The unintended consequences of parliamentary involvement: Elite collusion and Afghanistan deployments in Canada and Germany

Introduction

Legislative control of military deployments has been the subject of notable debate recently. Many contributors in these discussions have assumed that increasing parliamentary control of deployments provides a net democratic benefit. ¹ Having legislators vote on operations, it is argued, increases democratic legitimacy, augments debate, and provides a check on the executive (Dunn, 2007; Granatstein, 2009; House of Lords (HL), 2006; Joseph, 2013; Lord, 2011). Although there is variation among the formal institutional structures of democratic systems and the level of control legislatures exercise (Dieterich et al., 2010; Wagner et al., 2010), ensuring parliamentary votes on these decisions is understood to produce largely positive results. On the other hand, critics of greater parliamentary involvement argue that security policy often requires executive discretion, swift decision-making, and secrecy to be effective. The latter position has a long pedigree in political theory and has repeatedly been brought up in these debates (cf. Damrosch, 2003; Ku and Jacobson, 2003; Peters and Wagner, 2011).

Reflecting the aims of the special issue (Mello and Peters, this issue), this article explores the effects of parliamentary participation on security policy. The article questions the notion that parliamentary votes on military deployments necessarily produce positive deliberative and democratic outcomes. Analysing two ‘most-different cases’, Canada and Germany, we find that these votes yielded unintended consequences that ran counter to the intent of establishing greater parliamentary control and democratic deliberation. In examining their missions to Afghanistan, we find that elite collusion allowed governments to ignore public opinion and media scrutiny. Even though majorities in both countries opposed the war, mandates were continually renewed and troop numbers increased. We argue that parliamentary votes incentivised political parties to cease debating the aims, costs, and benefits of the deployment, reducing overall parliamentary scrutiny and questioning.

While an important strand in the literature claims that elite consensus on military operations tends to strengthen public support (e.g. Berinsky, 2007; Zaller, 1992), we find no such evidence for the involvement of Canada and Germany in Afghanistan. Major parties consistently supported the missions, whereas public opinion deteriorated over time and from 2009 onward majorities in both countries opposed involvement in the war. This resonates with the argument by Kreps (2010: 191) that elite consensus ‘inoculates leaders from electoral punishment’ and allows decision makers to defy public opposition and media scrutiny.

In explaining these outcomes, the article relies on rationalist institutionalism. We argue that deployment votes were approached in an instrumental fashion by political actors. Governing and opposition parties focused on how the votes would benefit them politically, rather than seeing parliamentary control as an inherent democratic good. The deployment votes deterred actors from working towards the normative goal of the votes: stronger parliamentary debate and legislative scrutiny. This resulted in decisions that were removed from public opinion and weakened legislative scrutiny of executive decisions pertaining to the war.

The article offers three novel contributions to the study of parliamentary war powers and legislative scrutiny of military operations. First, the article highlights how the incentives of parliamentary actors can undermine the normative goals of having parliament vote on military operations. Second, the article finds that these incentives can be seen across two notably different democratic systems. And third, the article’s findings suggest that rationalist institutionalism merits as much attention as regime type comparisons and normative theories in the study of parliamentary war powers and their effects.

We begin with a discussion of the parliamentary war powers debate and assumptions on how legislative voting affects political dynamics surrounding military deployments. Next, the article outlines how rationalist institutionalism offers an alternative account of how actors will behave before and after these votes. Third, the article applies a rationalist account to the German and Canadian parliamentary votes on Afghanistan. We conclude with a discussion of how its findings challenge existing assumptions about the role of parliaments in military decision-making.

Debating parliamentary involvement in security policy

Parliamentary influence on security policy has received increased attention in recent years (Raunio and Wagner, 2017). A new strand of literature on ‘parliamentary war powers’ shows that democracies are characterised by substantial variance in the formal institutional legislative control and oversight of military deployments (Born and Hänggi, 2005; Ku and Jacobson, 2003; Mello and Peters, 2018; Peters and Wagner, 2011). Moreover, some countries have seen a trend towards legislative votes on military deployments, despite the absence of formal requirements (Lagassé, 2017; Mello, 2017; Strong, 2015, 2018). These studies provide a critical insight into democratic foreign policy processes and identify sources of variation among democracies neglected in previous work. Notably, parliamentary war powers are analytically closer to military deployment decisions than abstract indicators of ‘institutional constraints’ that merely differentiate between regime types. While there is substantial work on the conflict behaviour of parliamentary and presidential democracies (e.g. Leblang and Chan, 2003; Reiter and Tillman, 2002), few have investigated the concrete involvement of legislatures in decision-making on the use of force. Yet, recent work shows that significant variation exists beyond the parliamentary–presidential distinction. Works in this vein suggest that it is rather the degree of legislative participation in deployment decisions that can, under certain preconditions, reduce war involvement (Dieterich et al., 2015; Kesgin and Kaarbo, 2010). The ‘parliamentary peace’ hypothesis suggests that, given a war-averse public, countries with wide-ranging parliamentary war powers abstain from military participation (Dieterich et al., 2015).

The most recent empirical test of the parliamentary peace yields ‘modest evidence’ for the expected relationship, qualified by the type of military mission (Wagner, 2018). This resonates with Haesebrouck (2016: 15) who finds the absence of a parliamentary veto a ‘core causal condition’ towards military participation in the fight against Daesh. While Dieterich et al. (2015) show a correlation between war powers and non-involvement in the Iraq War, others have demonstrated that this pattern might have been unrelated to parliamentary involvement and rather a matter of constitutional restrictions (Mello, 2014).

The discussion about the effects of parliamentary deployment votes is less nuanced in public debates and parliamentary studies. As committee reports from the British Parliament highlight, there are strong assumptions that granting the legislature a greater role in military deployment decisions will lead to enhanced parliamentary and public accountability, stronger democratic legitimacy and deliberation, and a necessary check on executive power (Bolt, 2015; HL, 2006; House of Commons (HC), 2004). The 2013 House of Commons vote against British military involvement in Syria strengthened these claims. Involving parliamentarians in military deployment decisions is assumed to produce net positives, save for concerns about limiting the executive’s ability to respond to crises and emergencies, and possibly subjecting military operations to judicial review (Lagassé, 2017).

A rationalist-institutionalist account

Rationalist institutionalism offers another perspective on parliamentary deployment votes. It makes three claims that are of value here. First, the school argues that political actors maximise utility, meaning that they are motivated by a desire to fulfil their preferences (Peters, 2012; Shepsle, 2008). The content of these preferences will vary depending on the actor and institutional setting, but rationalism is united in the view that their behaviour is driven by efforts to achieve them. Second, actors seek to fulfil their preferences in settings where they can either complement or conflict with the preferences of others. Actors take others into account to identify ways of working with those who have complementary preferences and prevail over those with whom they are in conflict. This means that actors act strategically and exercise foresight (Shepsle, 2010: chapter 6). Third, rationalist accounts recognise that rules can lead to suboptimal or perverse outcomes. Certain institutional designs and incentive structures can encourage actors to achieve their preferences through manipulation or deception (Riker, 1986). Institutional designs or reforms that ignore preferences or incentive structures may not achieve their stated aims or may create ‘unanticipated consequences’ (Pierson, 1996: 136).

For individual politicians, utility maximisation can involve re-election, promotion, and power, and maintaining one’s reputation (Dewatripont et al., 1999; Fenno, 1978; Fiorina, 1987; Frank, 1987; Müller and Strøm, 2008; Schlesinger, 1966). Political parties can also be treated as unified, self-interested political actors. Indeed, in parliamentary systems with strong party discipline and centralised party decision-making, it can be fruitful to treat parties as single actors. Work on party preferences has established how these political actors can vary in terms of how they understand their utility. Strøm (1990) has categorised parties based on whether they are vote-seeking/office-seeking, or policy-influencing. Vote-seeking parties aim to increase their electoral prospects and policy-seeking parties look to shape government decisions.

Vote-seeking parties will be particularly mindful of how their policies and decisions reflect public opinion. When vote-seeking parties act in ways the public supports, they will highlight their responsiveness to popular opinion. But when they act contrary to prevailing opinion, they will seek ways to shield their decisions from scrutiny or attention. Vote-seeking parties will react in a similar manner to media scrutiny. Policy-seeking parties, on the other hand, will aim to turn public opinion and media attention towards their preferred policy outcome. They therefore have an incentive to draw attention to their policy proposals, in the hope of generating support for their ideas or converting sceptical voters and media to their point of view.

Turning to legislative approval of military operations, our application of a rationalist approach leads us to the following expectations. First, regardless of whether a legislature is formally or informally involved in sanctioning military action abroad, political actors will approach the votes from a utility maximising perspective. While the rhetoric surrounding the votes may appeal to democratic ideals and the importance of parliamentary scrutiny (cf. Müller and Wolff, 2011), legislative actors will approach the votes strategically.

Second, when parties with vote-seeking preferences endorse a military deployment, their members are expected to gravitate towards less contentious debates after the matter has been settled. The debate diluting effect of a consensual legislative vote will be especially noticeable when those who voted in favour face a public that is sceptical or opposed. In such cases, vote-seeking political actors have an incentive to collude to avoid discussing the mission. When vote-seeking parties collude in this way, they can present a common front in the face of negative public opinion or media scrutiny. This effect will be further amplified when there is an election on the horizon. Under this situation, vote-seeking parties that supported the deployment will collude to keep the mission out of the campaign, out of the media’s coverage, and hence out of public debates.

Third, policy-seeking parties will aim to keep contentious debates over a military deployment alive after a vote, particularly if they opposed it. However, their ability to achieve sustained legislative scrutiny will depend on their relative strength in parliament; smaller parties are expected to have more difficulty focusing parliament’s attention on a deployment they oppose, particularly if larger parties support the mission. Fourth, individual political actors may occasionally vote against their party or parties may not be able to prevent their members from splitting into blocs for or against a military deployment. In such instances, the preferences of individual politicians will explain their behaviour, but the impact will depend on how other actors react.

Regardless of democratic system type, therefore, we hypothesise that military deployment votes incentivise vote-seeking political actors to engage in collusion, leading to less contentious parliamentary debate and legislative scrutiny when vote-seeking parties endorse a military deployment. Hence, parliamentary votes do not necessarily strengthen parliamentary debate and legislative control of military operations. While policy-seeking parties and individual politicians may attempt to sustain parliamentary debate about the deployment, their impact will vary depending on their relative influence in the legislature and how other parties respond to their efforts. Similarly, negative public opinion and media scrutiny may not encourage electoral accountability from parties who voted in favour of a mission, since they will share an incentive to not talk about the operation. When this collusion occurs, parties will engage in less democratic deliberation and be less responsive to negative public opinion and media scrutiny.

Research design and methods

To analyse the unintended consequences of parliamentary votes, we select Canada and Germany as two ‘most-different’ cases (Gerring, 2008). These cases differ in many dimensions, including electoral rules, political culture, government type, and, importantly, when it comes to formal parliamentary involvement. Canada’s House of Commons is elected under a single member plurality system that typically results in single party majority or minority governments, and the Canadian legislature is not formally required to approve or vote on military deployments. Germany’s Bundestag is elected via a mixed-member proportional system that produces coalition governments, and the German legislature is formally required to approve military deployments. Yet, both countries witnessed similar degrees of elite collusion, which led to an overall reduction of parliamentary scrutiny and a disconnect from public opinion. We argue that the similarities between Canada and Germany reinforce rationalist arguments that incentives and preferences can take precedence over historical structures and societal/political norms in explaining institutional outcomes.

The article primarily draws on a qualitative assessment of parliamentary votes and political decision-making on the Canadian and German Afghanistan missions. This is complemented with quantitative indicators of parliamentary activity, public opinion, media coverage, and troop levels. Parliamentary data were gathered from the Parliament of Canada’s official record and the German Bundestag Documentation System. The supplementary document contains all data used for this article.

Parliamentary votes in Canada

Parliament’s role in military deployments has been a long-standing source of debate (Hillmer and Lagassé, 2016), as has the legislature’s relative weakness towards the executive in defence affairs generally (Bland and Rempel, 2004). Looking at Parliament’s role during the war in Afghanistan specifically, Stephen Saideman (2016) found that the legislature was poorly placed to exercise effective oversight of the mission, that parliamentarians were easily distracted by secondary issues related to the deployment, and that there was little desire on the part of Members of Parliament to improve this situation. ² However, it is noteworthy that Parliament’s involvement in Canadian military affairs was deliberately increased during the Afghan war. Although the government was not required to consult the House of Commons to extend Canada’s mission there, the executive chose to do so. Rather than constraining the government, these votes gave the executive greater freedom of action (Hillmer and Lagassé, 2016). As important, these votes were followed by less parliamentary deliberation after they were held.

During Canada’s 2006 election, the Conservative Party pledged to ‘Make Parliament responsible for exercising oversight over […] the commitment of Canadian Forces to foreign operations’ (CPC, 2006: 45). The political incentives and preferences held by the Conservatives explain why they championed parliamentary war powers. Previous Liberal governments had been criticised for ignoring Parliament and overly centralising decision-making in the Prime Minister’s Office (Hillmer and Lagassé, 2016). Emphasising parliamentary oversight of military operations linked a high-risk international mission with discontent about executive dominance. With increased media attention of the Canadian mission in Afghanistan that followed the Liberal government’s decision to deploy the military to Kandahar, Afghanistan in 2005, stressing the need to increase parliamentary oversight allowed the Conservatives to differentiate themselves from their principal opponents about a mission that they supported and that attracted steady media attention (Figure 3).

The Conservatives carried the 2006 election, though only as a minority in the House of Commons. This result meant that voters would head back to the polls before long. Major parties in the Canadian House of Commons face incentives to engage in vote-seeking behaviour as a general rule (Lagassé and Saideman, 2017), and this propensity was amplified by relatively unstable minority government situation. To this end, Stephen Harper, the new Prime Minister, visited Canadian soldiers in Afghanistan shortly after Election Day. He expressed his unwavering support for the mission during the trip. Upon his return, Harper declared that he planned to extend the Kandahar mission. The Liberals had committed the Canadian Forces to a 2007 end date, but Harper decided the deployment should go on to 2009. In keeping with his party’s electoral platform and past effort to increase the role of the Commons, he announced that the extension would be brought to the House for a vote.

Harper’s incentives for holding the vote were threefold. First, he could fulfil his party’s electoral pledge to improve the Commons’ role in military deployment decisions. Second, the vote would allow Harper to further highlight his government’s support for the Canadian military and their deployment to Afghanistan. Third, the vote promised to divide or embarrass the Liberal Party. Since the Liberals had initiated the Kandahar mission when they were in power, they would be ill-placed to oppose an extension; yet, a number of Liberal parliamentarians were opposed to an extension. The left and right wings of the party were split over the mission, and with only an interim leader following the election, the divide could be exploited. If the Liberals voted against the extension, they could therefore be made out to be hypocritical, and if the party split, it would highlight divisions and incoherence within the party (Jockel, 2014: 76).

The Liberals, for their part, were placed in a disadvantageous position by Harper’s pledge to hold a vote. On the one hand, the right wing of the party and those individual members of Parliament (MPs) who had previously supported the Kandahar mission faced the prospect of a damaged reputation if they voted against the extension. On the other hand, if they voted with the government, their ability to criticise the Conservatives for their handling of the mission would be hampered thereafter. Canada’s other main opposition party, the New Democratic Party (NDP), faced no such dilemma. They were opposed to the mission from the outset.

Parliament was summoned on 3 April 2006. Harper’s extension was vigorously discussed until the vote was held on 17 May 2006. In these 6 weeks, the Afghanistan mission was mentioned 139 times in the House. On the day of the vote, the 103-member Liberal party caucus, with 30 high-ranking MPs, including the interim leader, voted in favour of the extension. Thanks to their defection, the vote narrowly passed 149 to 145.

Having been outflanked and divided by the Conservatives, the Liberals had an incentive to remain largely silent about the Afghan war following the vote. From 18 May 2006 to the end of the year, the official opposition only raised the mission 6 times in the House of Commons. The New Democrats, in contrast, raised the topic 15 times, and the Bloc Québécois, the separatist party from French-speaking Quebec, addressed the mission 9 times as well. The governing Conservatives, on the other hand, discussed the mission 15 times, either in response to the New Democrats and to highlight mission successes. Overall, from late May to the end of December 2006, mentions of Afghanistan fell to 45, far fewer than in the weeks leading up to the vote (Figure 1).

OPEN IN VIEWER

Figure 1.

Speeches on Afghanistan in the Parliament of Canada (2003–2014).

Debate about the future of Canada’s mission in Kandahar began to amplify again in April 2007. In December 2006, the Liberals had selected Stéphane Dion as their leader, a former minister from the party’s left wing who had voted against the extension in 2006 and was opposed to a further prolongation (Jockel, 2014: 76). Furthermore, Canadian casualties had begun to grow and there was growing public and media concern that the mission was facing difficulties (Figures 2 and 3). For the first time since the 2006 election, all three opposition parties were engaged in pointed critiques of the mission and questioning the government about an exit strategy. The Conservatives seemed vulnerable on the issue, making it an attractive target for opposition parties focused on the next election. From April to October 2007, the Afghanistan mission was mentioned 322 times in the House and the issue of an exit strategy coming up 47 times. In contrast to their behaviour in 2006, the Liberals prompted 112 of those mentions, whereas the New Democrats were involved in 38 exchanges and the Bloc Québécois in 39. Thought they were on the defensive, the Conservatives raised the operation 141 times, as they tried to vaunt the mission’s successes.

OPEN IN VIEWER

Figure 2.

Canada: International Security Assistance Force (ISAF) troop deployments and public support (2003–2014).

OPEN IN VIEWER

Figure 3.

Canada: media coverage (2003–2011).

By the summer of 2007, Harper was determined to further extend Canada’s deployment in Kandahar. However, given his government’s minority standing in the House and the possibility of an election in the near term, he realised that making the decision without having the Liberals onside was politically risky (Hillmer and Lagassé, 2016; Saideman, 2016). When the next election came, it was in the Conservatives’ interest to have the Liberals publicly supporting their policy on Afghanistan. If the Conservatives were alone in supporting the Afghan war, or if the government extended the mission without a supporting vote in the Commons, they alone would shoulder responsibility for the mission and possibly suffer at the polls, given that the operation was losing popular support and media scrutiny remained strong (Figures 2 and 3). The Conservatives’ vote-seeking preferences demanded that they either abandon the idea of a further extension or find a way to get the Liberals to openly endorse the policy.

The Conservatives’ strategy to co-opt the Liberals involved the establishment of an independent panel chaired by former Liberal minister John Manley (Jockel, 2014: 77–80; Saideman, 2016: 44–48). On 20 January 2008, the Panel released its report and recommended that the government extend the mission indefinitely (Independent Panel on Canada’s Future Role on Afghanistan). Armed with this recommendation, Harper approached the Liberals about a compromise. Harper and Dion settled on a final 2-year extension to 2011, to be voted on in March 2008. The reason why the Liberals agreed to the further extension remains speculative. However, from a rationalist perspective, a case can be made that the party feared another division when the matter was brought to the House. Indeed, Manley’s views on the need to remain committed to the mission likely resonated with a good number of Liberal MPs. The Liberal Party would not be particularly well-placed to campaign against the Conservatives’ Afghanistan policy if a sizable number of Liberal MPs voted with the government, and Dion’s standing as party leader would be weakened going into an election if he appeared unable to unify his caucus. Finally, those Liberals opposed to the mission could take solace in the fact that the compromise involved a last, 2-year extension rather than the open-ended extension recommended by the Manley Panel.

The months between the establishment of the Manley Panel and the lead up to the vote on the second extension saw an explosion of debate on Afghanistan in the House of Commons. The mission was mentioned 763 times between October 2007 and March 2008. Intent on explaining and justifying the extension, the Conservatives prompted 382 of these mentions. The Liberals followed with 197 mentions, while the New Democrats were involved in 123, and the Bloc Québécois in 83. On 13 March 2008, the motion to extend the mission carried 198 to 77, with the Conservatives and nearly all Liberals voting in favour, ³ and the New Democrats and Bloc Québécois voting against.

From the spring to fall of 2008, when Parliament was dissolved for an election, discussions on Afghanistan fell sharply in the Commons. The topic was broached 271 times when the House sat during this time. The Conservatives were involved in 119 of these mentions, defending the mission, pointing to progress, and reminding critics that the House voted in favour of a second extension. The Liberals raised the war 93 times during this period, honouring wounded and fallen soldiers, and asking for clarifications from the government and offering soft critiques of the Conservatives’ handling of the mission. The New Democrats continued to criticise the mission, though their interventions fell to 39 during this period. Similarly, the Bloc Québécois largely abandoned the topic, discussing the matter only 26 times.

Collusion between Conservatives and the Liberals following the vote ensured the Afghan deployment was discussed and debated little during the 2008 general election (Massie, 2016). While the NDP and the Bloc Québécois continued to critique the mission, and Canadian media kept the story alive, the two major parties focused on other issues. Although Canadians were increasingly sceptical of the mission (see Figure 2), the March 2008 vote ensured that neither the Conservatives nor the Liberals had an incentive to highlight the war. Instead, the vote led them to collude to keep the topic off the table as they both sought to carry the election.

The Conservatives were re-elected with a slightly larger number of seats in 2008, but not a parliamentary majority. Over the next year and a half, the NDP continued to lead opposition debates about Afghanistan, owing to the relative silence produced by the Liberal and Conservative collusion. Finally, in late 2010, Harper revealed that Canada would deploy the military on a training mission to Kabul after the mission in Kandahar came to an end the following year. Although the NDP protested that Harper was reneging on his pledge to withdraw from Afghanistan in 2011, the Liberals supported the Conservatives once again, ensuring that the two major parties colluded to keep debate to a minimum. The Liberals and Conservatives agreed that a vote was not necessary in this instance. Both parties sought to avoid even the short-term spike in debate that typically led up to the votes. The Kabul training mission began in May 2011 and ended in March 2014. This was Canada’s last deployment to International Security Assistance Force (ISAF).

In summary, Canada’s Conservative government introduced the practice of parliamentary deployment votes and made strategic use of the practice in 2006 and 2008. The Conservatives presented the votes as an effort to strengthen parliamentary war powers and to enhance legislative control of the executive (Hillmer and Lagassé, 2016). Yet, the votes resulted in fewer debates about the war in Afghanistan in the House of Commons (Figure 1). The votes deterred the official opposition from debating the mission or scrutinising the government after they took place. In May 2006, the vote exposed a rift in the Liberal Party ranks, leading the party to regroup and refocus for the remainder of the year. Part of that recalibration included keeping relatively quiet about Afghanistan, lest their apparent incoherence and divisions come to the fore again. In 2007, under a new leader, the Liberals began debating and scrutinising the government about Afghanistan again. With public support for the war waning and faced with sustained media scrutiny, the Liberals’ vote-seeking incentives encouraged a stronger critical stance. The Conservatives responded by targeting the Liberals’ reputational concerns and solidarity again. By establishing a panel headed by a former Liberal minister, the Harper government sought to compel the Liberals into supporting a further extension of the Kandahar mission. Faced with the prospect of another vote that would split their caucus, and anticipating an election in the near future, the Liberals opted to align with the Conservatives. The two parties voted together on a motion that outlined a 2011 termination date for the Kandahar deployment. Following the vote, both the Liberals and the Conservatives shared a vote-seeking incentive to keep talk of the mission to a minimum because the public were turning against the mission. This left the role of scrutinising and debating the deployment in Parliament to the NDP and Bloc Québécois, the two small opposition parties. It also meant that when Harper called an election in September 2008, both the Conservatives and the Liberals made a point of not highlighting the war (Massie, 2016). Their collusion led to less parliamentary and electoral debate, despite the mission’s increasing unpopularity. Notwithstanding the votes, the relative weakness of Canada’s Parliament in military affairs was a striking feature of the war (Saideman, 2016).

Parliamentary votes in Germany

Parliamentary war powers in Germany go back to a seminal ruling of the constitutional court in 1994, which led to a deployment law that became effective in 2005. In line with the ruling and its conception of the Bundeswehr as a ‘parliamentary army’, all major military deployments have since been placed before parliament for up or down votes, ⁴ which has led observers to categorise the Bundestag as an ‘exceptionally powerful and active parliament in controlling the deployment of armed forces’ (Wagner, 2017: 60). ⁵

Germany’s involvement in Afghanistan started in November 2001, when the Schroeder government of Social Democrats and Greens partook in Operation Enduring Freedom (OEF) with 3900 soldiers (including 100 special forces in Afghanistan) and, 1 month later, deployed 1200 troops to ISAF in Kabul. The mandates for these operations were renewed numerous times and ISAF was enlarged to an eventual force of 5350 soldiers. In 2008, the Bundestag ended the OEF involvement in Afghanistan, whereas ISAF was prolonged until the end of 2014. With its extensive timeframe, Germany’s military engagement in Afghanistan has become the largest and most costly military operation in the history of the country’s post-1945 armed forces with estimated expenses of 8.8 billion Euros for ISAF alone (Thiels, 2015).

Apart from the initial vote on OEF, which came down to a 10-vote difference, all other mandate decisions—a total of 26 parliamentary votes between 2001 and 2014—received overwhelming majorities in parliament. The largest number of individual speeches was given in December 2001, when the initial ISAF mandate was decided upon (56 speakers). Afterwards, there has been a slight decrease in parliamentary activity on Afghanistan until late 2006, when the number of speeches picked up again noticeably. While the Kunduz airstrike of 4 September 2009 marked a watershed for Germany’s Afghanistan policy, the frequency of parliamentary speeches did not change substantially (see Figure 4). From 2011 onward, parliamentary activity decreased markedly.

OPEN IN VIEWER

Figure 4.

Parliamentary speeches on Afghanistan in the German Bundestag (2003–2014).

Between 2001 and 2014, the Bundestag contained four to five parliamentary parties, depending on the legislative term. While the political decisions on military involvement in Afghanistan were made under a leftist coalition of Social Democrats and Greens, the mandates were renewed under the succeeding conservative–centrist governments of Chancellor Merkel, including two Grand Coalitions of Conservatives and Social Democrats (2005–2009, 2013–ongoing) and a Conservative–Liberal coalition (2009–2013). This means that all major parties except the socialist The Left have been in government at one time or another during the Afghanistan campaign and shared responsibility for the military deployments.

The initial debate about German involvement in Afghanistan nearly spelled the end of the red–green coalition. The parliamentary vote on 16 November 2001 resulted in the closest vote on military deployments in the history of the Bundeswehr, with 336 to 326 parliamentarians voting in favour of the mandate. Schroeder had set the tone with his declaration of ‘unconditional solidarity’ with the United States in the face of the 11 September 2001 terrorist attacks (Bundestag, 2001: 182–193). Following 9/11, it became clear that unconditional solidarity included military participation in the fight against terrorism. However, the pacifist wing of the Greens and a sizable group of Social Democrats were strongly opposed to any involvement in OEF, which the United Kingdom had initiated together with the United States on 7 October 2001. Twelve Green MPs and four Social Democrats threatened to vote against the motion, which meant that there was no governmental majority (Weinlein, 2011).

To rein in dissenters from within the red–green government, Schroeder combined the decision to deploy armed forces to OEF with a parliamentary vote of confidence. Since 1949, there have been five confidence votes and the 2001 incident was the only such motion combined with a decision on the use of force. Intended to secure a ‘Chancellor’s majority’, the confidence vote meant that all opposition MPs would vote against the motion, even though Conservatives and Liberals had already expressed their support for the fight against terrorism and would have supported a simple vote call. Forced by Schroeder’s vote of confidence, the Green Party leadership around Foreign Minister Fischer faced the dilemma of reconciling antimilitarist sentiments among the party base and substantial concerns regarding the US-led military operation with the party’s desire to remain in government. To Fischer (2011: 49), it was ‘political suicide’ to abstain from military involvement, as parts of his own party preferred. Several Green critics published a ‘position paper’ that expressed strong doubts about the supposed mission in Afghanistan (Buntenbach et al., 2001). Ultimately, the critics decided on continued support for the Chancellor, but signalled their disagreement with the military deployment by splitting their votes, which resulted in four Greens voting against the proposal. The Social Democrat Loercher decided to leave the parliamentary group and vote against the motion as an ‘unaffiliated’ MP (Bundestag, 2001).

The politics surrounding the initial OEF mandate demonstrate that Social Democrats and Greens adopted a strategic approach, primarily driven by vote-seeking rather than policy concerns. Indicating the political controversy of the Afghanistan mandate, 74 MPs from the governing coalition submitted written statements for the parliamentary record in which they explained the reasoning behind their own vote. Nonetheless, the fact that the Social Democrats, apart from one MP, voted entirely in favour of the motion and the Greens decided to display their concerns without seriously risking their position in government shows that vote-seeking behaviour took precedence. Four of the authors of the Green position paper voted against their expressed policy position, solely to remain in government. ⁶ The combination of a confidence vote with a substantive motion effectively undermined parliamentary war powers, as MPs had to decide between remaining in government or following their own conscience. Unlike in Britain where institutional changes were made to prohibit the combination of confidence votes and substantive motions, the German system still allows for this option. ⁷

Public support for German military involvement in Afghanistan has experienced a slow but steady decline between 2001 and 2014 (for a timeline of the North Atlantic Treaty Organisation (NATO) mission, see Figure 5). At the time of the initial deployment decisions, about 56% of the public supported the engagement (Mello, 2014: 123). Through 2006, there was still a majority in favour of the operations but since then a plurality of respondents spoke out against military involvement in Afghanistan (Mader, 2017: 168). In 2012–2013 almost 60% of the public were undecided or against German military involvement in ISAF (ZMSBw, 2014: 52). These trends in public opinion contrast markedly with parliamentary votes on the Afghanistan mission, which—apart from the initial mandate on OEF—continuously gained 70% or more support from Bundestag MPs (cf. Schoen, 2010). Moreover, there is an evident disconnect between public opinion and military involvement in Afghanistan. While public support diminishes over time, troop levels are continuously increased, peaking at about 5000 soldiers in 2011 (see Figure 5).

OPEN IN VIEWER

Figure 5.

Germany: ISAF troop deployments and public support (2003–2014).

One particularity of the German deployment was the placement of significant restrictions on military commanders and troops on the ground in Afghanistan. Most of these were formulated in the parliamentary mandates, which were themselves the result of negotiation and consultation between the major parties. But sometimes coalition politics led to the imposition of additional restrictions that were then added to the mandate as a formal note. This included severe limitations of permissible tasks for the deployed soldiers, a geographical restriction to the northern part of the country (Kunduz district), after the extension of ISAF, a strict separation of OEF and ISAF mandates, and the abstention from counter-narcotic operations. ⁸ As Fischer explained when the initial ISAF mandate was passed, the planned mission was a ‘peace operation’ solely aimed at stabilising the interim Afghan government in Kabul and the surrounding area (BT 2002, Dec. 22). By implication, OEF was thus the ‘combat mission’ but it received substantially less mention by decision makers in Berlin. This distinction characterised German politicians’ approach towards Afghanistan for many years. In November 2008, the Bundestag eventually decided to end OEF participation in Afghanistan, though Germany continued to support anti-terror operations elsewhere. The imposed caveats are another unintended consequence of parliamentary votes. The main parties clearly acted as vote-seekers who sought to shield the mission against potential negative media coverage and public opinion. Rather than trying to reach specific policy goals in Afghanistan, parties placed heavy restrictions on the military to reduce the risk of electoral backlash in the wake of undesirable incidents.

A critical event occurred in the early morning hours of 4 September 2009, when the German commanding officer of the Provincial Reconstruction Team Kunduz, Oberst Klein, ordered an airstrike against gasoline trucks that had been abducted by Taliban forces and were stuck on a sandbank in the Kunduz River. This attack caused the death of insurgents but also killed civilians including many underage boys, some of whom had apparently been forced to help syphon gasoline from the trucks. Separate investigations reported between 99 and 125 deaths (Bundestag, 2011). The ‘Kunduz’ affair single-handedly changed the domestic debate about Afghanistan as it became clear that ISAF was not simply a ‘peace operation’ but that the country was engaged in war-fighting in a hostile environment. Figure 6 shows the quantitative effect on German media coverage, which reached an all-time high after Kunduz. While Colonel Klein faced no legal consequences for his actions, the political-military leadership in Berlin was in turmoil. First, Defence Minister Jung, who had by then moved to another ministry, had to step down as a government minister, partly because of his inadequate handling of the affair. His successor in the defence ministry, zu Guttenberg, decided to relieve the country’s top military and political leaders, Chief of Defence Schneiderhan and Deputy Minister of Defense Wichert. Both had apparently withheld critical information from their superiors. On the upside, the ‘post-Kunduz shake-up’ led to a process of generational change among top-level decision makers, as noted by Rid and Zapfe (2013: 210).

OPEN IN VIEWER

Figure 6.

Germany: media coverage (2003–2014).

According to parliamentary practice, mandates for military operations are usually revisited in the Bundestag on a yearly basis. However, for operations in Afghanistan, it is evident that electoral concerns and vote-seeking behaviour affected the timing of parliamentary debates. Two cases stand out. In October 2008, the major parties agreed to prolong the mandate for an additional 3 months to keep Afghanistan out of the upcoming general election campaign. The Merkel government justified this unusual step arguing that the decision to prolong the next mandate should be given to the newly elected legislature (Bundestag, 2008: 7). Nonetheless, many commentators saw it as an attempt to keep the increasingly unpopular Afghanistan engagement out of the election campaign (Brummer and Fröhlich, 2011: 16). A similar case happened in 2010, when it became public that the government intended to let the Bundestag approve a renewal of the ISAF mandate 6 weeks earlier than originally planned for. Allegedly, Merkel had insisted on an earlier date to keep the Afghanistan issue from regional elections in three German Länder in early 2011 (Spiegel, 2010). Both episodes indicate that political decision makers adopted a strategic and instrumental approach to parliamentary votes, typical of vote-seeking parties, rather than seeing the parliamentary votes as a forum for deliberation and debate, as policy-seeking parties would conceive of them.

To sum up, Germany’s military involvement in Afghanistan was characterised by consensus politics typical of coalition governments. Except for The Left, all parliamentary parties had been involved in the decision-making and shared government responsibility at one time or another between 2001 and 2014. This meant that all major parties were implicated due to their previous support for the respective mandates and they were not in any position to seriously question or criticise government policy. Due to parliamentary and coalition politics, decisions on mandates and operational aims had to be agreed-upon first within the governing coalitions and then among a majority in the Bundestag and its respective committees. In this sense, parliamentary involvement led to the anticipation of legislative preferences. The government, and especially the ministry of defence, regularly consulted with key members of relevant factions and legislative committees to find agreement on contentious policy positions before a mandate was finalised. This is one reason, unlike in the United Kingdom, why there has never been a veto against a suggested military deployment in Germany. However, parliamentary involvement also led to the creation of extensive caveats of which the operational restrictions in the mandates and formal notes are only the publicly available information. This caused problems for troops on the ground and their operability within the multilateral context of NATO. While there is little evidence for a debate dampening effect, parliamentary votes have become routine exercises that spark little political controversy. Moreover, the analysis shows that despite the formal empowerment of parliament, the executive still dominates the agenda. This power was exercised, for instance, to shift parliamentary votes away from regional or general elections. And if push comes to shove, the government can always combine a confidence vote with a substantive vote on a military deployment to enforce a governmental majority in parliament.

Conclusion

This study has argued that parliamentary involvement in military deployment decisions does not necessarily lead to greater democratic deliberation or debate. Two factors that weigh against greater deliberation and debate are vote-seeking tendencies by parties in the legislature and collusion on the part of these parties to insulate themselves from an electorate that it opposed to a deployment they support. The findings of this article reinforce the conclusions of the recent literature on parliamentary security policy. While normative arguments about the value of involving parliaments in deployments treat legislatures as unified, corporate bodies, it is more fruitful to see parliaments as partisan battlegrounds composed of self-interested political parties (Raunio and Wagner, 2017). Focusing on parties as opposed to parliaments as a whole helps explain why legislatures do not necessarily live up to the aspirations laid out by normative theorists: political parties may have an incentive to avoid deliberation and debate rather than engage in it. When parties are motivated by vote-seeking behaviour, for instance, their incentive to deliberate and debate an issue will be shaped by how the policy will affect their electoral prospects, rather than democratic principles. Moreover, as Lagassé (2010) has argued with respect to Canada, and Carr (2017) with respect to Australia, a bipartisan approach to foreign and defence policy comes with disadvantages, especially when parties agree about a policy that the public opposes. In such cases, as this study highlights, it may lead these parties to collude to avoid electoral accountability.

The effects of vote-seeking party incentives also call into question the normative position that more parliamentary deliberation should lead to better deployment decisions. If parliamentary parties are primarily concerned with the electoral consequences of their votes, the actual merits of a deployment may not feature prominently. This aligns with Saideman’s (2016) findings about the irrelevance of Canada’s Parliament during the Afghan war, notwithstanding the fact that the legislature was asked to vote on Canadian deployments in 2006 and 2008. For Germany, Brummer (2014) has argued that the strong formal powers of the Bundestag are not matched by its actual influence on military deployments. Furthermore, collusion on the part of parties who are in favour of a deployment can diminish debate about missions with serious flaws, as Massie (2016) notes. In these cases, it may be that debate and deliberation would be improved if parliament played a less direct role in deciding deployments, since parties who are not part of the government or governing coalition would be freer to critique a mission they actually support. Future research is therefore warranted to see whether parliamentary debate and deliberation are different when a legislature is not directly involved in deployment decisions.

Finally, this study questions critics’ concerns that involving parliament will slow or inhibit military deployment decisions. Provided that major parties are in agreement, bringing military missions to the legislature for votes of approval or support does not necessarily make it more difficult for countries to send their forces abroad. Here again, the issue is less the involvement of parliament per se as it is the incentives and behaviours of the parties that make up the legislature.