry Kissinger misunderstood World War I as an inadvertent conflict. This misreading has profound implications for how to gauge the likelihood of small great power disagreements accidentally spiraling into major war.²⁹

In some cases, rare ones but with high impact, a discontinuous event takes an article off the shelf. Should China invade Taiwan in 2025, an in-depth piece written in 2020 in a journal like International Security might be among the most comprehensive external guides to a policy challenge that would consume world attention, even if it does not discuss this specific crisis. Naazneen Barma and James Goldgeier note that Swedish Nobel laureate Gunnar Myrdal's The American Dilemma, a masterful 1944 study of race in the United States, was largely ignored when published, and its sponsors disowned it for years. In 1956, however, its findings helped shape the Brown v. Board of Education decision, one of the most consequential Supreme Court rulings in U.S. history.³⁰

Because academic articles are often explicitly theoretical, they also offer insights into new or related but distinct situations. For example, the 2011 Arab Spring upended long-standing policies toward area regimes. Even though pre-2011 articles on democratic transitions, the impact of military coup-proofing, civil war resolution, and similar topics that are common in security-related academic journals do not necessarily focus on the conditions in Egypt, Tunisia, or other affected countries, they nevertheless offered many potential insights during the turbulence.³¹ Policymakers might have learned ways to promote successful elections, avoid (or predict) coups, prepare for civil violence, and so on. The key concepts endure, even if the dates and places are different.

Bureaucracies often amass considerable knowledge on complex issues, and policymakers are rarely stupid. Yet many policies seem foolish and ill-informed, especially in hindsight. To understand why the policy outcome often differs from the ideal, it is vital to understand the many pressures and restrictions that policymakers face. Academics should not treat these difficulties as immutable or even excusable, but recommendations should reflect an understanding of the policy itself and how to improve outcomes. Identifying the actual policy is a difficult aspect of research that should accompany the broader academic research process. Perhaps most important, academics should approach policy influence with considerable humility: in providing advice to smart, knowledgeable people, often the academic is not aware of all, or even most, of the challenges confronting the policy community.

Policymakers make their own policies, but they do not make them just as they please. Some factors, such as geography or the polarity of the international system, are invariable. Although many elements that constrain decision-making and agency are malleable, these constraints are often tight, making it hard for policymakers to break out of a narrow set of options.

Policymakers are beholden to their publics and to elites. These limits apply in both democratic and authoritarian systems, albeit in different ways and to different degrees.³² It is tempting to urge “leadership” as a recommendation, and at times policymakers do go against the preferences of their constituents and supporters. But leaders are understandably wary of jeopardizing their political status.

Competing priorities and limited resources also constrain policymakers. The many issues that scholars examine—civil wars in sub-Saharan Africa, refugee flows in Southeast Asia, the destabilizing effects of new weapons systems, and problems with security assistance, to name only a few—compete with one another and with numerous other concerns. For the most senior policymakers, they also compete with domestic priorities, which are usually more salient. Recommendations that call for more aid to a region, more training to an ally's military, and so on all come at a cost, with other priorities receiving less money and attention as a result.

Policymakers also must act with only limited information. Colin Powell recalled that if he waited for enough facts to be 100 percent right, it meant it was too late to act.³³ Although U.S. intelligence proved remarkably prescient about the likelihood of the 2022 Russian invasion of Ukraine, the Joe Biden administration did not know if Ukrainian forces would hold out, how key allies in Europe would respond, or how a then-unpopular leader like Ukrainian President Volodymyr Zelensky would respond when greatness was thrust upon him.³⁴ Policymakers also had to worry about less probable but potentially catastrophic concerns like nuclear escalation. Recommendations must recognize the many uncertainties and scenarios that policymakers are likely to face after a terrorist attack, as a civil war is breaking out, or when a peaceful movement seeks to overthrow a dictatorial regime.

In short, policymakers are in a proverbial box, constrained by politics, resources, competing priorities, and limited information. For academics to think outside this box, they must understand why policymakers are in it, which parameters are possible to shift, and which are likely to hold firm.

Understanding policy takes time, and it is easy to get wrong or caricature. First, policies often embody multiple strategic, bureaucratic, and political interests. Second, an administration may provide conflicting or confusing signals as to what its true policy is. Third, the resulting complexity makes coding difficult. Fourth, policies are often bad because alternative policies are worse; criticism should recognize this reality.

One barrier to understanding policy is that a single issue may be interwoven with a wide array of interests. Take, for example, negotiations with Iran as embodied in the Joint Comprehensive Plan of Action. In this instance, U.S. policymakers balance a range of goals, including: stopping Iran's nuclear weapons program altogether; pausing the program for several years; shoring up international regimes like the 1968 Treaty on the Non-Proliferation of Nuclear Weapons (NPT); leading allies in Europe and elsewhere that also oppose Iran's nuclear program but are more eager for commercial ties to Tehran; reassuring regional allies that are suspicious of Iran; condemning Iran's support for militant and terrorist groups in the greater Middle East; supporting Iranian demonstrators seeking regime change; and winning over a domestic population that is highly suspicious of any relationship perceived as forgiving to Iran.³⁵ Judging success is difficult, as some of these goals are incompatible. Policymakers maximize some interests, satisfice others, and “fail” on still others.³⁶ A recommendation that improves the odds for success in one area may hinder it in others. For example, the Barack Obama administration succeeded in nuclear negotiations with Iran in part because U.S. policymakers avoided entangling the nuclear discussions with demands regarding Iran's support for terrorist groups.

Multiple signals from an administration make determining the actual policy difficult. Governments issue public strategy documents but engage in private and even covert diplomacy that can be more consequential. Diplomats may join treaty negotiations but have private instructions to ensure that negotiations fail. Officials use public statements not only to delineate a policy but also to close off rival approaches. For example, an administration publicly condemning the assassination of Saudi journalist Jamal Khashoggi on one day would find it more difficult to sell Riyadh weapons the next. At times, the public statement is virtue signaling, staking the moral high ground even as most administration policies either do not follow through on lofty goals such as promoting human rights or opposing aggression or even go in the opposite direction. Western leaders, for example, called for Syrian dictator Bashar al-Assad's ouster but did not provide the Syrian opposition with sufficient military support to make it happen.³⁷ Was regime change truly their goal?

The many interests involved in policy decisions, the multiplicity of signals coming out of governments, and the possibility of virtue signaling all pose coding problems. It is challenging for scholars to determine success or failure or to assign values to other binary measures of policy that are often used in large datasets. Similarly, the signaling confusion makes it hard to for researchers to know which coding they should use.

For academics, the many interests and confusing signaling also pose a data problem. Official strategy documents can be great sources, but they can also be misleading. President Donald Trump's 2017 National Security Strategy seemed to bear little relationship to his administration's overall foreign policy. Indeed, in some ways actual policy contradicted the guidance, with the strategy embracing a strong role for U.S. leadership in the face of great power competition when President Trump was highly critical of traditional U.S. allies and was cozying up to Moscow.³⁸ Often deliberations occur in private before a formal meeting. The meeting record is thus a ratification, not a reflection of discourse. Understandably, scholars often have a bias toward the written word, whereas it is the briefing or private conversation that matters the most to many policymakers. Some written policy documents may accurately reflect the authors’ views. But policymakers are less likely to document their political and bureaucratic interests, which leads scholars to have a bias toward strategic explanations.

Indeed, almost every policy addressing a complex problem is insufficient and often the only alternatives are bad ones—so which bad one is the best? As David Baldwin notes in his assessment of economic sanctions, “If the menu of choice includes only the options of sinking or swimming, the observation that swimming is a ‘notoriously poor’ way to get from one place to another is not very helpful. And if the principal alternative to economic sanctions is appearing to condone communism, racism, terrorism, or genocide, the observation that they are a ‘notoriously poor tool of statecraft’ may miss the point. In the context of the logic of choice, the evaluation of one policy alternative in isolation from others makes little sense.”³⁹ Policymakers may have an ambitious declared goal (“stopping human rights abuses”), but in practice they may settle for a range of lesser achievements, such as slightly reducing human rights abuses by making it more difficult for a regime to access resources, signaling disapproval to gain allied support, avoiding pressure to use military force that may backfire, and so on. Recommendations that do not at least acknowledge the poor range of options available to policymakers will not be convincing.

Taylor Fravel and Charles Glaser's work on the South China Sea and U.S. policy is an excellent example of how scholars can avoid these traps.⁴⁰ Fravel and Glaser are careful not to caricature current policy as they describe alternatives such as greater retrenchment and more intense military resistance. They detail the conditions under which different alternatives might be appropriate and suggest specific policies to accompany the alternatives, such as clarifying ambiguous treaty arrangements, imposing substantial economic sanctions, implementing shaming measures when China violates norms, and deploying surface and air forces, among many others. Overall, the reader is left with a better understanding of the balance that current U.S. policy is trying to strike, which Chinese actions would suggest the policy is failing, and the many potential downsides of different approaches, particularly how more aggressive efforts risk unwanted escalation.

Scholars must understand the policymakers’ dilemma and factor it into their recommendations, but they should not be bound by it. Pointing out the weakness of a policy in addressing a problem is an important service, and strong arguments can help policymakers advocate for more resources, shift priorities, or even take political risks in the face of domestic and elite opinion. Even better, however, is offering a plausible alternative. A particularly important role for scholars is to help policymakers reconsider factors that the policy community sees as insurmountable rather than malleable, such as identifying ways to overcome long-standing animosities, to reshape public support, to reprioritize regional objectives, and so on.

Scholars are taught to dissect complex problems, but often they focus on a lacuna in the literature rather than the policy implications of their research findings.⁴¹ If they focus more on the advantages of a particular method or on why a variant of one major paradigm is better than another, such elements by themselves are of little interest to the policymaker. Below I both offer advice for authors as they begin to craft a piece designed to increase policy influence and list factors to consider as their research progresses. Some of these steps may prove useful even if the scholar sees the research as primarily academic.

For those interested in speaking directly to policymakers as well as academics, start with the “so what” that motivated the research in the first place to determine how it fits into the policy world. What factors shape current policies, how malleable are they, and what alternatives are on the table? What will policymakers learn from reading the article that may improve their understanding about the problems they face and that may offer potential solutions? The article's focus should help solve a problem that is in a policymaker's inbox or, just as important, should be in their inbox. As the then deputy secretary of state James Steinberg noted, policymakers are “desperate” for ideas and solutions.⁴²

Before scholars begin their research, they should examine the issues that different kinds of policymakers or policy influencers see as important. Make sure to include those topics in the essay in a way that their audience can recognize, understand, and appreciate. Some policymakers write strategic documents, others evaluate weapons systems, still others negotiate treaties, and so on. Sometimes scholars decide that their work will have the greatest impact if they inform the media. The research findings should help reporters enlighten the public and ask nuanced and informed questions of government officials. Likewise, if the research will have the greatest impact by shaping the thinking of undergraduates and masters’ candidates—the next generation of policymakers—consider how to structure the whole essay, and possibly the research more broadly, to be most useful and effective for that audience.

With the issue in mind, identify the target policy audience. For those writing on a common topic like alliances and war, North Atlantic Treaty Organization (NATO) countries’ ministries of defense and foreign affairs, the leaders of Asian democracies, and intelligence chiefs are some (ambitious) possibilities. But authors should think broadly, beyond just governments, even if their piece is focused on traditional interstate security issues. For example, social media companies have emerged as important players in the information realm. With three billion monthly active users, Facebook's decisions on who to allow on its platform and what can be discussed are often more consequential than various foreign ministries’ statements. Companies’ content moderation policies and crisis response protocols, or lack thereof, can hinder or enable genocide.⁴³ Schools can teach new subjects (or old subjects differently), affecting reconciliation between once-embittered communities and public attitudes toward age-old problems like the impact of discrimination. Civil society and advocacy organizations such as the International Campaign to Ban Landmines can shift discourse, rally domestic coalitions (remember the box and how domestic politics shapes it!), and advance international law. Also consider where on the policy food chain your audience is. The U.S. president can change things more quickly than the assistant secretary of state for the Bureau of International Security and Nonproliferation, and the assistant secretary has more power than a desk officer. But guess which one has more time to read your article?

Authors who want to speak directly to policymakers should think of ways to infuse their entire project—puzzle, theoretical approach, hypothesis testing, and presentation of findings—with answers to both scholarly and policy questions. In addition to presenting rigorous methods and building on existing literatures, an article's theoretical and hypothesis sections must also make sense to a curious policymaker. Why is comparing many states’ approaches to military training better than delving deep into one example (or vice versa)? Why present some explanations but not others? A policymaker working on this issue should nod her head as she goes along rather than wonder why vital, obvious details that are necessary to make progress on a problem are missing from an article's analysis.

Scholars can also emphasize certain variables in their analyses to identify ways that policymakers can achieve the best results. If civil wars are correlated with ethnic disputes, economic inequality, or poor governance, then policies that foster ethnic harmony, greater equality, and reduced corruption should be encouraged. Policymakers would eagerly listen to scholars who provide specifics on what has worked. Articles that focus on methods can also be useful to policymakers, though drawing policy insights from such works often requires a bit more effort. Nicolas Sambanis, for example, argues that changes to both the threshold of violence used to define a civil war and the coding of intrastate, interstate, and extrastate wars can dramatically alter findings regarding peace duration and the causal power of economic problems, among many others.⁴⁴ If articles that draw on such findings are not robust, as Sambanis's research suggests, then policy recommendations built on them should be reconsidered.

Some scholars may prefer to skip the above steps and instead focus on the article's contributions to the academic literature. If they do so, however, they can still write useful policy recommendations. Having read a scholar's research, the intended current or future policymaker may be more informed about the causes of war, why alliances fracture, barriers to ethnic reconciliation, and other grave problems and perhaps better equipped to reduce these dangers. The author should now ask, “Given the findings from my research, what makes desirable policy outcomes more likely?” This might involve pointing out tensions between different policy approaches. Lindsey O'Rourke, for example, finds that policymakers prefer covert regime change over overt measures because they can deny responsibility for failures and reduce criticism for meddling in general. Efforts to ensure deniability, however, make the operations less likely to succeed. Highlighting this trade-off between two competing benefits (deniability versus improved chances of success) is important and increases the article's utility to policymakers.⁴⁵ This step tends to be easier if scholars have designed their work with policy concerns in mind from the start. Even so, articles that are more academic facing may still make desirable policy outcomes more likely.

Scholars’ recommendations should flow from their analyses. Such a point seems straightforward, even obvious. But often policy recommendations stray from the analytic foundations on which they claim to rest. For example, the 9/11 Commission issued a powerful report condemning intelligence failures and calling for major structural reforms, particularly to centralize intelligence under a director of national intelligence.⁴⁶ As Richard Posner pointed out, however, among intelligence agencies it was only the Federal Bureau of Investigation (FBI) that demonstrated major structural failures in combating terrorism.⁴⁷ The 9/11 Commission nevertheless recommended major institutional changes elsewhere in the intelligence community but not in the FBI. Although the report indicated that post-9/11 information sharing worked well and that Central Intelligence Agency (CIA) paramilitary operations were effective and agile, it recommended centralizing information sharing and moving the CIA's operations under the Defense Department. To be clear, such recommendations may have been good ones, but they did not match the substance of the commission's findings on information sharing or paramilitary operations.

Finally, avoid offering hackneyed recommendations. A recommendation that urges policymakers to focus on economic growth may be vacuous. A call for more intelligence may be hard to enact. Instead, provide thoughts on why intelligence is currently lacking.⁴⁸

Depending on the research topic, there are many practicable ways to influence policy. It is useful to consider all options, even if most do not end up being relevant to your findings. The U.S. military teaches its students that the instruments of power are summarized by the acronym DIME (diplomatic, informational, military, and economic)—and that is one place to start.⁴⁹ What might a diplomat do to sail the ship of state in the right direction? Is more intelligence needed? If so, could government education or propaganda campaigns help? What about the many variations of military force? Do economic tools such as sanctions help? Sometimes the answer to such questions is a clear and quick “no.” But combining these tools can often move policy forward. There are numerous variations of DIME, such as MIDFIELD (military, informational, diplomatic, financial, intelligence, economic, law, and development), which brings in financial and economic tools, international and domestic law, and so on.⁵⁰ In other cases, changes to education policy might be appropriate. Regardless of the preferred abbreviation, if any, thinking through a list of tools is a useful way to start.

Combinations are particularly important. Policymakers rarely rely on one instrument, and saying that such an instrument succeeds or fails, by itself, is less persuasive than discussing combinations.

Another approach is to think of a checklist for policymakers. Alexander George's writing on coercive diplomacy, for example, offers both contingent generalizations on when it works and factors for policymakers to think through. In essence, George shows how structured, focused comparisons of past cases help policymakers assess what they need to know and do for current developments.⁵¹

Some policy recommendations may be highly effective but also involve high costs in lives or money. Others help solve one problem but introduce others. Considering the findings, what current policies are making things worse or simply wasting money? Similarly, what costs and trade-offs are likely if a policymaker implements the article's recommendations? In general, it is always useful to ask, “Why is this not being done already?”

The findings that promoting regime change in Iran would lead to many casualties and would foster anti-Americanism would be useful to share with policymakers. Yet a recommendation to avoid foreign intervention might mean accepting a hostile, nuclear-armed Iran. It is easier for policymakers to dismiss research that fails to consider the latter possibility. Other policies are just expensive. Asking Asian countries to respond to China's rise by vastly increasing their anti-access/area denial capabilities may be sensible, but doing so is costly. Political leaders have other uses for the money.

By contrast, examining the cost of existing policies can generate new recommendations. For example, Kenneth Pollack finds that U.S. efforts to train Arab militaries using a U.S. military model is a recipe for failure given different political, cultural, and institutional settings. This seems like a finding that would lead to a recommendation to stop training altogether. Even though ineffective training is often useless in a military sense, stopping it would anger allied elites and harm bilateral relationships. Pollack thus recommends that the United States shift how it trains foreign militaries to better recognize these differences.⁵²

With the above in mind, make recommendations that are distinctive and clear. In her work on military training, Renanah Miles Joyce contends, “Liberal providers should emphasize building institutions that help to regulate military behavior rather than prioritizing individual or unit-level training with a normative component tacked on.”⁵³ This sentence packs a lot of substance: it identifies the actors (liberal states that provide military assistance), the policy that needs to change (prioritizing individual and unit-level training), and the proposed alternative (building institutions).

Another approach is to think about the policy box and where you stand in relation to it. It is tempting for scholars to ignore politics and simply point out the best answers. Yet recommendations that incorporate political realities are potentially more influential. Often, a mix is best. A scholar might note that today's political reality makes the ideal policy infeasible. Instead, the author might recommend a suboptimal but still useful set of steps: “As long as U.S. domestic politics makes a return to the Trans-Pacific Partnership [TPP] difficult, a less effective but valuable step would be to engage in a series of bilateral trade agreements that, cumulatively, offer lesser but still important security benefits similar to TPP.” This recommendation acknowledges the preferred solution (TPP) and offers a more politically plausible middle ground (bilateral agreements) that a policymaker could consider. Similarly, it is important to acknowledge resource constraints but not be bound by them: “Ideally, Taiwan would purchase a suite of anti-access/area denial capabilities rather than rely on more traditional systems like tanks, and it should begin with Harpoon anti-ship missiles.”⁵⁴

After conducting exhaustive research and developing informed recommendations, a scholar can repackage their research for podcasts and as shorter pieces for magazines like Foreign Affairs, outlets like Lawfare, and newspapers. These shorter pieces come to the attention of policymakers at multiple levels and make it more likely that at least a few of them may engage the longer work. After Keir Lieber and Daryl Press published their deeply researched findings on challenges to nuclear deterrence in International Security, they reached broader audiences by sharing their results in Foreign Affairs and the Atlantic.⁵⁵

Believe it or not, many editors welcome pitches from informed, serious scholars, even if the scholars have not previously written for popular publications. Most outlets have information on where and how to submit on their websites. Emailing editors directly is also an option.⁵⁶ In many cases, busy editors will not respond to inquiries or will otherwise not give a pitch the time it deserves. So be it. Curse to yourself, move on, and submit elsewhere, repeating as necessary. After initial contact is made, subsequent submissions are often easier, especially if an author proves to be authoritative, responsive, and otherwise easy to work with. I edit the “Foreign Policy Essay” at Lawfare and regularly feature content that draws on long academic articles. A simple email to me usually leads to a response—the author and I discuss if a piece might be suitable and, if so, how it might draw on the original research but reach a different, policy-focused audience.

After following some form of steps one through six, there are several important questions that scholars should pose to themselves as they draft policy recommendations for their articles. These questions have no right answers—but considering them will help properly situate the research in ways that policymakers find useful.

It is fine to think big, and it is also fine to think small—each category has different audiences and different impacts. Mid- or senior-level officials are more likely to act on smaller, more fine-grained recommendations. They can use their bureaucratic power to advocate purchasing a particular weapons system, strengthening an international organization, or using financial tools instead of military force to coerce an adversary. They cannot, however, easily establish a new norm on a controversial topic, jettison the 1947 National Security Act, or dump a long-standing ally in favor of a new one. Yet such broad recommendations, even if infeasible in the short or medium term, are part of what academics contribute to a debate. By changing public and elite perceptions over time, scholars can give policymakers more agency to overhaul their approach, thereby loosening the constraints of the policy box. Another factor to consider is the timing of a recommendation. For example, proposing that the European Union change its aid recipients as the deadline for doing so approaches might be more influential than making that same recommendation months or years after the deadline has passed.

A related question is whether a recommendation focuses on either solving or mitigating a problem. The former, obviously, is better, but in most cases it is unrealistic. If scholars have solutions for how to finally end civil wars, reconcile embittered ethnic groups, or ensure that nuclear war is an impossibility, then they should propose them! Yet small improvements in dangerous situations are tremendously valuable. Recommendations that make a civil war a little less likely, reduce the odds of a counterproductive intervention, or minimize wasted time or resources have measurable consequences. Offering a recommendation that reduces the number of refugees from one million to 950,000 is less consequential than preventing the disaster that created one million refugees. But it is still monumental to have fifty thousand fewer refugees, even if the reduction seems insufficient to the scale of the suffering. Most scholarship at best slightly shifts policy, and academics should be comfortable, indeed proud, that it does so. Academic authors should recognize that sometimes only limited progress is possible given resources, the limits of policy instruments, and political realities.

Policymakers and government analysts are often wrong. And so are academics. A powerful advantage of academics, however, is (or should be) that mistakes are learning opportunities. Scholars can revisit foreign policy decisions and, by understanding why people were wrong, identify neglected variables or scope conditions. This power, however, comes with responsibility. Just as methods and sources have inevitable limits and gaps, so do policy recommendations. Small variations in findings—“sanctions always fail” versus “sanctions usually fail”—have profound policy implications, and those variations should be made clear. In addition, scholars should reevaluate their work and highlight their mistakes as ways to ensure their integrity.⁵⁷ Scholars can and will be wrong, and refusing to acknowledge this by making excuses or otherwise avoiding responsibility misses a learning opportunity and reduces the scholar's credibility. In conveying uncertainty, scholars should strike a balance between showing humility by acknowledging research limitations and offering policymakers clear advice despite unknown or conflicting variables.

Although this article focuses on policy recommendations, an important (and at times easier) variant is to consider policy implications: How do research findings inform existing policies? This differs from a recommendation, which asks: “What should policymakers do differently in light of the research findings?” Consider this statement: “There is no need to stop Iran from acquiring nuclear weapons because it is unlikely to use them.” If policymakers are convinced by the argument, they might abandon efforts to coerce Iran or otherwise dramatically change their approach. Another variant is to warn of possible problems with a current approach: “Efforts to stop Iran from acquiring nuclear weapons will face challenges from Iran's insecurity about its own defense capabilities, from the different interests of Iran's trading partners, and from a distrust of U.S. credibility following the U.S. withdrawal from the Joint Comprehensive Plan of Action under President Trump.” This latter example offers no new policy but warns current policymakers about potential obstacles to success. Presumably, policymakers could try to mitigate these constraints, such as by providing side payments to trading partners, but the author is not making a specific recommendation.

By contrast, consider an approach that focuses on recommendations: “To stop Iran from acquiring nuclear weapons, the United States and its allies should focus less on military pressure and more on tightening economic sanctions.” Ideally, a scholar would provide examples for how to do such tightening. When in doubt, it is better to be direct than to let others draw their own conclusions.

Pointing out both policy recommendations and policy implications is valuable for the reasons discussed above. Policymaking is difficult, and solutions are not always obvious. Relatedly, sometimes a particular recommendation is uncomfortable (e.g., don't do humanitarian intervention or otherwise help a vulnerable population). This discomfort does not mean that scholars should avoid highlighting unpopular policy implications. Ideally, scholars would embrace this role as they are less likely than someone in government to suffer career harm from an unpopular position.

Table 1 presents five International Security articles that were published in the last five years. These examples highlight a wide range of topics that might matter to policymakers.⁵⁸ Some focus on a clear policy issue like civil-military relations or on an instrument like military training. Others seem more abstract, examining the nature of the international system or hostile uses of water. The authors represent a mix of both senior scholars and people at earlier career stages. Without claiming expertise on any of these issues myself, I identify potential audiences, policy recommendations, and other policy platforms for each article. (Note: the scholars themselves may disagree with my read on the implications of their work.)

To illustrate the framework and some of the points above more fully, consider a hypothetical article that examines alliances, a staple topic in international relations. The scholar asks, “What causes major power alliances to fail?” To make the work more policy relevant, the author should investigate the policies of the United States or other relevant countries, seeking to understand why they do what they do. Why did alliances with a particular focus, strength, and scope emerge, and what limits did they have? This investigation might involve reviewing government records, interviewing diplomats, and otherwise treating this baseline question as its own research topic. When doing interviews, it is useful to ask counterfactuals to determine why different results did not occur: Why were certain desirable countries excluded or problematic countries included in the alliance? Why was a particularly difficult coordination mechanism included or an alternative excluded? Overall, the scholar should try to get a sense of why the status quo emerged the way that it did.

With this background in mind, it is time to look forward. The question—what causes major power alliances to fail—is of obvious interest to a U.S. or an Asian diplomat, a NATO leader, or another official who might be involved in strengthening alliances. Nonetheless, it is hard to consider a specific audience for this topic. If the piece is highly relevant to the United States, the audience might be the regional bureaus at the State Department, which manage diplomatic relations for their parts of the world. Another option is the Office of the Secretary of Defense for Policy, which has departments that focus on the Indo-Pacific region, Strategy and Plans, and International Security Affairs, among others. Other entities, perhaps less central but also important, might include the International Finance office at the Department of Treasury and various shops within the intelligence community that monitor relations with countries around the world. The more scholars learn about these audiences’ agendas, remits, and resources, the better scholars’ recommendations will be.

Although the variables in play will of course depend on the research, it is worth considering two hypothetical alternative variables: shared interests versus institutional design. The former, of course, is out of the hands of almost all policymakers. But at least some (very senior) policymakers have input into institutional design.

This hypothetical example also illustrates how recommendations and implications may differ. The implications of different interests may lead to problems that are difficult to solve but must be anticipated and managed, perhaps to the point of not relying on allies under certain conditions or expecting only fitful cooperation. In contrast, a scholar may recommend a specific change to institutional design, such as a new entity, or greater powers for or new members of an existing entity. Here, as in other instances, it is important to consider the scope of the recommendation. Academics might rightly propose an entirely new alliance structure, such as an Asian version of NATO. Or they might focus more narrowly (but with more chance of influencing the debate) on how to tweak an existing structure to make it more effective.

Similarly, it is useful to consider how different elements of national power might help, and drafting a basic policy menu is a useful first step. The scholar should ask how diplomats, intelligence officers, the military, and economic actors like the Treasury Department might contribute. Imagine holding a meeting (or, ideally, interviewing people from different agencies) and think about how each might play a role.

It is also valuable for scholars to think ahead about likely problems with their recommendations. If, say, the recommendation is more resources to help gain the goodwill of a particular country, the trade-off is one that senior policymakers always face: fewer resources for other countries. But there may be less obvious costs and trade-offs. Might strengthening the alliance alarm a neighbor, perhaps leading to a dangerous spiral? Might the ally become more aggressive, creating a moral hazard, or, conversely, fear being chain-ganged into a conflict? Such possibilities need not be covered exhaustively, but it is important to acknowledge the limits of a recommendation. Again, interviewing and engaging with relevant policymakers can highlight these limits.

When the research is completed and published, it is time to consider additional publishing options. Many of these should be tied to current events: For a scholar writing in early 2024, what does research on alliance weakness tell us about how the Australia-United Kingdom-United States alliance might hold up or how Sweden and Finland's accession to NATO might be best managed? Leading newspapers might find these topics of interest, as would more specialized outlets like Foreign Affairs and Foreign Policy. When possible, scholars should give briefings on their work or otherwise promote it.

Writing policy recommendations can seem daunting, and in many ways it is. It can be done poorly and at times even counterproductively. When done well, however, recommendations can help guide decision-makers and the public on the world's more difficult issues.

In many ways, the process is the same for crafting both a better policy recommendation and a better article. Use clear, jargon-free prose and structured arguments to make recommendations more convincing.⁵⁹ Authors should seek out criticism, ideally from those with policy experience as well as from fellow scholars. The editors at International Security are an invaluable resource: they can help scholars think through and fully consider both ideas and implications. By making policy recommendations, scholars join a broader community that seeks to make the world a better place. It is not an easy task, but it is a necessary and rewarding one.

The author would like to thank Michael Desch, James Goldgeier, Matthew Kirchman, Ines Oulamene, Kenneth Pollack, Jeremy Shapiro, and the anonymous reviewers for their comments and excellent feedback on previous versions of this article.