Democratic Backsliding and Party System in Interwar Japan

1. Introduction

Compared to Northern and Western European economies in the late 19th and early 20th centuries, Japan was a latecomer to democratisation, and its first democratic experiment was rather short-lived. After the First World War, the Taisho Democracy Movement decimated the power of Meiji oligarchs, and the Universal Male Suffrage Law in 1925 created an opportunity for these previously disenfranchised social groups to enter into party politics (Banno, 1982; Ito, 2002; Kitaoka, 1999). For the first time in Japan’s political history, a semi-democratic system, one that closely resembled early twentieth-century British liberalism, emerged. In this era, the two major political parties, the conservatives (the Seiyukai) and the progressives (the Kenseikai/Minseito), engaged in contested elections and alternated turns holding the premiership (Benson, 2001: 34; Banno, 2000: 108; Awaya, 1983).(1) 

The regime was by no means liberal democracy by today’s standards—rather it was either semi- or illiberal democracy at best. Indeed, during the interwar years, the protection of individual rights, including freedom of speech, assembly, and association, was not consistent enough to qualify as “liberal.” The regime, however, constructed a political space in which citizens could take part in contested elections with at least moderately-inclusive suffrage and the absence of major fraud. Despite the non-democratic nature of the Meiji Constitution, this political environment fostered democratic accountability, because after the Second Movement for Constitutional Government (Dainiji Kensei Yogo Undo) of 1924, it became a practice that premiership would automatically be given to the party that won a preceding election. This informal rule, constitutional normality (Kensei no Jodo), was particularly important because it provided political parties with a means to bypass the oligarchy (and by extension the Emperor) and translate their electoral victory into premiership, effectively preserving democracy’s majority principle (Banno, 2006).

From its inception, however, this democratic experiment faced a series of challenges, both political and military, from radical rightist groups centred around anti-democratic agendas. These groups’ challenges were motivated by and rooted in their interests in expanding Japan’s presence on the continent (Sakai, 1992: 46-7). Their aggressive expansionism, primarily against Manchuria, China, and the Soviet Union, initially called for rapid arms buildup, and later the installment of a controlled economy through national-social integration. Yet political parties and democratic institutions were major obstacles to these objectives. Under these circumstances, dismantling political parties and democratic governance itself became a central part of the military’s political strategies, and a recessionary crisis in the early 1930s provided them with opportunities to launch these attacks. In May 15, 1932, in the midst of a recession, Prime Minister Tsuyoshi Inukai was assassinated by hard-line officers in the Navy, and Japan’s young semi-democracy entered into a phase of profound instability (Kitaoka 1999, 222-225; Awaya 1983, 237-239). There is no clear consensus among Japanese historians about when interwar semi-democracy ended, but it is generally accepted that the political parties almost completely lost their influence over the executive and the legislature branches at the onset of the Second Sino-Japanese War in the summer of 1937. (2) This implies that Japan’s first democratic experiment was dismantled in, at most, just under fourteen years.

The democratic backsliding in interwar Japan poses a particularly interesting conundrum to the analysts of democratisation. (3) Why precisely did political parties fail to jointly stem the tide of military dictatorship if neither a military takeover nor electoral competition were politically tenable for military radicals? Indeed, these radicals engaged mostly in a power struggle against the political parties until 1936, when some military officers attempted a coup against the party government (Kitaoka, 1999: 236-44). Yet the coup turned out to be half-baked, and was immediately repressed by the remaining forces in the military, who did not support an out-and-out takeover of the government. Likewise, military radicals never successfully subjugated political parties through electoral competition. As the following analysis suggests, if the political factions had been united, they could have prevented the formation of a military administration in the critical juncture of the 1930s. This brings us back to the puzzle: why were political parties unable to act together and preserve the new democratic institutions?

This essay addresses this question by taking up the last fourteen years (1924-37) of the interwar regime, and seeks to identify necessary conditions that turned Japan’s first democratic experiment into a military dictatorship. The first section of this article outlines two alternative hypotheses on democratic backsliding in the existing literature, and investigates the empirical limitations of these existing approaches with reference to the Japanese case. I will demonstrate that the proposed causal chains between political conditions and this power reversal are rather tenuous in these studies. This discussion suggests that the crux of the downfall of the semi-democratic system was the regime-supporting political parties’ inability to form a free-standing counter-coalition against the Army. I then proceed to show that the critical conditions of their inability to cohere in the times of politico-economic crisis can be found in the specific patterns of political contestation and labour incorporation under the interwar two-party system. I conclude with a brief discussion of how analysts will benefit from tracing these patterns even further back to the preceding period and as well as investigating the role of party system and political incorporation in the other cases of democratic backsliding.

2. Theoretical Literature on Democratic Backsliding

Past comparative studies point to the possibility that the mechanisms of democratic breakdown could be causally distinct from those of consolidation (e.g., Przeworski, et al., 2000; Svolik, 2008). These studies have suggested that while economic development, as measured by per capita income, does increase the chances of democratic survival and, by extension, of democratic consolidation, low levels of development do not by themselves set off the process of authoritarian reversals. Rather, when poor economies preclude consolidation, democracies become susceptible to proximate causes, such as recessionary crises (Gasiorowski, 1995), that start separate causal sequences of backsliding into authoritarian regimes.(4) However, since an economic crisis in less-developed countries will not always dismantle transitional democracies, some “other conditions” (Przeworski, et al., 2000) must be working in tandem to cause democratic breakdown in those cases. When such backsliding sequences coincide with a particular set of social, political, or economic conditions, democratic regimes' chances of survival are greatly diminished. Thus, any convincing explanation of democratic collapse will necessarily specify the background conditions that translate a politico-economic crisis to a deadly blow to transitional democracies.

A series of seminal works in the field sought to specify what such critical conditions might be, focusing on the exhaustion of the “easy” stage of import substitution industrialisation induced by economic development and the subsequent formation of authoritarian alliance among dominant classes (O’Donnell, 1973, 1978) and multiple socio-political conditions, including the fragmentation of parties, societal heterogeneity, and the loyalty of military forces (Linz, 1978). Other works narrow down a range of plausible conditions to one or two variables in order to clarify the ambiguities surrounding the relationships among multiple conditions. Examples include analyses of the veto power held by military forces (Agüero, 1992), the fragmentation of party systems in presidential regimes (Mainwaring, 1993; Stepan and Skach, 1993), and the experience of past military or monarchical regimes (Cheibub, 2007; Svolik, 2008).

Other analysts have suggested that the relationships between executive and legislative institutions (Beliaev, 2006; Clark and Wittrock, 2005; Mainwaring and Shugart, 1997) seem to have a profound effect on regimes’ susceptibility to democratic erosion. Collier and Collier’s work on Latin American cases in the 20th century (Collier and Collier, 1991) analyses the effect of different types of party systems on regime durability. According to Collier and Collier, different modes of labour incorporation produced four types of party systems in Latin America, and the configuration of these party systems profoundly influenced the regime's chances of survival in the periods that followed the establishment of party systems.
More recent works by Morgan and Seawright reintroduce the party system to the debate of democratic backsliding and argue that the collapse of traditional party systems in Venezuela and Peru was precipitated by the party systems’ failure to represent voters ideologically and programmatically (Morgan, 2011; Seawright, 2012). In the 1980s and 1990s, traditional parties in both countries became less capable of programmatically capturing the voters, maintaining a corporatist representation, and/or satisfying voters’ clientelistic demands. This motivated voters to leave the parties and instead to vote for political outsides. Although Seawright and Morgan differ on how they conceptualise the linkages between parties and citizens and also on what causes such linkages to become tenuous, the broken linkages between parties and voters, when combined with corruption or bad economic performance, play a central role in both studies.

Among existing literature, which varies greatly in its emphases, there are two distinct analytical responses to the question of democratic backsliding in interwar Japan. These examine labor militancy and the subsequent consolidation of anti-democratic coalition (Ohara, 2001; Gordon, 1991) and acts of violence combined with the legacies of past authoritarian institutions (Awaya, 1983; Berger, 1977; Scalapino, 1975). All of these studies share the premise that recessionary crises are integral to the democratic backsliding, but differ on what additional conditions lock an economically-distressed regime into a spiral of democratic erosion. More importantly, the notion that the military’s political victory over pro-democratic factions was the final straw is, as the following review suggests, implicit in all of these hypotheses. This premise is particularly important for testing these hypotheses; if a reversal of power to anti-democratic military factions was indeed a final, irreversible step in the backsliding process, then one would expect these additional conditions, in conjunction with a recession, to generate shifts in favour of anti-democratic groups.

Contrary to the existing accounts that emphasise labour militancy and military violence, the essay argues that the political parties’ inability to form an enduring counter-coalition, the critical condition of the regime’s downfall, must be located in the specific patterns of political contestation and labor incorporation under the interwar two-party system. In particular, the rest of the essay demonstrates that the regime’s downfall was a function of two important conditions: (1) polarising urban-rural cleavages, embedded within a two-party system whose political parties were structurally divided over democratic and fiscal principles, which precluded the construction of an enduring counter-coalition against anti-democratic agents, and (2) the lack of broad labor incorporation at the early stage of party development, which caused the working classes to give political support to the military when an attempted counter-coalition fell apart. The goal of this essay is to substantiate, both theoretically and empirically, the causal chains connecting these critical conditions to the regime outcome.

3. Labour Militancy

Ohara (2001) offers an alternative account suggesting that the crux of democratic breakdown in interwar Japan was not recessionary crises themselves, but rather the subsequent radicalisation of working classes, which ultimately altered the balance of power between pro- and anti-democratic groups in favour of the latter. Ohara makes an explicitly causal claim, and argues that, when propertied classes in the urban and rural sectors faced the surge of labour/tenant dispute during a protracted recession, these groups perceived this as a serious threat to their economic interests and began to assess the relative costs of democratic institutions and feasible authoritarian alternatives (Ohara, 2001: 121-122). Seeking remedies and protections from the repressive apparatus of the state, whose members were generally inimical to democracy, propertied classes joined an anti-democratic coalition and thereby shifted the power balance.

Ohara’s emphasis on the political incorporation of propertied classes closely parallels Moore’s (1966) and Rueschemeyer, Stephens, and Stephens’s (1992) analyses of an anti-democratic triad, which suggests that democratic survival depends upon the absence of an anti-democratic coalition of state, large landholders, and bourgeoisie. In all of these works, the key factor is the inclusion of propertied classes both in cities and the countryside into a cross-sector anti-democratic coalition, and according to Ohara, this type of inclusion is a function of recession-induced labor militancy, which aggravates class cleavages in both rural and urban areas and drives vulnerable propertied classes to ally with the military.

Figure 1.  Tenant and Labour Dispute

 Source: Nomusho Tokeihyo Dai 22 ji, 1945; Nihon Keizai Tokei Sokan, 1930

The strength of this explanation is its potential to account for the timing of the regime’s downfall. In the first six months of 1937, when improving economic conditions coincided with a tight labor market and the consolidation of labor movements “disputes soared in number throughout Japan” (Gordon, 1991: 302). Prior to this, the unemployment rate declined from 6.88 (1932) to 4.35 percent (1936), and the level of exports increased by 240 percent (Sakai, 1992: 89; Banno, 2006: 194). The restoration of Japan's economic health in the rural sector took much longer, but the respective prices of rice and silk rose by 67 and 64 percent between 1931 and 1936. This quick recovery can be partly attributed to Takahashi’s emphasis on a large-scale fiscal stimulus (Matsumoto, 1985: 102-5). As Gordon describes, had it not been for the Second Sino-Japanese War, “the year 1937 would have been by far the time of greatest labour protest in Japan’s history...[T]hese strikes were far more assertive… more intense, more politicised,” than in previous periods.(5) Similarly, in the countryside, the number of tenancy disputes peaked throughout 1935 and 1936. In response to the rise of labour militancy, “military men and bureaucrats... truly feared that domestic social order might collapse” (Gordon, 1991: 239). The surge of labour militancy in 1937 and the perceived need for government/military intervention coincided with the timing of democratic backsliding as pro-democratic groups were politically subjugated by radical rightists in the Army.

The question is whether a shift in the balance of political power, the critical intervening variable, can be deduced from the propertied classes’ reactions to radicalised labour. Evidence indicates that the deepening cleavage in urban and agrarian communities was not necessarily translated into the military’s political influence over the regime-supporting political parties. Despite the radicalisation of labour and tenancy disputes, the working classes never successfully established themselves as a viable third party in the Diet until April 1937; even in 1936, the Social Masses Party (Shakai Taishuto) could secure only 18 seats out of 466 and it was 1937 before they finally managed to consolidate their organisations and attained a casting vote with their 37 seats (Nihon Kingendai-shi Henshu Iinkai, 1978: 930). The lateness of labour mobilisation suggests that, until the very last phase of the authoritarian reversal, the surge of labour dispute posed virtually no political threat to the propertied classes in interwar Japan.

Figure 2.  Participants per Dispute
 
Source: Nomusho Tokeihyo Dai 22 Ji, 1945; Nihon Keizai Tokei Sokan, 1930

Perhaps because of this, the actual political damage from the decline of propertied classes’ support was marginal; in 1936, the progressives and the conservatives jointly secured 82% of the seats in the House and 76% in 1937, both of which are more than sufficient to dismantle any non-democratic administration (Ibid.). The declining trend indeed suggests that the rise of labour militancy might eventually have caused the reversal of power in the Diet through elections, had the democratic order not ended merely three months after the election of 1937. In the actual case, however, it occurred too late to have any such effect. Moreover, as shown in Figure 2, despite the surge of tenant dispute in 1935-1936, the political threat from tenant farmers was rather mitigated because the number of participants per dispute was in a declining trend. This implies tenant dispute became smaller and more fragmented in the last years of semi-democratic regime.

4. Legacies of Authoritarian Institutions and Violence

Historians provide a simple, and yet powerful alternative account—the country under the stalled economy was locked into a spiral of democratic backsliding by two factors, a series of coup attempts and assassinations combined with the design flaw of the Meiji Constitution (e.g., Awaya, 1983; Koketsu, 2005; Scalapino, 1975). The argument is generally presented in two steps. First, the military’s effective monopoly over the means of violence sets off a series of violent reactions to the rise of mass parties. Second, despite the political parties’ majority status in the House, these violent acts eliminated their political capacity to prevent the Army’s political exploitation of weak civilian control in the Meiji Constitution, and produced the reversal of power in 1937.

Historians generally suggest two particular incidents as the most devastating blows to the interwar regime. First, in 1932, the foundation of the semi-democratic order was threatened by the May 15th Incident, in which hard-line expansionists in the Navy and rightist civilians assassinated Prime Minister Inukai in protest against the London Naval Treaty and abruptly terminated the Seiyukai administration. The second incident occurred in February, 1936, when radicals in the military attempted a coup d'état and assassinated three major political figures; this is now known as the February 26th Incident.(6)  Immediately following the incident, Army soft-liners used the coup attempt by the hard-liners as a pretext to usurp executive power in the name of national security. The soft-liners insisted that only generals and admirals on active duty could be appointed to hold ministerial positions in order to prevent the reintroduction of newly-retired generals of the soft-liner faction who staged or supported the coup. Nullifying the informal taboo on the appointment of active-duty officers to the cabinet, the soft-liners were able to monopolise the positions of Ministers of the Army and the Navy (Nakamura, 1977: 19; Berger, 1977: 91).

Figure 3.  Military Expenditure, 1930-1937
 
Source: Ryoichi Miwa, “Takahashi Zaiseiki no Keizai Seisaku,”(1979), 117.

One merit of this argument is that it provides a plausible account of how, despite the parties’ dominance in the House of Representatives, the Army managed to overpower the political parties. The trend of military spending provides an idea of the power balance between the two opposing camps during budgetary sessions. As shown in Figure 3, the sudden increase in the military expenditure in 1937 suggests that the February 26th Incident, particularly the assassination of Finance Minister Takahashi, the politically-powerful Seiyukai opponent of arms expansion, decimated the parties’ ability to resist the military’s exorbitant demands. The data also suggest that the Manchurian Incident of 1931 and the May 15th Incident of 1932 contributed to the military’s leverage over the political parties, as it helped compromise the integrity of the democratic order by significantly expanding the presence of nonelected actors in the cabinets. Although the level of military expenditure is only indirect evidence of this inference, the bottom line is that the coup attempt of 1936 seems to be strongly correlated, if not causally linked, with the reversal of power between the two opposing camps.

The credibility of the argument largely depends on the premise that these violent acts eventually stripped off enough of the political parties’ capacity to resist the Army’s countervailing power to enable the 1937 overthrow. However, between 1931 and 1937, despite these violent attacks, the progressives and the conservatives still had opportunities to work together and preserve the control of the Diet. Three incidents are critically important. The first opportunity arose after the October Incident of 1931 (Jugatsu Jiken), in which young military radicals and the extremists of the right-wing group Ketsumeidan planned a coup against the progressive’s administration (Kitaoka, 1999: 220-2; Awaya, 1984: 231-3; Sakai, 1992: 18-9). Although the coup was detected and preempted by the military’s staff headquarters, the progressives and moderates of the conservative party discussed the possibility of forming a grand Conservative-Progressive coalition (Kyoryoku Naikaku) to consolidate their power. The Manchurian Incident of 1931, the October Incident, and the political parties’ subsequent perception of threat from radical rightists temporarily made the conservatives and the progressives closer and made a jingoist-pacifist divide less evident (Banno, 1985).

The second opportunity arose between late 1933 and early 1935. By this time, the three politico-economic crises—the repercussions of the Great Depression, the Manchurian Incident, and the May 15th Incident—had settled down, and this created a window of opportunity for the two parties to form a unified front and jointly restore their control of the cabinet, the power that they lost after the May Incident (Sakai 1992). Third, even after February 1936, due to the impracticability of another violent attack and their majority in the Diet, the political parties could have impeded the ongoing process of democratic breakdown (Banno, 2006: 194). Indeed, a diary record indicates that the assassination of the ministers angered the emperor, who was rather sympathetic to the liberal constitutional order, and the emperor even attempted to lead the Imperial Guard himself to suppress the rebels (Honjo, 1967: 275-7). Given the emperor’s outright opposition to the Army’s coup attempt and the subsequent purge of the hard-liner Imperial Way faction, attempting another coup was impractical for the remaining rightist radicals in the Army who were instilled with the ideology of imperial sovereignty.

On the other hand, the political parties had overwhelming support from the masses who were averse to the Army’s violence, fascist rhetoric, and proposed tax increase. In the April 1937 election, the progressives, the conservatives, and the Social Masses Party together managed to secure approximately 390 seats out of 466 in the House, which would have been more than sufficient to dismantle any Army-led administration if they had been united (Abe, 1996: 184-7). The progressives remained the largest party in the House with 180 seats, while the conservatives followed them with 174 seats. At the same time, the Social Masses Party secured 37 seats in the House and continued to increase their seats in local elections across the country.

What makes these opportunities to form a counter-coalition important is the fact that, in the first and second cases, their proposed anti-fascist administrations could have suppressed violent reactions from the Army; Ugaki Kazushige, a retired Army General who was rather pro-democratic and politically powerful enough to subdue radical elements in the Army, was willing to ally with them at these times (Sakai, 1992: 141-2; Matsuo, 1976: 84; Ito, 1987: 4). Matsuo Takayoshi, Takahashi Susumu, and Miyazaki Ryuji, prominent scholars on Japanese history, consider that the interwar regime could have survived if the conservatives and the progressives had coordinated at these critical moments in fending off challenges from military radicals (Miyazaki and Takahashi, 1985).

In the third case, according to Banno Junji (perhaps one of the most eminent experts on Taisho democracy), if the progressives, the conservatives, and the working class party had jointly opposed the Army’s proposals in 1936 and early 1937, the political stalemate between the regime-supporting camp and the Army would have continued so that succeeding Army-led administrations would not have been stable enough to persist (Banno, 2006: 203). Even if creating such a supermajority was implausible, if the progressives had allied with the working class party and some independents to create a majority in the House, it would have been possible to deter the process of military takeover. Of course, if these coalition attempts fell apart precisely because of military violence, the thesis of military violence would still hold. However this was not the case. I explain why in the next section.

5. The Structure of Two-Party System and Labour Incorporation

If these coalition strategies fell apart for reasons other than the violent assaults on which the accounts I've discussed are based, knowledge of why the parties could not cohere in these critical moments can greatly increase our understanding of the democratic breakdown. The party system provides an answer to the core of this puzzle; the structure of a fractious two-party system played a decisive role in preventing the construction of a counter-coalition between the conservatives and the progressives. There are two key elements central to this thesis: (1) a partisan divide over democratic and fiscal principles reinforced by underlying rural-urban cleavages and (2) the lack of broad labour incorporation at an earlier stage of party development. The series of attempted coalitions in the 1930s fell apart precisely because these two aspects of the party system jointly eliminated the margin of interparty coordination. After explaining each, I will discuss their relevance to the breakdown of attempted counter-coalitions against the Army.

Figure 4.  Explanation of Democratic Breakdown in Interwar Japan  

6. Fiscal Policies and Rural-Urban Cleavages, 1931

Between 1900 and 1924, the two parties had taken very different trajectories of party development; the Seiyukai conservatives had deeply entrenched themselves among rural landowning constituencies, whereas the progressives had catered to the interests of big business and urban upper-middle classes (Kimbara, 1975: 270-271; Banno, 2006, 124-6). This rural-urban cleavage embedded in the two-party system produced a new interparty divide, which was characterised by two distinctively different policy orientations—the conservatives’ emphasis on expansionary fiscal programs and the progressives’ insistence on contractionary policies (Awaya, 1983: 90-3, 98-101; Najita, 1974).

The Seiyukai adopted expansionary programs, primarily because the rural-based conservatives were susceptible to the political gain that would accrue from frivolous pork-barrel spending directed to underdeveloped provincial economies. Indeed, especially after 1914, under the leadership of Hara Takashi the Seiyukai actively allocated public spending to the rural sector through the distribution of public works, such as railroad, telecommunication network, river improvement, and agrarian subsidies (Kimbara, 1975: 267-71, 281-2; Masumi, 1988: 166).(8) These efforts paid off politically, as they allowed the Seiyukai to regain a majority status in the House of Representatives in 1917 (Kimbara, 1975: 279). The Seiyukai’s pork-barrel tactics also led to their victory in the nationwide local elections of 1919. As a consequence, by 1924, expansionary fiscal policies directed at the rural sector had become an integral part of the conservatives’ platform.

Figure 5.  Total Expenditure, 1914-1936
 
Source: Akira Hara, “1920 Nendai no Zaiseishishutsu to Sekkyoku Shokyoku Ryoseisaku Rosen,” (1981), 80-81.

By contrast, catering to their big-business constituencies, the urban-based progressives were in favour of austere fiscal policy and the rationalisation of government, military, and industry (Kimbara, 264: 292; Mizunuma, 1976: 67-8; Awaya, 1983: 58). Even after 1925, although the logic of democratic competition provided the progressives with an incentive to extend social welfare programs to newly-enfranchised working classes, the progressives were unable to reverse big-business’s grip on their party executives and continued to uphold austere fiscal principles. For example, in 1931, due to opposition from a capitalist interest group, the Joint Association of the Capitalists’ Organisation and Industrialists’ National Organisation, the Hamaguchi administration was unable to legislate labor rights through the Labour Union Law (Rodo Kumiaiho) (Mizunuma, 1976: 66; Gordon, 1988: 239; Yoshii 1985, 55-6).  In the progressives’ second Wakatsuki administration, the progressives could not even offer relief measures for unemployed workers in the midst of the Great Depression (Banno, 2001: 147-9).
The progressives were unable to press for labour rights even in the mid-1920s. During the First Wakatsuki cabinet (1926-7), the progressives failed to pass the labour union law due to their financial reliance on Zaibatsu (Takeda, 1985: 76; Yasuda, 1994: 126-30). Kase writes, “The Kenseikai Cabinet… was slightly more positive about unemployment policy than was the succeeding Seiyukai cabinet… but over the period of both administrations the planning of unemployment insurance did not progress” (Kase, 2004: 208). By 1926, the progressives’ preferential treatment toward the bourgeoisie was already evident (Matsuo, 1976: 92). Although, when compared to the conservatives, the progressives were, in principle, supportive of political inclusion of middle and working classes, the bourgeoisie’s intraparty dominance over the labour sector established prior to the suffrage expansion in 1925 locked the progressives into a pattern of fiscal austerity.

By 1925, these rural-urban cleavages were deeply embedded into the two-party system and had a lasting impact on the parties’ expenditure policies in the following years. As shown in Figure 5, the period between 1925 and 1935 can be divided into five phases—the progressives held premiership in (A), (C), and (E), whereas the conservatives were in power in (B) and (D). The trend indicates that the government increased expenditure when the conservatives were in power, whereas it either reduced or maintained the status quo under the progressives’ control. To isolate the effect of military spending, Figure 6 focuses on expenditure on civil engineering, transportation, and industry. The two distinct spending patterns are also evident in these policy areas.

Figure 6.  Fiscal Expenditure on Civil Engineering, Transportation, and Industrial Investment
Source: Akira Hara, “1920 Nendai no Zaiseishishutsu to Sekkyoku Shokyoku Ryoseisaku Rosen,” (1981), 80-81.

These political reactions to rural-urban divisions entailed the formation of a new policy divide over fiscal principles, and the grand coalition plan of 1931 was ultimately shelved precisely because of this policy gap. When moderate members of both parties attempted to build a Cooperative Cabinet (Kyoryoku Naikaku) in the aftermath of the October Incident of 1931, both parties recognised that it was in their interest to prevent any further escalation of the Manchurian Incident, which could jeopardise Japan’s relations with the US and Britain. Although both factions preferred to limit the military’s position in the state, overcoming the divide over fiscal programs was impractical for them because doing so would inevitably undermine their structurally-given base of support. “We could support the Minseito’s foreign policy,” Tsuyoshi Inukai, the president of the Seiyukai, confided to an oligarch just two months after the Manchurian Incident (Harada, 1950: 127-8). “Still, because of our differences in (fiscal) policy, it would be difficult to form a coalition,” Inukai concluded (Ibid.). Ugaki Kazushige, a retired Army General who could help subdue radical officers, later attempted to persuade both parties to set aside their differences in fiscal policy to contain the Army, and still the coalition plan never came through due to this policy divide (Ugaki, 1970: 818).

7. Ideological Divide and Rural-Urban Cleavages, 1933-1935

The second critical aspect of the two-party system is the ideological divide over key democratic principles, such as popular government, political freedoms, and civilian control. This divide was also induced by rural-urban cleavages. A major episode that demonstrated the Seiyukai’s ideological orientation occurred in 1924 when a rightist clique in the Seiyukai party went as far as to support the Kiyoura cabinet—the authoritarian administration that opposed the introduction of universal male suffrage. This administration was exclusively composed of aristocrats, bureaucrats, and military officers (Kitaoka, 1999: 36-7; Ito, 2002: 240; Yamamoto, 1980: 168-9). The Kiyoura cabinet was the waning oligarchy’s last attempt to battle the political parties and retake control of the executive branch.

Prior to this decision, given the overwhelming public support for further democratisation, centrist conservative leaders urged their party members to align with a democratic bandwagon and press for suffrage expansion in an upcoming election.  However, the rightist clique regarded popular sovereignty as either premature or undesirable and successfully induced the party’s majority to support the authoritarian government (Ito, 1988: 160; Ugaki Kazushige Nikki, vol., 1, 1968: 259; Kimbara, 1972: 287). In 1925, the conservatives’ rightist position attracted an authoritarian Army general, Giichi Tanaka into the party, and their leaders handed him the party leadership (Matsuo, 1976: 87; Ito, 1988: 137-9; Ishigami, 1960).

The subsequent consolidation of the rightist clique in the conservative party entailed an important consequence—by 1927, the conservative party’s overall ideological disposition had gravitated further toward the right, although not to the extent of fully embracing authoritarianism.  In 1926, the Seiyukai’s shift to the right led the conservatives to exploit the Kokutai ideology during the Park Yeol Incident. The Park Yeol Incident set a precedent of exploiting the ideology for political purposes, which, according to Matsuo, later led the Seiyukai to take advantage of the dispute over the London Naval Treaty to dismantle the progressive administration (Matsuo, 1976: 87-90). This ideological shift translated into a series of semi-authoritarian policies, such as the revision of the Peace Preservation Law and the national expansion of the Special Higher Police in 1928, both of which curtailed civil liberties and compromised the integrity of the democratic order (Kimbara, 1975: 292; Scalapino, 1975: 275-6).

The fact that the majority of the conservatives embraced this far-right position was also conditioned by the rural-urban cleavages. When nationwide movements for democratisation developed in the aftermath of the First World War, for the rural-based conservatives who had blatantly ignored disenfranchised city-dwellers for almost two decades, expanding suffrage in hopes of cultivating a new base of support among hostile urban classes was politically impractical.  Fearing also that radically expanding the number of eligible voters would undermine their electoral advantage established among propertied classes in the countryside, the majority of the conservatives were even more motivated to oppose broader political inclusion, and their ambivalence toward democratisation eventually became a conduit for the introduction of authoritarian agents into the party (Ishigami, 1960: 348-50; Yamamoto, 1980; Koketsu, 2005). Thus, the conservatives’ ideological deviation from democratic principles in the 1920s and the 1930s was also induced by the socioeconomic characteristics of the support base the party had established prior to the suffrage expansion.

By contrast, during the interwar years, progressive factions were by and large consistent in upholding democratic principles. Influenced by Yoshino Sakuzo, the most influential democratic thinker in modern Japan, the progressives were ideologically indoctrinated to the idea of popular government and greater political freedoms.(14)  The prevailing mood of the progressives’ political discourse in the 1920s was eloquently summed up by a chief secretary of a progressive faction, who confirmed their firm commitment to suffrage expansion and popular government:

By introducing (universal male suffrage), it is our purpose to reform the parliament with limited political participation, rectify the politics of inequality by privileged classes… and to realise fair and rigorous politics by the great majority of the people.(15)

While democratic principles became the pillars of the progressives’ ideological platform, the party had strategic reasons to embrace these ideas: in the late 1910s, suffrage expansion became an effective method of competition against the conservatives. Indeed, when the post-war inflationary spiral and the widening income disparity induced disenfranchised masses in cities to call for suffrage expansion, political competition drove the progressives to transform into an urban-based mass coalition that defined itself against a conservative party narrowly based on the interests of rural landowning classes.  As described earlier, ultimately the progressive party could not politically incorporate working classes due to the bourgeois’ intraparty dominance, but it was this political opportunity structure that propelled the progressives to commit to democratic principles centered around suffrage expansion. Taken together, the cleavages between cities and the countryside shaped both parties’ ideological dispositions and jointly redefined the contentious nature of the interwar two-party system.
The opportunities for counter coalitions against the Army in 1933-1935 never came to fruition precisely because of this entrenched ideological divide. In late 1933-early 1934 and late 1934-early 1935, the Conservative-Progressive Coalition Movement (Seimin Renkei Undo) gained momentum among the moderate members of both parties as a means to curtail the Army’s presence in the cabinet (Banno, 2006:178). By 1933, because the progressives’ deflationary prescriptions to the recession of 1930-1931 had been widely recognised as disastrous to the depressed economy, both parties’ disagreement over fiscal policy had become less salient.

The problem, however, was that constructing an enduring counter-coalition also required that a majority of the members of such a coalition share a commitment to basic democratic principles, because the whole purpose of this coalition was to restore democratic government. At these critical moments, the Seiyukai’s mainstream clique (Sosai-ha) shelved the coalition plans precisely because of their ambivalence about defending the democratic order (Yamamoto Jotaro Denki, 1943: 872). In the period between late 1933 and early 1934, the Sosai-ha prioritised a single-party cabinet over jointly forming a cabinet with the progressives even at the expense of foregoing their long-term chances of preserving democratic institutions.  In the second case (late 1934-early 1935), the Sosai-ha even pretended to go along with the plan while secretly conspiring against the coalition movement, and at the last minute, defected from the progressives as a sanction against their own moderates who prioritised the restoration of democratic government.  Thus, at these profoundly important moments, divergence among ideological outliers in each party precluded more moderate members of both factions from convincing the majority in the Seiyukai to cohere for the restoration of popular government.

8. Lack of Labour Incorporation, 1936-1937

After the February 26th Incident of 1936, without a progressive majority to independently form an administration, resisting the process of breakdown could only be accomplished by the alliance among the progressives, the Social Masses Party, and some independents. This could have been a viable strategy, given that proletariat organisations were no longer fragmented or marginal; in the March 1937 election, working classes managed to increase their share of seats in the legislature, securing a casting vote in the House of Representatives (Awaya, 1984: 266). But, instead of joining with the progressives to defend the democratic institutions, members of the labour party turned to the Control Faction (Tosei-ha) of the Army for an expansionist fiscal policy and war procurement, which deprived the progressives of their last viable strategy (Nishida, 1997: 261; Sakai, 1992: 139; Yasuda, 1994). On controlled economy, see also Nakamura 1977, 13. ). Labour defection in the post-February 26th period thus demands an explanation.

This turnaround of working classes was conditioned primarily, but not exclusively, by the lack of broader labour incorporation in the previous two decades. While the mobilisation of Japanese labor began in the 1910s, labour incorporation, i.e., the formalisation of the relationship between a body of workers and political leadership, primarily through institutional, policy, or political arrangements, never fully occurred before the regime’s downfall in 1937. After 1925, due to a series of repressive state interventions, Japanese socialists remained organisationally fragmented and politically weak until 1932 when multiple proletarian factions combined to form the Social Masses Party (Nimura, 1975: 125-131).

The Social Masses Party represented a broad segment of working classes in both cities and the countryside, but without the prior experience of mobilising a mass labour base, the party failed to electorally penetrate the Diet, or form an alliance or strike a deal with the progressives, resulting in the continuing absence of labour incorporation.  As I’ve discussed above, the Japanese working classes remained politically subordinated by the bourgeoisie throughout the better part of the interwar political development. To working classes, reconstructing their political ties with the progressives who had previously betrayed their commitment to workers’ rights was simply infeasible (Banno, 2001: 184). As Banno writes, “the Minseito and the Seiyukai completely watered down the labor movement’s central demand, the Law on Retirement Reserve Fund and Retirement Benefits (Taishoku Tsumitate Kin Oyobi Taishoku Teate Hou)… under these conditions… idea of uniting the young and aggressive segments of the Minseito and the Seiyukai and the Social Masses Party to form an anti-Fascist popular front against the Army was nothing more than a desk plan.”

Instead, socialist leaders were more receptive to the idea of the military soft-liners’ administration on the condition that the military government would help labour legislate the retirement fund bill and the Labour Union Law, reduce the income inequality between the proletariat and the bourgeoisie, and offer a large-scale economic stimulus through arms expansion and a controlled economy. At this point, the Social Masses Party had no choice but to ally with the military and protect the livelihood of working families. Berger writes, “the Social Masses Party… by 1937 had abandoned it goal of social democracy in favour of supporting the army’s program of state economic controls” (Berger, 2005: 131). Labour defection, made possible by the lateness of labour mobilisation, thus effectively ended the last chance to defend the interwar regime in 1937.

9. Conclusion

This article has explored the roots of democratic breakdown in interwar Japan. I first elaborated the empirical and theoretical limitations of the existing explanations of the regime’s downfall, critically assessing the tenuous connection between their explanatory variables and the critical intervening element, the military’s political victory over pro-democratic factions. I then demonstrated that the democratic breakdown was a function of two structural factors, rural-urban cleavages and the lateness of labour mobilisation, and argued that the two conditions entailed the contentious two-party system and the lack of broad labour incorporation respectively, both of which precluded the construction of counter-coalitions in the 1930s.
While the literature on democratic breakdown continues to grow, many of the empirical works are developed with the intent of clarifying the comparatively-fragile nature of presidential systems in the post-1945 period. To elaborate a more general framework that goes beyond this temporal restriction and its methodological focus on presidential regimes, however, I have found it useful to reexamine the parliamentary cases of interwar years, particularly the case of Taisho Japan, a case that had rarely been revisited by the recent macrocomparative literature on this issue. Likewise, while the correlation between recessionary crises and regime outcomes has been well documented by scholars, the generalizable causal connection linking additional conditions and the regime outcome hasn’t been fully explored in these empirical works. If we explore the causal relationship between the party system and labour incorporation on one hand and the regime outcome on the other in other cases, the generalisability of the causal pattern found in interwar Japan can be ascertained.

While rural-urban cleavages had a profound influence on the subsequent processes of the democratic breakdown, it leads to another question: why, over the course of party development prior to 1925, did the progressives and the conservatives come to associate with such contrasting socioeconomic groups? Likewise, although the lack of labour incorporation was integral to the breakdown processes, this lack also leads us to question how exactly the Japanese bourgeois preempted the mobilisation of labour and consolidated their position within the progressive party. Resolving these questions perhaps requires us to trace the origins of the two-party system and the pattern of labour mobilisation even further back in history, and doing so might reveal that the failure of Japan’s democratic order could have been embedded in the factors or conditions that existed long before the interwar years. Investigating such hypotheses not only will help analysts locate more fundamental causes of the regime’s downfall, but also will help us develop a deeper understanding of the analytical constitution of such variables, which could be located among institutional, structural, ideational, or strategic factors. Hence, their effects on regime durability points to possible channels for future research.

References

Abe, Hirozumi. (1996). Nihon Fashizumu Ron. Tokyo: Kage Shobo.

Agüero, Felipe. (1992). "The military and the limits to democratisation in South America." In: Scott Mainwaring, Guillermo O’Donnell, and J. Samuel Valenzuela, eds., Issues in democratic consolidation: The New South American Democracies in Comparative Perspective. Notre Dame, IN: University of Notre Dame Press.

Awaya, Kentaro. (1983). Showa no Rekishi 6 Showa no Seito. Tokyo: Shogakukan.

Banno, Junji. (1982). Taisho Kaihen. Tokyo: Minerva Shobo.

———(1985). "Seito Seiji no Hokai." In: Junji Banno and Masato Miyaji, eds., Nihon Kindaishi ni Okeru Tenkanki no Kenkyu. Tokyo: Yamakawa Shuppan Sha.

———(2001). Nihon Seiji Shippai no Kenkyu. Tokyo: Kobosha.

———(2006). Kindai Nihon Seijishi. Tokyo: Iwanami Shoten.

Beasley, W. G. (1963). The Modern History of Japan. New York: Praeger.

Beliaev, Mikhail V. (2006). "Presidential Powers and Consolidation of New Postcommunist Democracies." Comparative Political Studies, 39 (3), 375–98.

Berger, Gordon M. (1977). Parties Out of Power in Japan, 1931-1941. Princeton: Princeton University Press.

———(2005). "Politics and Mobilization in Japan, 1931-1945." In: Peter Duus, ed., The Cambridge History of Japan, Volume 6, The Twentieth Century. New York: Cambridge University Press.

Bernhard, Michael, Timothy Nordstrom, and Christopher Reenock. (2001). "Economic Performance, Institutional Intermediation, and Democratic Survival." Journal of Politics, 63 (3), 775–803.

Cheibub, José Antonio. (2007). Presidentialism, Parliamentarism, and Democracy. New York: Cambridge University Press.

Clark, Terry D. and Jill N. Wittrock. (2005). "Presidentialism and the Effect of Electoral Law in Post-Communist Systems: Regime Type Matters." Comparative Political Studies, 38 (2), 171–88.

Collier, Ruth Berins and David Collier. (1991). Shaping the Political Arena. New Jersey: Princeton University Press.

Dore, Ronald P. (1959). Land Reform in Japan. London: Oxford University Press.

Gasiorowski, Mark J. (1995). "Economic Crisis and Political Regime Change: An Event History Analysis." American Political Science Review, 89 (December), 882-897.

Gordon, Andrew. (1991). Labor and Imperial Democracy in Prewar Japan. Berkeley: University of California Press.

———(2003). A Modern History of Japan. New York: Oxford University Press.

Hara, Akira. (1981). "1920 Nendai no Zaiseishishutsu to Sekkyoku Shokyoku Ryoseisaku Rosen." In: Nakamura Takafusa, ed., Senzenki no Nihon Keizai Bunseki. Tokyo: Yamakawa Shuppansha.

Harada, Kumao. (1950). Saionji-ko to Seikyoku, vol. 2. Tokyo: Iwanami Shoten.

Honjo, Shigeru. (1967). Honjo Nikki. Tokyo: Hara Shobo.

Ishigami, Ryohei. (1960). Harakei Botsugo. Tokyo: Chuo Koron.

Kano, Masanao. (1975). "Taisho Democracy no Shiso to Bunka." In: Iwanami Koza Nihon Rekishi 18 Kindai 5. Tokyo: Iwanami Shoten, 333-376.

Kase, Kazutoshi. (2004). "Unemployment Policy in Prewar Japan: How Progressive was Japanese Social Policy?" Social Science Japan Journal, (7)2, 199-221.

Kawato, Sadafumi. (1994). The Government and Politics of Japan. Tokyo: University of Tokyo Press.

Kimbara, Samon. (1975). "Seito Seiji no Tenkai." In: Iwanami Koza Nihon Rekishi 18 Kindai 5. Tokyo: Iwanami Shoten, 255-96.

Kitaoka, Shinichi. (1999). Seito Kara Gunbu-e, 1924-1941. Tokyo: Chuo Koron Shinsha.

Koketsu, Atsushi. (2005). Kindai Nihon Seigun Kankei no Kenkyu. Tokyo: Iwanami Shoten.

Kyoto Daigaku Bungakubu Kokushi Kenkyushitsu-hen. (1958). Nihon Kindai-shi Jiten. Tokyo: Toyo Keizai Shimposha.

Linz, Juan J. (1978). The Breakdown of Democratic Regimes: Crisis, Breakdown, & Reequilibration. Baltimore: Johns Hopkins University Press.

Lust, Ellen and David Waldner. (2018). “Unwelcome Change: Coming to Terms with Democratic Backsliding,” Annual Review of Political Science, 21, 93-113.

Mainwaring, Scott. (1993). "Presidentialism, Multipartism, and Democracy: The Difficult Combination." Comparative Political Studies, 26 (2), 198-228.

Mainwaring, Scott and Mathew Shugart. (1997). "Conclusion: Presidentialism and the Party System." In: S. Mainwaring and M. Shugart, eds., Presidentialism and Democracy in Latin America. New York: Cambridge University Press, 394–439.

Masumi, Junnosuke. (1988). Nihon Seijishi 2 Hanbatsu Shihai, Seito Seiji. Tokyo: Tokyo Daigaku Shuppankai.

Matsumoto, Hiroshi. (1985). "Senji Kokka Dokusen Shihonshugi eno Iko." In: Koza Nihon Rekishi 10 Kindai 4. Tokyo: Tokyo Daigaku Shuppankai.

Matsuo, Takayoshi. (1976). "Seiyukai to Minseito." In: Iwanami Koza Nihon Rekishi 19 Kindai 6. Tokyo: Iwanami Shoten.

Minobe, Tatsukichi. (1912). Kenpo Kowa. Tokyo: Yuhikaku.

Miwa, Ryoichi. (1979). "Takahashi Zaiseiki no Keizai Seisaku." In: Tokyo Daigaku Shakai Kagaku Kenkyusho, ed., Fashizumu-ki no Kokka to Shakai 2, Senji Nihon Keizai. Tokyo: Tokyo Daigaku Shuppankai.

Miyazaki, Ryuji and Susumu Takahashi. (1985). "Seito Seiji no Teichaku to Hokai." In: Junji Banno and Masato Miyaji, eds., Nihon Kindaishi ni Okeru Tenkanki no Kenkyu. Tokyo: Yamakawa Shuppansha.

Mizunuma, Tomokazu. (1976). "Kin Kaikin Mondai." In: Iwanami Koza Nihon Rekishi 19 Kindai 6. Tokyo: Iwanami Shoten.

Moore, Barrington Jr. (1967). Social Origins of Dictatorship and Democracy: Lord and Peasant in the Making of the Modern World. Boston: Beacon Press.

Najita, Tetsuo. (1974). Hara Takashi. Tokyo: Yomiuri Shimbunsha.

Nakamura, Akira. (1977). "Tosui to Seiji tono Koso." In: Rekishi Kagaku Kyogikai, ed., Nihon Fashizumu Ron. Tokyo: Azekura Shobo.

Nihon Kingendai-shi Henshu Iinkai. (1978). Nihon Kingendai-shi Jiten. Tokyo: Toyo Keizai Shimposha.

Nimura, Kazuo. (1975). "Rodosha Kaikyu no Jotai to Rodo Undo." In: Iwanami Koza Nihon Rekishi 18 Kindai 5. Tokyo: Iwanami Shoten, 93-140.

O'Donnell, Guillermo A. (1973). Modernization and Bureaucratic-Authoritarianism: Studies in South American Politics. Berkeley: Institute of International Studies, University of California.

———(1978). "Reflections on the Pattern of Change in the Bureaucratic-Authoritarian State." Latin American Research Review, 13(1), 3-38.

Ohara, Masayo. (2001). Democratisation and Expansionism: Historical Lessons, Contemporary Challenges. London: Praeger Publishers.

Przeworski, Adam, Michael Alvarez, and José Antonio Cheibub. (2000). Democracy and Development: Political Institutions and Material Well-Being in the World, 1950-1990. Cambridge: Cambridge.

Rueschemeyer, Dietrich, Evelyne Huber Stephens, and John D. Stephens. (1992). Capitalist Development and Democracy. Chicago: University of Chicago Press.

Sakai, Tetsuya. (1992). Taisho Democracy Taisei no Hokai: Naisei to Gaiko. Tokyo: Tokyo Daigaku Shuppankai.

Scalapino, Robert A. (1975). Democracy and the Party Movement in Prewar Japan: The Failure of the First Attempt. Berkeley: University of California Press.

Stepan, Alfred and Cindy Skach. (1993). "Constitutional Frameworks and Democratic Consolidation: Parliamentarianism versus Presidentialism." World Politics, 46(1), 1-22.

Suetake, Yoshiya. (1984). "Seito Naikaku no Seiritsu to Hokai." Kindai Nihon Kenkyukai, ed. Tokyo: Yamakawa Shuppansha.

Svolik, Milan. (2008). "Authoritarian Reversals and Democratic Consolidation." American Political Science Review, 102 (May), 153-168.

Takenaka, Harukata. (2002). Senzen Nihon ni Okeru Minshuka no Zasetsu: Minshuka Tojotaisei Hokai no Bunseki. Tokyo: Bokutakusha.

Ugaki, Kazushige. (1970). Ugaki Kazushige Nikki, vol.2. Tokyo: Misuzu Shobo.

Yamaguchi, Yasushi. (2006). Fashizumu. Tokyo: Iwanami Gendai Bunko.

Yamamoto Jotaro Denki Hensan Iinkai. (1943). Yamamoto Jotaro Denki. Tokyo: Hara Shobo.

Yamamoto, Shiro. (1980). Nihon Seitoshi (Ge). Tokyo: Kyoikusha Shuppan Service.

Yoshii, Kenichi. (1985). "Manshu Shinryaku to Gunbu, Seito." In: Koza Nihon Rekishi 10 Kindai 4. Tokyo: Tokyo Daigaku Shuppankai.

Young, A. Morgan. (1938). Imperial Japan, 1926-1938. London: Allen & Unwin.

Notes

Article copyright Hiroyuki Yamamoto.

Back to top