An open access publication of the American Academy of Arts & Sciences
Fall 2022

Specific Sources of Trust in Generals: Individual-Level Trust in the U.S. Military

Max Margulies and Jessica Blankshain
View PDF

This essay explores the individual-level determinants of trust in the U.S. military. Prior research has identified five possible drivers of societal trust in the military: performance, professionalism, persuasion, personal connection, and partisanship. Using data from the American National Election Studies and the General Social Survey, we emphasize the importance of understanding trust at an individual level, as perceptions of military performance and professionalism are not objective but mediated by individual-level factors. Our findings reinforce mixed support for trust being linked to assessments of military success on or off the battlefield, and undermine arguments that relate high trust to a widening gap between the military and civilian society. We also present new evidence for generational and ideational sources of military trust consistent with recent speculation that trust in the military is declining. Overall, we show that individual-level trust may be difficult to change, but that public trust in the military has consequences for a variety of defense-oriented policies.

Max Margulies is Director of Research and Assistant Professor at the Modern War Institute at West Point. In addition to his primary interests in military personnel policies, he studies and writes broadly on civil-military relations, strategy, and conflict.

Jessica Blankshain is Associate Professor of National Security Affairs at the U.S. Naval War College. She is co-author of Decision-Making in American Foreign Policy: Translating Theory into Practice (with Nikolas K. Gvosdev and David A. Cooper, 2019), and writes for both academic and popular publications.

In August 2021, commentators debated whether the fall of Kabul following the American exit from Afghanistan was President Biden’s “Saigon moment.”1 Many juxtaposed photos of desperate crowds at Hamid Karzai International Airport with those of lines of evacuees struggling to board American helicopters during the fall of Saigon in 1975. The Taliban’s recapture of Afghanistan came approximately a year and a half after The Washington Post published the “Afghanistan Papers,” previously confidential internal Department of Defense interviews associated with the Special Inspector General for Afghanistan Reconstruction, the office responsible for overseeing Afghanistan reconstruction and relief projects. The interviews contradicted public statements by civilian and military officials about the United States’ accomplishments in Afghanistan. The title, of course, evokes another secret wartime historythe Pentagon Papers, leaked and published in 1971 as the “secret history of the Vietnam War.”2

In the 1970s, the combination of Vietnam and Watergate led to a crisis of confidence in U.S. governmental institutions, including the military. Within a few decades, the military had successfully regained public trust while other governmental institutions, by and large, had not. Will the fallout from the withdrawal of troops from Afghanistan, and the prolonged conflicts of the post-9/11 era more broadly, similarly reduce public confidence in the military? There is some evidence that confidence is already eroding. Most often cited is a November 2021 survey conducted on behalf of the Ronald Reagan Institute that finds Americans’ confidence in many institutions, but especially the military, has fallen since 2018. Other surveys, like Gallup polls and the General Social Survey, also show a decline, though not as starkly.3

To understand where public confidence in the military is likely to head in the future, we need to understand what drives it. Who has trust in the military, and why? Answering these questions will also help us understand whether and why trust in the military matters. If trust in the military is consistently much higher in some segments of the population than in others, there is the risk of not only increasing polarization between these communities, but also that some will have a harder time making their voices heard in the policy-making process. This is particularly worrisome if trust is also associated with policy preferences. To that end, we must also examine the extent to which trust is related to specific military policies or democratic accountability. Do people with high trust in the military show more support for policies preferred by the military? And does their high trust translate to greater confidence in the use of military force abroad?

Scholarly investigations of Americans’ high trust in the military in the post–Cold War era identify five interrelated Ps as possible drivers of public trust in the military: performance, professionalism, persuasion, personal connection, and partisanship.4

The first two determinantsperformance and professionalismreflect rationalist explanations for public trust. In essence, the military earns public trust by demonstrating competence and character, and can lose trust through operational and ethical failures. Accordingly, we would expect to see public trust in the military vary in response to major events, such as battlefield victory or defeat, and highly publicized acts of heroism or scandal. Polling data from the Vietnam War era through the first Gulf War lend some support to this theory.5 Through the 1980s and 1990s, in particular, public trust was highly correlated with military performance, with a notable boost from the 1991 Gulf War.6 And yet, high trust remains despite two decades of U.S. military occupation in Afghanistan and Iraq that was “bloodier, slower, and less decisive than the American public had come to expect.”7

Of course, even by this performance-based logic, what matters is not objective military performance but rather the public’s perception of performance. The public may blame operational failures not on the military, as it did in Vietnam, but on the decisions of civilian policy-makers.8 In a 2019 Pew survey, 90 percent of respondents said military leaders “do a good job preparing military personnel to protect the country” all/most or some of the time.9 This was the highest assessment of performance for any occupation surveyed, with members of Congress coming in last at 46 percent. It is difficult to disentangle cause and effect here. Do people trust the military because they believe the military has performed well and is not to blame for any failures? Or does the public lay blame for operational failures on civilian leaders precisely because they trust the military more than civilian government?

The “professionalism” determinant has a clear relationship to performance, but also encompasses a wider array of issues, including “the way the institution has responded to social problems” like drug abuse and barriers to racial and gender integration.10 To this list, we might add barriers to the integration of other underrepresented groups (including gay and transgender service members), crime, and other ethical scandals. The perceived professionalism of the post-1980s all-volunteer force (AVF) was a marked contrast to the scandals and turmoil associated with the Vietnam War–era and early AVF-era military.11 But as with operational performance, high levels of trust have continued despite major scandals, including prisoner abuse at Abu Ghraib, the Fat Leonard corruption investigation, and high-profile reports of sexual assault and right-wing extremism within the ranks.

The public may view some of these issueslike sexual harassment and assaultas reflections of broader societal problems, rather than as failures specific to the military.12 People may also forgive ethical lapses because they believe the military is better than civilian society at holding perpetrators accountable.13 This view is somewhat supported by the results of a 2019 Pew survey investigating trust in power and authority. While 50 percent of respondents thought the military behaved unethically at least some of the time (the lowest of any occupation), 57 percent thought the military faced consequences for these lapses (the most of any occupation). For comparison, 81 percent thought members of Congress acted unethically and only 25 percent believed they faced consequences.14 There are many, however, who do not share this view of military accountability, as evidenced by continuing congressional efforts to reform the military justice system. Again, this argument may be somewhat circular: Do people trust the military because of its track record for accountability? Or do they believe in military professionalism and accountability, regardless of objective evidence, because they trust the military as an institution?

The degree to which perceptions of performance and professionalism are malleable highlights the possible role of the third P: persuasion. Persuasion encompasses public depictions of the military, including “the careful use of advertising, movies, and the news media to portray the military’s improved performance and professionalism in the best possible light.”15 Military recruiting efforts are one key source of persuasion. With the introduction of the AVF and associated need to compete in the labor market, military recruiting shifted toward selling a product (military service) to customers (potential recruits), which involved “sophisticated and expensive military advertising campaigns” and “intensive market research.”16 Such efforts are aimed not only at those of recruiting age, but also at parents and other influencers, and may also boost the broader public’s image of the military.17

Beyond targeted recruiting efforts, the Cold War’s militarization of foreign policy and culture more broadly, from criminal justice to fashion, may have also contributed to positive views of the military: “The climate of comfort with military imagery and military organization certainly does no harm to the military’s image.”18 While fewer Americans than in the recent past have personal experience with military service, images of the military are everywhere, from popular movies and TV shows to professional sporting events and commercial advertising. By this logic, trust in the military remains high despite the wars in Iraq and Afghanistan in part because of “the popular militarism that is a key part of modern American culture” and shows no signs of fading.19

Persuasion may have long-lasting effects, with formative sociocultural experiences in early adulthood shaping generational attitudes for decades afterwards.20 Twenty years ago, public policy scholar David King and author Zachary Karabell argued that “Generation Xers and Millennials are far more likely than their Baby Boomer parents to have confidence in the military,” because baby boomers’ views were shaped by Vietnam, while younger generations were shaped by the perceived successes of the 1990s, such as the 1991 Gulf War and the 1994 intervention in Haiti.21 The 1990s also saw the fiftieth anniversary of World War II and prominent celebrations of the heroism of American soldiers fighting in “The Good War.”22 According to this generational consequence of persuasion, we might expect more skeptical cultural depictions of the military during the era of inconclusive “forever wars” to lower trust among Generation Z more than for preceding generations, whose average lifetime trust may be anchored to different formative experiences.

This brings us to a fourth potential driver of trust: personal connection to the military. A personal connection to the military could affect trust by mediating perceptions of performance and professionalism, or the effects of persuasion. But personal connection could also affect trust through distinct psychological mechanisms. In the late 1990s, there was considerable support for the idea that societal trust in the military would decline as fewer members of society had a personal connection to the military.23 As part of the landmark survey of civil-military attitudes from the Triangle Institute of Security Studies (TISS), political scientists Paul Gronke and Peter D. Feaver tested this hypothesis and found that among elites, military experience was linked to greater confidence in the military.24 Particularly in the AVF era, a positive correlation between personal connection and trust could in part result from self-selection: those with more trust in the military might be more likely to serve or encourage their family members to do so. Trust and confidence could also be linked to the character of the experience with or connection to the military, although this could be positive or negative.

By contrast, a recent popular hypothesis, sometimes called “patriotism-lite” or “thank you for your service” culture, posits that the military’s high polling numbers are driven by those who are not connected to the military.25 For those who have not served, knowing that “others bear that burden [of military service] leads to guilt and gratitude, which become expressed through superficial demonstrations of appreciation and pride.”26 Security scholar David Burbach has argued that high levels of confidence in polling do not necessarily translate into real support for the military in terms of resources or policy deference.27 Patriotism-lite is often described as a post-draft and distinctly post-9/11 phenomenon, as long-running wars have been fought by a small segment of the population.

Finally, a fifth P, partisanship, is becoming increasingly prominent as a source of trust in the military. Burbach found that partisanship is the strongest predictor of trust in the military, with Republicans having higher trust than Democrats, and both sets having higher trust than independents.28 In data from the General Social Survey, the partisan split begins in 1980 and widens over the subsequent thirty-five years until it overwhelms other demographic variables.29 Individuals are also likely to report higher trust in the military when their co-partisan is president, and when presidential approval is high.30 A 2019 Pew survey similarly found a partisan gap, with Republicans having more faith than Democrats in military leaders.31 One possible explanation for this partisan split is that Republicans and Democrats get their news from different sources, and process it using different cognitive biases, resulting in fundamentally different perceptions of the military’s performance and professionalism arising from their distinct persuasive environments.32 There is some evidence that Republicans view the military as a part of their “in-group” in a way that Democrats and independents do not.33

There are, however, signs these partisan dynamics may be changing. As discussed above, partisans at both ends of the spectrum may trust the military less today than only a few years ago. In a 2019 survey, political scientists Ronald R. Krebs and Robert Ralston found that, during the Trump administration, Democrats were more likely than Republicans to endorse deference to the military, presumably as a check on a president they did not trust.34 In a follow-up survey, Krebs and Ralston found that Democrats were less deferential when Biden, their co-partisan, became president, though Republicans’ deference did not rise as they had expected.35 They also found a large decline in the percentage of Americans who strongly agree that “Members of the military should be allowed to publicly express their political views just like any other citizen,” from 55 percent in the 1990s TISS survey to 28 percent in their 2021 survey; as well as decreases in support for public-policy advocacy or criticism of civilian officials by members of the military, particularly among Republicans.36 While less deference to the military is not inherently bad, it could undermine public support for the norm of civilian control if it is driven by the belief that the military is a partisan opponent.37

We now turn to two large, long-running surveys of the American public’s attitudesthe American National Election Studies (ANES) and the General Social Survey (GSS)to understand better the evidence supporting each of the drivers of confidence proposed above. The ANES uses a feeling thermometer toward the military, asking respondents to rate the military from 0 to 100: higher the warmer/more positively they feel toward it, and lower the colder/more negatively they feel. Warmth is a good measure of positive attitudes, but it is not specifically a measure of trust or confidence. To the extent people generally value trustworthiness, we should expect the feeling thermometer to correlate positively with confidence in the military. While the ANES extends further back in time to examine attitudes during the Vietnam War, questions about feelings toward the military were excluded from surveys after 2012. The GSS has current data that reflect whether respondents have “hardly any,” “some,” or “great” confidence in the leaders of the military from the end of the Vietnam War through the present, giving an up-to-date view on trends. Unfortunately, the GSS has few questions on foreign-policy preferences we would expect to correlate directly with confidence in the military.

Performance and professionalism may correlate with confidence in the military, but data from both the ANES and GSS show that they do not tell the whole story. On one hand, the ANES feeling thermometer shows warmth toward the military decreasing over the course of the Vietnam Warand afterexcept for a brief blip as the war was drawing to a close in 1972. Consistent with Brady and Kent’s findings earlier in this volume, our analysis of both GSS and ANES data shows increases and relatively higher points in the early 2000s, when the wars in Iraq and Afghanistan seemed to be nearing successful conclusions. The GSS high point for confidence is 1991, from a survey fielded during and immediately after the Persian Gulf War, which was widely perceived as a resounding success. The ANES did not field a survey in that year, but its closest surveys both before and after show that any spike it may have picked up would have been fleeting.

While these trends seem consistent with the performance hypothesis, it is odd that both the GSS data on confidence and the ANES feeling thermometer reflect low points in the 1980seven lower than the years immediately after the Vietnam War. Confidence dipped during the worst of the Iraq War, but it never dropped as low as during the Vietnam War, and the ANES shows no similar dip for warm feelings. Of course, this might simply be because, from a U.S. casualty perspective, the Iraq War never came close to going as poorly as the Vietnam War. Nonetheless, even within the context of the post-9/11 wars, the performance hypothesis has a hard time explaining the GSS high point for post-9/11 military confidence in 2018, as well as the consistently high ratings in the ANES from 2004–2012.

The performance hypothesis does not fare much better when we examine the ANES questions about whether invading Vietnam or Iraq was a mistake. Figure 1 shows that in general, a greater proportion of respondents viewed invading Vietnam as a mistake as the war dragged on. However, average warmth only dipped from 75 percent in 1964 to 70 percent in 1972. While the performance hypothesis expects positive attitudes toward the military to track closely evaluations of the military’s success, this slight decline in warmth is not commensurate with the large shiftfrom 25 percent in 1964 to 57 percent in 1972in the public’s belief that the war was a mistake. To the extent that the public may have blamed policy-makers, rather than the military, for starting an ill-advised war, this could support the argument that public trust in the military is resilient when blame is shifted to civilian policy-makers.

We expect changes in performance to affect attitudes toward the military most among people who pay the most attention to relevant military events. People who report an interest in military policy or national news might be more likely to follow events that demonstrate both military performance in foreign conflicts and professionalism in upholding domestic or organizational values. Here, the performance hypothesis again finds mixed support: while confidence in the military tracks Iraqi civilian casualtiesmeaning that when casualties are lower, public confidence is higherthere is no obvious difference in how confidence changes between those who report being “very interested” in military affairs versus “not at all interested.”

The data in Figure 2 suggest that people who pay a lot of attention to what the military is doing do not respond to events differently from people who pay no attention. The ANES questions that directly ask about respondents’ attention to what was happening in Vietnam reach a similar conclusion. In 1968, the Tet Offensive led many Americans to realize the fight in Vietnam would be much harder than they had been led to believe, despite the military’s tactical successes. As a result, we might expect the warm feelings toward the military in 1968 to be lower for people who pay attention to the war than those who did not. In contrast, a significantly greater proportion of people who reported paying attention to Vietnam had positive attitudes toward the military than did those who reported not paying attention, 79 to 70 percent.38 There is no clear evidence that paying attention to events affected attitudes toward the military over the course of the Vietnam War.

While there is mixed support for the performance and professionalism views, there is strong evidence to support the hypothesis that personal connection is positively correlated with trust in the military. During the height of the Vietnam War, the ANES asked about respondents’ and their families’ or friends’ recent military service within “the past five or six years.” Data from 1968 show that people with connections to the military feel more positively toward it.39 This outcome is particularly noteworthy given that many respondents were connected to the military by an unpopular draft, which if anything should create a negative impression. There is no significant difference in mean warmth toward the military between those with Vietnam War–era military service and those without. However, respondents who knew anyone who had recently served in the military (“in the past five or six years”) rated the military five points higher on average (76 versus 71) than those who did not know anyone who had served. Knowing someone who had served in Vietnam had the same effect (77 versus 72). Respondents who knew anyone in the military (including themselves) were also more likely to report feeling very warmly toward the military than respondents who did not know anyone in it: 34 percent of those who knew someone in the military rated it 90 or higher, compared with just 23 percent of respondents who did not know anyone in the military. In fact, the closer connections the respondents had to the Vietnam War, the more likely they were to have very positive feelings toward the military: while 37 percent of respondents who knew anyone who served in Vietnam rated the military 90 or higher, compared with just 25 percent of respondents who did not know anyone, this number rises to 41 percent for respondents who served in Vietnam themselves or had family who did.

The post-9/11 era shows similar patterns. As seen in Figure 3, mean warmth toward the military was roughly five points higher for respondents with military service than for those without in 2004, 2008, and 2012. Interestingly, the greatest difference is between respondents with no military service and older veterans: pre-9/11 veterans report the most warmth (86) and those with no military experience report the least (80). Average warmth among post-9/11 veterans was statistically indistinguishable from these other groups. Respondents with either a personal or familial connection to military service had the largest gap in warm attitudes in 2012: those with a connection to the military rated it 7 points higher on average than those without.

Likewise, the GSS data support the hypothesis that personal connection correlates with higher confidence in the military. There is no difference in the mean levels of military confidence between those with military service and those without in any years before 2010, except for 1988 and 1993, when respondents with military service had slightly more confidence. Further disaggregating military service into those who served more than four years and those who served fewer than four years provides more evidence that longer service is associated with greater confidence in the military: Not only do more years reflect this relationship (1975, 1977, 1988, and 1993), but the magnitude of the difference is also substantially larger in the latter two years. Between 2010 and 2018, there is a significant difference between those who served and those who did not in each year except 2012. Again, the most salient difference in this period seems to be between those who served more than four years and those who did not serve at all.

The clear relationship between military connections and higher self-reported levels of trust in the military undermines the expectations of the patriotism-lite argument by showing that familiarity does not, in fact, breed contempt. It is particularly notable that military service appears to correlate with warmer attitudes toward the military even during the Vietnam-draft era, though the difference has certainly become starker since 9/11. This correlation suggests that positive feelings toward the military are not solely a result of self-selection in the AVF era. Greater trust in the military is not driven by people with fewer connections to the military.

Our analysis also confirms a partisan divide in trust. While warmth toward the military has been on the rise among both Democrats and Republicans for decades, this shift has been most pronounced among Republicans.40 Trust among Democrats has only recently reached levels comparable to those before the collapse of trust in the late 1970s and early 1980s; trust among Republicans rose sharply around the year 2000 and has stayed high since, although there are now signs confidence may be falling.

It is not clear what has driven this sharp divide in partisan attitudes. One possibility is that political partisanship is interacting with other demographic or cultural shifts. We explore whether the partisanship and persuasion explanations interact to explain recent trends in trust: formative experiences shape and secure early impressions about the military but are filtered and amplified differently through partisan lenses. If this hypothesis is correct, we would expect partisan trends in trust in the military to differ across generations and demographic groups. Conversely, if the trends over time are similar for these different groups, that would indicate that something exogenous is driving trust for all groups, despite their different formative experiences.

As context, it is important to consider the change in national demographics reflected in these surveys. The percentage of respondents who self-identify as Democratsincluding independents leaning Democrathas declined since the 1950s, while the percentage who self-identify as Republican or leaning Republican has generally risen since around 1980. The average age has risen from a low point of around forty-four in the early 1980s to over fifty in 2020. This seems to be driven by the baby boomer generation, which has had the largest share of the survey sample size since 1980. For almost the whole period since 2000, boomers have made up at least 10 percentage points more of the sample sizeand often 15 points morethan the next largest group.

This finding is significant because age is strongly positively correlated with more trust in the military, and boomers and Generation Xers have together made up between 60 and 68 percent of the sample throughout the last two decades. At the same time, the average combined age of these generations from 2000 onward is fifty. Could attitudes from either or both generations be driving the recent rise in trust in the military? The ANES data show boomers are significantly more likely to be Republican than Democrat, and while the difference is slight, it increases over time. In addition, as Figure 4 shows, the partisan polarization that has increasingly characterized American society for the last two decades also started much earlier for boomers, at least with respect to attitudes toward the military: significant differences between Democratic and Republican confidence in the military begin in the 1980s for boomers and the early 1990s for Generation X. Conversely, the partisan divide is only evident in the silent generation starting in the year 2000.

Interestingly, millennial Democrats are the only group for whom there is a clear decrease in confidence in the military over the last twenty years. Millennial Republicans do not exhibit the sharp rise seen in older generations, and may even have declining confidence, though we should be cautious given the small numbers of Republican millennials in the sample prior to the mid-2000s. Overall, the partisan gap seems larger for the two youngest generations, though here too we must be cautious about interpreting the low number of respondents (seventy-two) from Generation Z across all survey years.

There is also evidence that exogenous factors shaped attitudes. While boomers show an unusually steep increase in warmth toward the military as they age compared with other generations, each generation saw a sharp rise in warmth regardless of party affiliation from the late 1990s through the early 2000s: when members of the silent generation were around sixty and baby boomers were approaching fifty. Together, the evidence indicates that increasing partisanship has played a role in attitudes toward the military, but the generational composition of American society, especially changes among baby boomers and Generation X, has also played a role (see Figure 5). The major, polarizing debates about the Vietnam War and the transition to the AVF during these generations’ formative years may have contributed to their distinct military attitudes.

Trust in the military may affect public policy by shaping what resources and roles the public envisions for the military. In this section, we use the ANES survey questions on a wide range of contemporaneous policy issues to examine the consequences of trust in the military. Is there any relationship between how individuals feel about the military and their policy preferences on related issues?

The public’s willingness to use military force abroad is one such issue. In 1992 and 1996the two years that the ANES asked both its military thermometer question and the general foreign policy question about “how willing the United States should be to use international force to solve international problems”there was a wide gap in military warmth between those who answered “extremely” or “very willing” to use force and those who answered “not very willing” or “never willing.” Respondents who were more willing to use military force felt more than 12 points warmer toward the military than those who were more reluctant to use force. In addition, more than half of the respondents who felt more warm than cold to the military were extremely or very willing to use force to solve international problems, but less than one-third of those who felt colder toward the military were willing to use force.

We should be cautious about assuming the same relationship holds today as it did in 1992 and 1996, the brief period of unchallenged American supremacy that lasted from the end of the Cold War until 9/11. While the ANES does not ask about confidence in the military in its most recent surveys, we can get a closer look at this relationship through the ANES polling about attitudes toward specific uses of force in Vietnam, Afghanistan, and Iraq. One such question asks whether respondents think the United States “did the right thing” in getting involved in the war, or, in the case of Iraq, whether it “should or should not have sent troops.” From 1968 through 1972, as well as during the Iraq War in 2008, respondents who thought sending in troops was the right thing to do viewed the military between 5 and 9 points more favorably than those who thought sending in troops was wrong. Similarly, in nearly every year with available ANES data, respondents with a positive view of the wars in Iraq and Afghanistan felt more than 10 points warmer to the military than respondents with a negative view of those wars. As Figure 6 shows, compared with the data gathered during the Vietnam War, respondents with both negative and positive views of war feel warmer toward the military today.

Warmth toward the military is also associated, or at least was associated during the Vietnam War, with a preference for more escalatory war strategies. While Figure 7 shows that the average difference in the military thermometer ratings is not consistently large, cross-tabulations reinforce the relationship. In 1964, 1968, and 1970, respondents who felt more warm than cold toward the military were at least 10 percentage points more likely to prefer escalation as a strategy. Even in the waning years of the war (1970 and 1972), the proportion of respondents who self-identified as hawks in favor of pursuing victory was nearly double among people with warmer views of the military (39 percent versus 20 percent in 1970 and 36 percent versus 17 percent in 1972).

Lastly, this relationship is also evident for more general policies, such as defense spending. Figure 8 shows a consistent fourteen-point or greater difference in warmth toward the military between those who prefer to increase defense spending and those who prefer to decrease it. More than 53 percent of respondents who felt warmth toward the military wanted to increase defense spending, compared with less than 25 percent who did not feel warmth toward the military. The GSS shows similar results over an even longer period. From 1973 to 2018, respondents who reported having “a great deal of confidence” were most likely to believe that the United States spends too little on defense. For most of this period, respondents with “some confidence” were also more likely than those with “hardly any confidence” to feel the same. This contrasts with Burbach’s finding that confidence may be superficial and unrelated to concrete policy preferences.

There is much debate about what drives public trust in the military. Our analysis leverages two different national surveys to provide a comprehensive evaluation of individual-level predictors of attitudes toward the military. The evidence does not point to any decisive factor but offers promising directions for further inquiry. While trust is somewhat responsive to major events that reveal new information about military performance or professionalism, these shifts are inconsistent and tend to be smaller than we might expect. Instead, trust may be resilient to changes in the news cycle because it is shaped by a deeper social and personal context. Formative experiences, such as those that define generations or stretch partisan divides, may create a strong foundation for how individuals assess the military over their lifetime. Perhaps most notably, under both the draft and the AVF, evidence shows that familiarity with the military is associated with greater trust in it. This has important implications for many debates today about the consequences of an increasingly active but insular military, though there remains much to learn about the relationship between personal connections to the military, attention to military affairs, and trust in the military.

At the same time, our analysis has not been exhaustive. We must continue to consider the ways these drivers interact. The drastic change in boomer trust, from the low levels in their youth to consistently high levels of support and trust as they age, shows that there is no simple story of some generations being more confident in the military than others: formative experiences matter but are not determinative. This points to the importance of differences in how individuals process broader events. Similarly, we must learn more about how the public conceives of the military to understand what the public trusts the military to do. There is evidence that at least some groups exaggerate their support for the military.41 Who cares more about demonstrating their trust, and how does this affect their policy preferences? In different times and places, people may trust the military based primarily on their idea of its warfighting capabilities or its morals, or even on their interactions with servicemembers in daily life. One important driver for future investigation is the military’s association with gender and racial norms. The U.S. military has historically been a bastion of (predominantly White) masculine ambition and ideals; does trust change when the public perceives deviations from these ideals?

The relationship between the American public’s trust in the military and other important outcomesthe effectiveness of American foreign policy, the health of American civil-military relations, the soundness of American democracyis also complex. Both high and low levels of trust in the military can have adverse consequences. Trust in the military must be considered in the context of the American public’s views of society and government. It can be counterintuitive to think there are downsides to the military enjoying consistently high levels of public trust, particularly after the experience of the Vietnam War. But this collective effort to keep trust in the military high may have blinded us to other civil-military dangers, which in turn endanger both military effectiveness and democratic foundations.42

A persistent and large gap between confidence in the military and confidence in civilian governmental institutions threatens to upend the hierarchical nature of proper democratic civil-military relations. We have seen calls for the military to have a larger role in policy-making, including on issues not directly related to military expertise or even foreign policy. We have also seen repeated calls for more veterans to enter government (likely coming at the expense of increased representation of other voices not already overrepresented in government), with some even campaigning on the platform that their military service makes them uniquely qualified as political leaders. We may also see a vicious cycle as civilian leaders, knowing that the military is more popular than their institutions, use the military as political shield/weapon when beneficial, which only serves to further elevate the military over civilian institutions and thereby further exacerbate the trust gap.

But there is also a danger that use of the military for political ends could ultimately have the opposite effect, turning the military into “just another political institution.” Politicization arising from high trust in the military may, down the line, cause a drop in trust. The Supreme Court may be a cautionary tale of how quickly trust can change when the public perceives that an institution has become too politically motivated.43  We may be beginning to see such a shift in attitudes toward the military. For example, partisan actors have used the debates over inclusivity, military justice, and vaccine policy to create perceived divides within the military, portraying woke leadership as undermining the true warriors.44 It remains to be seen whether this will have an enduring effect on who trusts the military and how much. Given the clear relationship between trust in the military and civilian views on important defense policy, it is paramount that we find a way to foster appropriate and balanced attitudes toward the military.

Authors’ Note

All views are the authors’ own and do not represent the official views or positions of their employers.


  • 1See, for example: Joe Concha, “Biden’s ‘Fall of Saigon’ in Afghanistan Presents Worst Moment Yet of Presidency,” The Hill, August 15, 2021; Nick Turse, “The Fall of Kabul,” The Intercept, August 15, 2021; and Joshua Zeitz, “Joe Biden’s Kabul Is Not Gerald Ford’s Saigon,” Politico, August 17, 2021.
  • 2Report of the Office of the Secretary of Defense Vietnam Task Force (Washington, D.C.: U.S. Department of Defense, 1969).
  • 3Ronald Reagan Institute, Reagan National Defense Survey (Washington, D.C.: Ronald Reagan Presidential Foundation and Institute, 2021), 4; and NORC at the University of Chicago, “The General Social Survey.”
  • 4Paul Gronke and Peter D. Feaver, “Uncertain Confidence: Civilian and Military Attitudes about Civil-Military Relations,” in Soldiers and Civilians: The Civil-Military Gap and American National Security, ed. Peter D. Feaver and Richard H. Kohn (Cambridge, Mass: MIT Press, 2001), 129–161; David C. King and Zachary Karabell, The Generation of Trust: Public Confidence in the U.S. Military since Vietnam (Washington, D.C: AEI Press, 2003); Andrew A. Hill, Leonard Wong, and Stephen J. Gerras, “‘Self-Interest Well Understood:’ The Origins & Lessons of Public Confidence in the Military,” Dædalus 142 (2) (2013): 49–64; and David T. Burbach, “Gaining Trust While Losing Wars: Confidence in the U.S. Military after Iraq and Afghanistan,” Orbis 61 (2) (2017): 154–171.
  • 5Burbach, “Gaining Trust While Losing Wars,” 160–161.
  • 6King and Karabell, The Generation of Trust, 31.
  • 7Burbach, “Gaining Trust While Losing Wars,” 162.
  • 8Ibid., 163; and Hill, Wong, and Gerras, “‘Self-Interest Well Understood,’” 55.
  • 9Claire Gecewicz and Lee Rainie, Why Americans Don’t Fully Trust Many Who Hold Positions of Power and Responsibility (Washington, D.C.: Pew Research Center, 2019), 7.
  • 10King and Karabell, The Generation of Trust, 32.
  • 11Burbach, “Gaining Trust While Losing Wars,” 164–166.
  • 12King and Karabell, The Generation of Trust, 61.
  • 13Ibid.; and Hill, Wong, and Gerras, “‘Self-Interest Well Understood,’” 57.
  • 14Gecewicz and Rainie, Why Americans Don’t Fully Trust Many Who Hold Positions of Power and Responsibility, 3–4.
  • 15King and Karabell, The Generation of Trust, 3.
  • 16Beth Bailey, “The Army in the Marketplace: Recruiting an All-Volunteer Force,” The Journal of American History 94 (1) (2007): 48.
  • 17Barbara A. Bicksler and Lisa G. Nolan, Recruiting an All-Volunteer Force: The Need for Sustained Investment in Recruiting Resources–An Update (Arlington, Va.: Strategic Analysis, 2009), 26.
  • 18King and Karabell, The Generation of Trust, 69.
  • 19Ronald R. Krebs and Robert Ralston, “More Deferential but Also More Political: How Americans’ Views of the Military Have Changed Over 20 Years,” War on the Rocks, November 17, 2021.
  • 20King and Karabell, The Generation of Trust, 3.
  • 21Ibid., 12.
  • 22For a discussion of American mythologizing of World War II, see Elizabeth D. Samet, Looking for the Good War: American Amnesia and the Violent Pursuit of Happiness (New York: Farrar, Straus and Giroux, 2021).
  • 23Gronke and Feaver, “Uncertain Confidence: Civilian and Military Attitudes about Civil-Military Relations,” 130.
  • 24Ibid., 139–141.
  • 25See, for example, James Fallows, “The Tragedy of the American Military,” The Atlantic 315 (1) (2015): 72–90.
  • 26Burbach, “Gaining Trust While Losing Wars,” 156–159. Burbach also credits patriotism-lite to Charles Moskos in “Patriotism-Lite Meets the Citizen Soldier,” United We Serve: National Service and the Future of Citizenship, ed. E.J. Dionne Jr., Kayla Meltzer Drogosz, and Robert E. Litan (Washington, D.C.: Brookings Institution, 2003), 33–42.
  • 27David Burbach, “Confidence without Sacrifice,” in Reconsidering American Civil-Military ­Relations: The Military, Society, Politics, and Modern War, ed. Lionel Beehner, Risa Brooks, and Daniel Maurer (Oxford: Oxford University Press, 2021), 160–161.
  • 28David Burbach, “Partisan Dimensions of Confidence in the U.S. Military, 1973–2016,” Armed Forces & Society 45 (2) (2019): 211–233.
  • 29Ibid., 221.
  • 30Ibid., 226.
  • 31Gecewicz and Rainie, Why Americans Don’t Fully Trust Many Who Hold Positions of Power and Responsibility, 7.
  • 32Michael Robinson, “Who Follows the Generals? Polarization in Institutional Confidence in the Military,” APSA Preprints (2019).
  • 33Ibid.
  • 34Ronald R. Krebs and Robert Ralston, “Why Conservatives Turned on the U.S. Military,” Foreign Affairs, September 28, 2021.
  • 35Ibid.
  • 36Krebs and Ralston, “More Deferential but Also More Political.”
  • 37Krebs and Ralston, “Why Conservatives Turned on the U.S. Military.”
  • 38The difference between these groups is statistically significant at a 0.04 level. All differences we report in this essay are statistically significant at a 0.05 level at minimum, unless noted otherwise.
  • 39American National Election Studies, “1968 Time Series Study” (accessed September 28, 2022).
  • 40Both the ANES and GSS ask respondents whether they “generally think of” themselves as Democrats or Republicans. We group respondents who leaned one way or the other in with their stronger co-partisans.
  • 41Meredith Kleykamp, Crosby Hipes, and Alair MacLean, “Who Supports U.S. Veterans and Who Exaggerates Their Support?Armed Forces and Society 44 (1) (2018): 92–115; and Burbach, “Gaining Trust While Losing Wars,” 169–170.
  • 42See, for example, Jessica D. Blankshain and Max Z. Margulies, “The Downside of High Trust in the Military,” The New York Times, September 16, 2021.
  • 43Jeffrey M. Jones, “Approval of U.S. Supreme Court Down to 40%, a New Low,” Gallup, September 23, 2021.
  • 44Max Boot, “The Disturbing Strategy behind MAGA Complaints about a ‘Woke Military,’” The Washington Post, September 26, 2022.