What leads some individuals to keep fighting during civil wars despite the risk it entails? Why do some organizations have many committed fighters, whereas others experience widespread desertion? In this exciting and thoughtful book, Theodore McLauchlin argues that these questions are intricately related. Individuals make decisions to desert or stay based on the nature of the small groups they fight in as well as the structure and ideological underpinnings of the broader organization.

McLauchlin develops a relational theory in which bonds of trust between combatants activate norms of cooperation. Fighters are more likely to fight if they trust that those around them are motivated by similar goals and will fight rather than flee. To explain how trust is generated, McLauchlin introduces a mechanism of costly signals whereby soldiers judge the commitments of others via shared experiences in prewar political networks, voluntary acceptance of the hardships of soldiering, or participation in training and combat induction.

Two additional mechanisms include whether fighters share social ties or are divided by factionalism. Social ties, without costly signals, are indeterminate and can even help fighters coordinate with others wishing to desert. When factional differences (such as ethnicity or class) introduce shortcuts about commitment, they can encourage even committed individuals to desert if stereotyped as disloyal. This is a valuable advancement: whereas existing research finds that coercion is often ineffective, McLauchlin shows that coercion applied indiscriminately, rather than based on individuals’ actual behavior, can provoke desertion.

These microlevel mechanisms aggregate. Organizations limit desertion by facilitating costly signaling through training, recruiting willing volunteers, or incorporating prewar political networks. By contrast, organizations that import or encourage factional competition and stereotyping are likely to generate mistrust.

The author tests the argument through analysis of the Spanish Civil War (1936–1939). Using quantitative and qualitative evidence, McLauchlin employs detailed data about foot soldiers and officers to conduct precise microlevel tests on the Republican side (chapters 4 and 5). For example, focusing on the transition to conscription in 1937, McLauchlin uses newly collected data about Republican battalions to assess how costly signals (measured by whether units were predominately volunteers or conscripts), social ties, and factionalism influenced desertion. The findings underscore the relational approach: "a volunteer serving among conscripts was more likely to desert than a conscript serving among volunteers" (p. 64). In the remaining chapters, he assesses macrovariation on the Republican and rebel side (Nationalist Army). He finds that the rebels had fewer deserters because they had many self-selected volunteers, clear military discipline, and some coercive capacity. Finally, he turns to a set of shadow cases in Syria, presenting some evidence that the relational approach explains patterns of desertion.

This book provides a powerful framework for capturing how the seemingly competing foundational concepts of ideology, individual commitment, and community networks influence the cohesion of armed groups. The relational theory explains when and how these variables matter. In McLauchlin’s words, the approach “draws attention to the content of the relationships among soldiers—that is, whether they have a link that suggests to each other that they want to fight” (p. 54). This is a valuable addition to our understanding of civil war processes. Social ties matter when they support organizational objectives, but they do not independently generate the will to fight. Ideology and socialization can create shared perspectives, but their effect is shaped by the relationships between primary group members as they enter and experience combat.

This impressive book has insights for many civil war processes beyond desertion. However, when one considers Desertion in the broader context of civil wars, there are some areas where it leaves unanswered questions. These are opportunities to further build the relational approach.

For example, although the Spanish Civil War allows for detailed analysis, as McLauchlin acknowledges (p. 39), it is a conventional conflict. Do the findings travel to other forms of substate conflict, such as guerrilla warfare? What are the theory’s scope conditions? The empirical tests include state and nonstate actors. Moving beyond conventional war, does the theory apply to all armed groups despite the notable difference in resources often plaguing rebels versus their state foes? If so, does the book meaningfully advance existing findings about how task-based cohesion generates trust in conventional militaries? McLauchlin observes the theory’s proximity to this research in military sociology (p. 10). I believe it does provide avenues for meaningful expansion—if nothing else, transporting this research to rebel groups is a necessary step.

However, the book should do more to engage with unique aspects of civil war. Military sociologists find that conventional armies build trust endogenously. Can insurgents do the same despite their limited resources? Do rebels have such agency, or are they bound by preexisting social and resource endowments? McLauchlin, in developing the theory, presumes that rebels exercise agency through training, discipline, and socialization. Because there are “rarely enough” committed fighters, organizations often must create commitment (p. 195). But those pathways are notably under-theorized, and the conventional nature of the Spanish Civil War makes it difficult to address this question in the substate context.

Indeed, much of the evidence of trust comes from committed fighters self-selecting into disciplined units. The sheer scale of the Republican and rebel forces made such large-scale self-selection feasible; both sides had large contingents of volunteers. Given this empirical reality, the author’s limited description of the content of training and socialization cannot sort out their causal role. Did they socialize uncommitted recruits or serve as focal points for committed fighters? Illustratively, McLauchlin sees self-selection as central in the major Republican transition to militarization during which they gave soldiers an option to depart. Many fighters left, so “staying around after the militarization order seems to have reduced much of the militias to a committed core” (p. 120).

Beyond this ambiguous evidence for the relational trust theory (are groups of self-selected fighters more committed due to bonds of trust or simply higher concentrations of overall commitment?), few insurgents could afford to let large swaths of fighters depart as a selection mechanism. The major example McLauchlin presents of creating commitment in the absence of self-selection is the resource-intensive coercion applied by the rebels (p. 152–158). Although he finds that this was partially effective because the Nationalists had state-like “capacity…for top-down monitoring and surveillance” (p. 156), few substate actors could rely on such an approach.

This focus on large-scale self-selection or coercion overlooks the success that many groups—including those mentioned by McLauchlin, such as ISIS or the People’s Liberation Armed Forces (PLAF, or Viet Cong)—have had in reshaping the commitment of diverse and often uncommitted recruits. McLauchlin’s dismissal of ISIS’s (chapter 9) ability to generate trust as the conflict in Syria progressed undervalues the sheer scale at which it adapted to create combat-capable units in the face of significant opposition. Similarly, due to the conventional focus, the theory does not engage with the reality that insurgents are often “swimming in the sea of the people” (p. 39). That opportunity (and challenge) shapes the way that insurgents build commitment; and, in fact, many groups shape individuals’ commitment before they join. The PLAF used its political infrastructure to prepare individuals for combat by using a tiered structure that promoted fighters from community defense forces to organized combat units.

Nevertheless, McLauchlin offers a substantial contribution to our understanding of how violence is organized. He makes an enormous step in unpacking the micro- and macrodynamics of insurgent organization and linking individual and group-level processes with larger combat units and organizations. Desertion should be read and discussed widely.