In their thousands, right-wing groups marched towards the Senate House. They marched under many different insignia and banners, including far-right, avowedly racist organisations. Some believed reports spread through right-wing media of a deep state conspiracy, others sought to show strength that would weaken or lead to the fall of the left-wing government, elected to power with a slender majority, which some equated with communism. Still others marched against the Jews. What gave the demonstration a deeper potential for escalation still was the presence of members of well-organised veterans groups. These veterans included decorated heroes who had sacrificed much in service to their country. But what of their commitment to democratic processes and values when they had chosen a side that lost an election?
There are obvious parallels between this description and the events in Washington, DC on 6 January 2021. But on 6 February 1934, the French police response to the riot in front of the Senate building in Paris was more aggressive than the US Capitol Police, and some have argued, reflected their panic. The police opened fire, and 14 demonstrators were killed and 236 wounded. The left viewed the events as an attempted coup, the right as a symbol of government oppression that had created martyrs to freedom. Thus began a ‘civil war’ in France that would severely weaken the country in face of the looming crises to come (Jackson 2001). So much so that six years later, one of the world’s most long-lived democracies would commit suicide, its representatives voting away their own power to a dictatorship headed by Le Maréchal, Philippe Pétain. A hero credited with saving France at the pivotal Battle of Verdun in the First World War, Pétain would head the authoritarian, racist, Vichy regime that would collaborate with Nazi Germany until France’s liberation in 1944.
France’s crushing military defeat in 1940 was only part of the story. Instead, it was arguably in part a symptom of an underlying process that had led to an undermining of democratic values. Unlike other democratic states that had fallen that year to the Nazis, France’s elected representatives in 1940 chose not to set up a legitimate government in exile. Instead, many appeared convinced that democracy was a necessary price to pay to facilitate a ‘national renewal’ of France.
A growing body of research points out how the failure of young democracies, such as that of 1930s Germany, often occurs when institutional gatekeepers allow would-be authoritarians to legally assume power in the name of short-term opportunism (e.g. Levitsky and Ziblatt 2018). However, democracies do tend to become more resilient as they survive. Some researchers emphasise the key role of the diffusion of democratic values among citizens in this resilience (Besley and Persson 2019). But much less is known about how support for democratic values erode, even in previously durable democracies. We argue that to help prevent the erosion of democratic values and to reduce political polarisation, a key dimension to understand is the importance of embeddedness in social and economic networks of individuals with the potential for political influence. And, throughout history, a common environment where such networks may emerge is through shared heroism.
In a new paper (Cagé et al. 2020b), we exploit a natural experiment that created an exogenous network of politically influential individuals in France – a hierarchical network of war heroes generated through the systematic ‘millwheel’ rotation of French regiments through the direct command of Pétain during the pivotal WW1 Battle of Verdun in 1916. We wed this to a unique recently declassified dataset on more than 95,314 individual extreme-right wing supporters and collaborators assembled by Free French army intelligence during the Pétain-led Vichy (1940-44) regime.1
An exogenous network of heroes: The rotation of regiments at the Battle of Verdun
Verdun, the ‘Stalingrad’ of WW1, became the symbol of French fortitude and willingness to resist (Jackson 2001) in which 305,440 soldiers on both sides were killed, almost a death a minute over the ten months of what became the longest battle in history.2 The profound significance of the simple phrase “J'ai fait Verdun” [“I did Verdun”], adopted broadly among its veterans, was understood throughout the country (Ousby 2007).
The surprise German offensive at Verdun took France completely by surprise, leading to the sacking of four generals in five days. A ‘snap decision’ was made to put Philippe Pétain in command (Horne 1962, p.129). At the start of WWI, Pétain was a 58-year-old colonel on the verge of retirement after an undistinguished career. He was assigned to Verdun only because he happened to be available at the time. Assigned command over an arbitrary rotation of regiments drawn from across France (see Figure 1), Pétain strengthened the logistics and stabilised the front before he was promoted away from direct command in May 1916.3 By the time the battle was declared over in December, 88.7% of France's 34,947 municipalities had raised a regiment rotated through Verdun, with 50.1% having served under Pétain's direct command. Pétain himself was already being fêted as the Hero of Verdun and was later made Marshal of France.
Figure 1 Rotation of regiments through Verdun, by month, February to December, 1916
Note: From the top left (February) to the bottom right (December), different regiments were dispatched to the Battle of Verdun. Pétain commanded between February and May 1st. The figure displays where all (dark blue), some (light blue) or none of the regiments from each municipality were rotated through Verdun each month.
Support for extreme right-wing parties and Nazi collaboration
To rescue France once more from Germany, Pétain was called to head France after her crushing defeat in WWII. On July 10th, 1940, the two legislative chambers granted the Cabinet the authority to draw a new constitution (Lacroix et al. 2019). Soon, Pétain assumed plenipotentiary powers as Head of State. Upon gaining power, Pétain's regime rapidly dismantled liberal institutions and adopted an authoritarian course. In October 1940, Pétain's collaboration took an overt turn, when a photograph of him shaking hands with Hitler was widely publicised. The regime rapidly took on an extreme right wing and racist agenda, which became even more repressive after the full military occupation of France by Germany in November 1942, when the Vichy regime sponsored the Milice [militia] to hunt down and kill the French Resistance.
Figure 2 maps the distribution of collaborators recorded in the declassified file to their home municipality in 1945, overlaid with their regimental combat experience in WWI. While there is significant regional variation in the shares of collaborators, these shares are disproportionately higher in Verdun-under-Pétain municipalities. These raw differences are robust in a regression framework – we show that exposure to Verdun-under-Pétain significantly increases collaboration rates by 6.9%, compared to otherwise similar municipalities within the same department. These effects appear across the spectrum of collaboration, from joining extreme right-wing organisations and deep economic collaboration to volunteering for Vichy paramilitaries that hunted Jews and the Resistance, and even for the Waffen SS heading to the Eastern Front in 1944, when it was already clear that the Nazis would lose the war. Verdun-under-Pétain municipalities were also 8% less likely to raise civilian members of the French resistance.4
Figure 2 Quintiles of the distribution of Nazi collaborators per capita overlaid with rotation at Verdun
Why did some of France’s greatest heroes become some of their gravest villains? We use novel hand-collected voting data to show that these extreme acts had their genesis in increasingly pronounced political choices that emerged in the post-WWI period and mimicked Pétain’s own views. Widely known as an anti-communist and with increasingly authoritarian sympathies, Pétain’s speeches, which initially focused on veterans’ groups, took a more overt political tone just before the elections in 1936. We show that municipalities that served under Pétain at Verdun, although politically similar to other municipalities in 1914, begin by voting against the communists in the 1924, and then increasingly vote for the right (and later the extreme right), particularly in the 1936 elections (Figure 3). This accentuated the severe polarisation that weakened France in the run-up to the next war (Jackson 2001).
Figure 3 The vote share for the right in French legislative elections, 1914, 1924 and 1936 in municipalities whose regiments served under Pétain at Verdun
Importantly, there are increasing differences over time in the vote for the right (and later for Nazi collaboration) in municipalities whose troops served under Pétain at Verdun, in contrast to those serving under Pétain’s command before the battle (Figure 4), or at Verdun after Pétain’s promotion. These patterns are consistent with complementarity (Milgrom and Roberts 1990) in political influence within the network – the political influence of Pétain’s heroic credentials was stronger when legitimised and diffused through the network of those that served with him at Verdun, and they in turn became more influential the more their leader was viewed as a hero.
Figure 4 Increasing differences in the effect on the right vote in inter-war elections in municipalities whose regiments served under Pétain at Verdun, rather than under Pétain’s previous commands (with 95% confidence intervals)
The presence of complementarities can help explain not only why the heroes of Verdun acquiesced in the demise of the Republic for which they had fought, but why some even became the staunchest supporters of the Nazis over time. If others that share a heroic credential are now considered traitors, this will reduce the value to each hero of their own heroic credential and identity. This is particularly the case for the most public face of the network, Pétain. As a result, the heroes of Verdun had more incentives to support their leader – it was costlier to turn against him than for others who lacked their common group identity. Further, they faced greater incentives to participate in organisations and other reinforcing devices that strengthened the value of their heroic credentials and the network as a whole. And the more individuals invest, the costlier it becomes to abandon the network.
These reinforcing incentives over time may explain why the home communities of the heroes of Verdun still supported Pétain even when it was clear that the Nazis were losing. They can also explain why these preferences and identities proved durable in France, even after the collaborationist regime would fall, far-right parties were banned, and Pétain himself was convicted of high treason.5 As we show, Verdun-under-Pétain municipalities reveal a durable average 6.8% increase in vote for the right in post-WWII elections until the end of the Cold War (Figure 5).6 These effects were particularly accentuated during the key moments of political crisis in post-war France – the Algerian War of 1958, the civil unrest of 1967-68, and the election of the first post-war Socialist president in 1988.
Figure 5 Effects on the left and right vote share in post-WWII elections in municipalities whose regiments served under Pétain at Verdun (with 95% confidence intervals)
In a different country, at a different time, mobilised by claims of a different rigged election, another armed mob gathered before the gates of a different State House. Concerned that the militia he commanded might themselves be politicised, the commander ascended the steps to talk to the crowd, unarmed and alone. “Men, you wish to kill me, I hear,’’ he said.
“Killing is no new thing to me. I have offered myself to be killed many times, when I no more deserved it than I do now. Some of you, I think, have been with me in those days… I am here to preserve the peace and honor of this State, until the rightful government is seated—whichever it may be, it is not for me to say. But it is for me to see that the laws of this State are put into effect, without fraud, without force, but with calm thought and sincere purpose. I am here for that, and I shall do it. If anybody wants to kill for it, here I am. Let him kill!” (Pullen 1999)
The commander who then threw open his coat to the mob in front of the Maine State House in 1880 was Joshua Chamberlain, who had commanded the 20th Maine Regiment that saved the Union line from being flanked at the Little Round Top during the Battle of Gettysburg. In a made-for-Hollywood moment, a veteran in the crowd then jostled through to the front saying “By God, old General, the first man that dares to lay a hand on you, I’ll kill him on the spot.” The crowd then melted away (Pullen 1999).7 Having demonstrated through (often immensely) costly sacrifice their willingness to give up their private wellbeing for the good of the country, it is not surprising that heroes, whether demonstrating sacrifice through war or through nonviolent civil disobedience, are sought after to represent the people in politics, and are trusted when they endorse a political position.8 This source of legitimacy can and has been used to sway opinion and strengthen democracy. Heroes’ credentials for courage can also help them reach out and make peace.9 When embedded in networks of others with such shared identities, this can lead to a potent group of politically engaged and organised individuals who can support one another in diffusing democratic values and pro-social behaviour.10 But those whose patriotism is hard to question may also use this credential to expand the ‘Overton window’ and legitimise views that would otherwise be proscribed and beyond the pale of public discourse. As France’s experience shows, such networks can also be manipulated and lead to reinforcing incentives that can entrench extreme positions over time and make even those who do change their minds subject to sanctions and pressure from others within the network. This makes it harder for individuals to reverse course, as some have tried to do in the days after the Capitol was stormed (NYT 2021). History teaches us the power of leveraging the legitimacy of networks of heroes. These can have great destructive force. But they can also support democratic values and peace.
Besley, T, and T Persson (2019), “Democratic Values and Institutions”, American Economic Review: Insights 1 (1), 59-76.
Bekkouche, Y, and J Cagé (2018), “The Price of a Vote: Evidence from France, 1993-2014”, CEPR Discussion Paper no. 12614.
Bhavnani, R, and S Jha (2012), “Gandhi’s Gift: On the Promise and Limitations of Non-Violent Civil Disobedience”, Economics of Peace and Security Journal.
Cagé, J (2020), “Media Competition, Information Provision and Political Participation: Evidence from French Local Newspapers and Elections, 1944-2014”, Journal of Public Economics.
Cagé, J, P Grosjean, and S Jha (2020a), “Army of Shadows: Organizing Resistance in World War 2 France” working paper.
Cagé, J, A Dagorret, P Grosjean, and S Jha (2020b), “Heroes and Villains: The Effects of Combat Heroism on Autocratic Values and Nazi Collaboration in France,” CEPR Discussion Paper no. 15613.
Dippel, C, and A Ferrara (2020), “Oh Captain, My Captain: Leadership and Team Cohesion in the Union Army”, UCLA working paper.
Gay, V (2017), “The Legacy of the Missing Men: The Long-Run Impact of World War I on Female Labor Force Participation”, mimeo.
Horne, A (1962), The Price of Glory: Verdun 1916, London: MacMillan.
Jackson, J (2001), France: The Dark Years, 1940-1944, Oxford University Press.
Jha, S (2018), “Trading for Peace”, Economic Policy, 485-526.
Jha, S, and S Wilkinson (2012), “Does Combat Experience Foster Organizational Skill? Evidence from Ethnic Cleansing during the Partition of South Asia”, American Political Science Review, 106 (4), 883-907.
Jha, S, and S Wilkinson (2019), “Marching towards the Bastille: Veterans as Vanguards in the French Revolution”, mimeo, Stanford.
Jha, S, and S Wilkinson (in progress), Wars and Freedoms.
Lacroix, J, P-G Meon, and K Oosterlinck (2019), “A Positive Effect of Political Dynasties: the Case of France's 1940 Enabling Act,” CEPR Discussion Paper no. 13871.
Levitsky, S, and D Ziblatt (2018), How Democracies Die, New York: Broadway Books.
Lormier, D (2017), Les 100 000 collabos: Le fichier interdit de la collaboration française Documents, Cherche Midi.
Milgrom, P, and J Roberts (1990), “The Economics of Modern Manufacturing: Technology, Strategy and Organization”, American Economic Review 80 (3), 511-28.
Milgrom, P, Y Qian, and J Roberts (1991), “Complementarities, momentum and the evolution of modern manufacturing”, American Economic Review: Papers and Proceedings 81 (2), 84-88.
Millington, C (2012), From Victory to Vichy: Veterans in Inter-War France, Manchester University Press.
New York Times (2021), “Graham Swarmed by Trump Supporters at National Airport”, 9 January.
Pullen, J J (1999), Joshua Chamberlain: A Hero’s Life and Legacy, Stackpole Books.
Williams, C (2005), Pétain: How the Hero of France became a convicted traitor and changed the course of history, Palgrave Macmillan.
1 Some important features of the file are summarised in Lormier 2017. Since the file, while declassified, has not been made public, we are grateful to Dominique Lormier for allowing us access to the original documents.
2 These figures can be compared to the 405,399 military deaths the United States suffered during the entire Second World War, and the 22,654 soldiers killed on both sides in the bloodiest battle in US history, Antietam.
3 The interchangeability of line regiments was a common feature in other militaries as well, including the British army (see Jha and Wilkinson 2012).
4 See our companion paper, Cagé et al. 2020a.
5 Pétain’s deputy, Pierre Laval, was executed, along with a number of other high-ranking Vichy ministers. De Gaulle, who had served under Pétain in WW1, commuted Pétain’s own sentence of execution to life imprisonment in recognition of Pétain’s military contributions in WW1. Pétain’s death in prison in 1951 sparked demonstrations in most French cities, orchestrated by veterans of Verdun (Williams 2005, p.271). On how complementarities can induce momentum and persistence in technological adoption and organisation, see Milgrom et al. 1991 and in political and social institutions, Jha 2018.
6 See Cagé 2020 and Bekkouche and Cagé 2018 for details on the post-war electoral data.
7 On the effects of leadership on unit cohesion in the civil war, including under Chamberlain’s command, see also Dippel and Ferrara 2020.
 On the role of punishment for nonviolent actions acting as a screening device for political leadership, see Bhavnani and Jha 2012.
 It is a common observation in international relations that politicians from relatively hawkish parties are, ironically, often better positioned to make politically risky overtures for peace with long-standing adversaries than their dovish counterparts. One example is Nixon's famous rapprochement with China. However, our interpretation resonates with the greater set of options available to war heroes to shape politics regardless of party. For example, Yitzhak Rabin, a commando in Israel's war of independence who rose to be the Army chief during Israel's victory in the 6 Day War, was also able to pursue the Oslo Peace Accords as head of the centre-left Israeli Labour party.
 For example, French veterans of the American war of independence played an important role in spreading and defending democratic values in the constitutional phase of the French Revolution (Jha and Wilkinson 2019, and in progress).