Jan Teorell, Determinants of Democratization: Explaining Regime Change in the World, 1972–2006 (Cambridge: Cambridge University Press, 2010).

Zoltan Barany and Robert G. Moser (eds), Is Democracy Exportable? (Cambridge and New York: Cambridge University Press, 2009).

Nathan Brown (ed.), The Dynamics of Democratization: Dictatorship, Development and Diffusion (Baltimore, MD: Johns Hopkins University Press, 2011).

Christopher Hobson and Milja Kurki (eds), The Conceptual Politics of Democracy Promotion (Abingdon: Routledge, 2012).

Judith G. Kelley, Monitoring Democracy: When International Election Observation Works, and Why it Often Fails (Princeton and Oxford: Princeton University Press, 2012).

Steven Levitsky and Lucan A. Way, Competitive Authoritarianism: Hybrid Regimes after the Cold War (Cambridge and New York: Cambridge University Press, 2010).

Staffan I. Lindberg (ed.), Democratization by Elections: A New Mode of Transition (Baltimore, MD: Johns Hopkins University Press, 2009).

Richard Youngs (ed.), The European Union and Democracy Promotion (Baltimore, MD: Johns Hopkins University Press, 2010).

By now the democracy wave(s) that began in Mediterranean Europe in the 1970s and then washed across most of the world's regions has generated a tidal flow of political science literature on democratization. The tide has continued unabated, contrary to growing signs that democratization itself had run out of steam and might even be undergoing a reverse, the ‘Arab spring’ of 2011 notwithstanding. Democracy promotion, or what is now often called support for democracy building or democracy support in order to disassociate it from the kind of coercive approaches that try to impose democracy or an excessive commitment to some very specific institutional architecture, began to gather pace from the late 1980s. A sizeable international democracy promotion community has evolved that comprises, in Laurence Whitehead's words (2012: 21), a ‘network of competing and partially overlapping institutions pursuing multiple agendas at the behest of a diverse set of sponsors’.

It is now passé to say that for many years democratization's international aspects were much neglected in the literature. Even so, the wave of scholarly interest in democracy support in particular took some time to gain momentum. Just as democracy promotion itself was seen to increase as a response to the emergence of democratizing trends on the ground so, as Nathan Brown (Reference Bermeo2011: 241) says from a US perspective, ‘academic interest in international dimensions of democratization increased in large part as democracy promotion became a central component of the foreign policy of the main Western powers’.Footnote ¹ The level of interest is higher now than ever and could well increase further, despite the serious setback that was dealt to democracy promotion's international standing, legitimacy and reputation following the use of military force to bring about regime change in Iraq. Indeed, of the eight books in this review, writing about democracy support or some feature of it is the major aim of four (Barany and Moser Reference Barany and Moser2009; Hobson and Kurki Reference Burnell and Gerrits2012; Kelley Reference Carothers2012; Youngs Reference Gallie2010a); three more speak quite explicitly to interests and concerns recognizable to democracy support (Brown Reference Bermeo2011; Levitsky and Way Reference Chirot2010; Teorell Reference Carothers2010) and Lindberg's book (2009) notes more briefly some democracy support implications too.

Back in 1997, Thomas Carothers, who began studying democracy promotion even before the 1990s and is probably the most widely read and most heavily cited of all observers writing on it, commented that a sound strategy for helping democracy spread must be grounded in a good grasp of what makes democratization happen and how it comes about. From his perspective as a senior figure in the Carnegie Endowment for International Peace, Carothers found that the model of democratization embodied in the US and by extension other democracy assistance ‘is not drawn from the domestic political experience of the United States or other established democracies. Neither is it borrowed from the world of academic theory’ (Carothers Reference Bunce and Wolchik1997: 117). Indeed, the standard model of democracy – liberal democracy – and its historical provenance were clear, but both the way democratization as a process of change was understood and how the assistance strategies drew upon a credible explanation of democratization were anything but clear. At that time the worlds of democracy promotion and independent reflection simply did not seem to meet. The onus for correcting this situation lay on academic research as well as on policymakers.

So, 15 years on, a pertinent question to ask is whether today's scholarly literature can furnish policymakers of democracy support with critical guidance. This is certainly not the only yardstick or, even, the most important one for assessing the literature. And it may well not be – and does not have to be – the main purpose of academics writing about the subject. Indeed, this writer has heard it said that policy engagement should not feature on their radar at all, although in fact the question of why democracy's spread should be supported – as distinct from whether it can be furthered, and how – is not one that features extensively in these books. Modest exceptions are the occasional forays into reasoning that say democracies tend not to make war on democracies (noted by Edward Mansfield and Jack Snyder, who in their chapter in Barany and Moser (Reference Barany and Moser2009) present their familiar argument that premature or incomplete democratization increases the chances of violent conflict) and the benefits that democracy can bring to development (where Brown (Reference Bermeo2011) maintains that the relations of cause and effect operate in both directions).

Titles such as Monitoring Democracy (Kelley Reference Carothers2012) and The European Union and Democracy Promotion (Youngs Reference Gallie2010a) obviously do not have a problem with terms of reference that call for policy relevance. For Determinants of Democratization (Teorell Reference Carothers2010), however, a crucial test in this regard could be whether the book identifies causal variables that democracy supporters potentially can act on – levers they can pull sufficiently to make a difference to democracy's prospects. In fact, we can ask of any of the books whether they help actors engaged in supporting democracy, as well as democracy promotion's many critics, address questions such as ‘when/when not’, ‘how/how not’, and ‘where/where not’ to do it. The questions are not new, but new insights are needed.

So whereas Marc Plattner's opening remark in Is Democracy Exportable? (Barany and Moser Reference Barany and Moser2009: 11) that ‘the prevailing tone of most of the book is one of relatively benevolent skepticism’ may come as no great surprise to some, Hobson and Kurki's invitation in The Conceptual Politics of Democracy Promotion (2012) to think critically about the ‘what’ – to reconceptualize the kind of democracy – in democracy promotion looks much more novel. According to Lindberg, in Democratization by Elections (2009: 315), ‘there seems to be little substantial debate any longer about the normative preference for representative liberal democracy’. This is precisely what Hobson and Kurki set out to challenge. For, while most existing literature on democracy promotion, including such examples as Monitoring Democracy (Kelley Reference Carothers2012) and The European Union and Democracy Promotion (Youngs Reference Gallie2010a), dwell on problems with the promotion in democracy promotion, The Conceptual Politics of Democracy Promotion problematizes the idea of democracy (and hence democratization) in democracy promotion. For this reason, that book along with the monographs by Teorell, Kelley, and Levitsky and Way merit more attention below than the other books, whose primary theme is more difficult to summarize, by virtue of their being collected works, each containing chapters by 10 or more different authors. The monographs are also distinctive in their employment of both large-n studies and selections of empirical cases to reach their findings. In contrast, Youngs's book (Reference Gallie2010a), for example, eschews quantitative study in a preference for narratives on the performance of democracy support in different countries or regions, as well as being the only one to comprise previously published (newly revised) pieces.

Jan Teorell's Determinants of Democratization (2010) in remarkably few pages (160 plus appendices) conducts a wide-ranging review of evidence from 165 countries and nine individual cases from 1972 to 2006, moving us closer to understanding what makes democratic change happen. The findings are multilayered. They provide support for several traditions in the field of comparative democratization studies, including a strategic focus on actors, which works well for making sense of transitions, and structural arguments that, together with the peaceful nature of uprisings and the institutional arrangements of the previous authoritarian regime, supply more forceful explanations of democratic development in the long run. The agency/short-run versus structure/long-run dichotomy is no great surprise. More unusual is Teorell's comparison of different authoritarian regime types’ different propensities for democratic change; it is an analysis that speaks to Brown and Kauffman's conclusion in The Dynamics of Democratization that to really understand what causes democracy to emerge and persist we need a more sophisticated understanding of the different varieties of authoritarian rule (competitive authoritarian regimes being one candidate and the subject of Levitsky and Way's book). Teorell claims that those which allow multiparty elections – however defective – have more promise compared to other types. This seems to fit well with the arguments of Lindberg in Democratization by Elections: although elections are often conceived as an attribute of democracy or even treated as the end point of transition, they can serve as a means of bringing about change towards democracy. Policy implications for democracy promotion are not hard to draw. A further feature of note is Teorell's examination of what causes de-democratization, where a particularly interesting finding is that wealthy countries are less vulnerable not because of their wealth but because of media proliferation.

Teorell converts his analysis of the determinants of democratization into accessible clues for democracy promotion. The idea that in the short run, at least, structural factors are not decisive means that democracy promotion might be able to make a difference by empowering the agents of change. In particular, helping non-violent protests against an authoritarian regime can cause splits within the regime to widen to the point that defections gather pace. In this context international support for multiparty elections can make a practical contribution. But violent protests do not promote democratization. And as the plight of Syria in mid-2012 suggests, where peaceful protests fail or are rebuffed by force then domestic pressures for violence can take over, either with or without external encouragement, and democracy promoters can do little in such situations.

Another striking finding from Teorell's examination of the determinants of democratization is his discounting of economic development. Economic crises are more likely to trigger authoritarian collapse. The paranoia that China's leadership displays about maintaining the country's growth looks justified, but the findings do little to support a reliance on trade or aid as indirect ways to advance eventual democratic breakthroughs in developing countries, even though these can be politically more convenient and less troublesome in other ways for the West. Democracy promotion should also be concerned by Teorell's finding that his statistical models perform dismally as predictors of democratic development in the short run, while accounting for 40 per cent of the variation in the level of democracy at the end of the period as a whole. This offers no encouragement to democracy promoters wanting to identify which of the many non-democracies they should focus their limited resources on next (with the exception of multiparty autocracies facing economic crisis). Democracy assistance seems to be presented with a dilemma. While the apparent unpredictability of short-run democratic development means it can only play catch-up, by reacting opportunistically where it can instead of anticipating and tipping the odds in favour of democratic transition in advance, the possibilities for influencing longer-run democratic development, where our powers to predict look more impressive, appear weak precisely because the more structural conditions then take main effect. However, the findings do suggest that international support for popular access to free media would be very appropriate in places where democracy has already been established. Tunisia might be a good example today.

Staffan Lindberg in Democratization by Elections sets the bar high both for theorizing democratization and for policy deliberation on democracy promotion, where he notes ‘innumerable possible combinations of pathways whereby a country can move forward, backward, and forward again, be stalled for a number of years, and so on, and that the outcome of such paths of transition varies. There is no easy way to depict the many possibilities’ (2009: 15–16). Having already bruited the idea that elections even if not wholly free or fair can be a mode of democratic change in respect of Africa, in this new work Lindberg now subjects the idea to closer examination for a wider set of countries, sharing insights from 15 mainly US-based specialists. The book reveals a more varied picture: a democratic effect is far from guaranteed; in some environments, such as strong authoritarian regimes, de jure competitive elections sustain authoritarian rule, and in other places can bring that outcome closer. Corroborating reasons to explain the anti-democratizing effects can be found in Roger Moser's chapter in Is Democracy Exportable? (Barany and Moser Reference Barany and Moser2009). Establishing the most important influences on the outcome and also their amenability to external influence becomes important not just for theorizing about democratization but for democracy promotion too. The cases described in Linderg's book identify the strength of the opposition parties as a key.

Democratization by Elections concludes with a tentative theory embracing democratization and autocratization or autocratic reproduction by elections, highlighting the processual and institutional incentives that face the opposing actors rather than the actors themselves. Although no recommendations for democracy promotion are formulated, there is an implication that helping strengthen pro-democratic opposition parties between as well as at elections could help democracy's chances. Similar advice can be read into Levitsky and Way's account (Reference Chirot2010) of competitive authoritarian regimes. However, democracy support has found this extremely difficult to do in the past, partly because compared to, say, election monitoring it more easily provokes great political sensitivities and retaliation from the regime. Where party assistance has been extended, the literature finds nothing like a transformative effect; at times the outlook for democracy may even be impaired (see Burnell and Gerrits Reference Brown2012; Carothers Reference Bunce and Wolchik2006). That leaves international election monitoring and, maybe, advice on the electoral systemFootnote ² as potentially more feasible avenues for translating Lindberg's own concluding theoretical insight: ‘iterative, multiparty elections change the costs of both oppression and toleration and thus become major events that affect the cost–benefit analysis for the incumbent as well as for reformers’ (Lindberg Reference Fish2009: 325) into an instrument for informing democracy support.

Judith Kelley's very impressive Monitoring Democracy (2012) is the first book here that is entirely about democracy promotion or one aspect of it. Her findings about when international election observation works and why it often fails constitute a major piece of scholarship that speaks clearly, directly and unequivocally to the needs of democracy promotion. It adds depth and critical nuance to statements like those of Susan Hyde (Reference Bermeo2011: 277) that because of democracy assistance's attention to elections in non-democracies, ‘Leaders today are more likely to hold elections and less likely to get away with election manipulation’. The research underpinning Kelley's book gathered data from over 600 monitoring missions and 1,324 national elections between 1975 and 2004, compiling information from monitoring organizations and countries including 15 case study presentations.

Kelley addresses two main questions: do monitors assess elections accurately and effectively? and do they improve the quality of elections? The overall finding is that monitoring (monitoring and observation are used interchangeably) can improve election quality but most of the time it has not done so. Sometimes it has been biased and contributed false legitimacy to the government. Monitoring is broken, but worth fixing. The book's added value then lies in pinpointing exactly which aspects of monitoring need improving, could be improved and how, and identifying the circumstances where monitoring should not even be attempted. At first sight the possibilities do not look good: monitoring bodies, whether intergovernmental or non-governmental, acquire political baggage and this – together with more practical constraints and legitimate normative concerns – blunts their effectiveness and can compromise claims to neutrality.

In elections where the simple answer to Kelley's first question is yes, the election's quality, whether good or bad, tends not to be in doubt anyway. But in other cases disagreements about the election emerge out of bias, not least in the ‘shadow markets’ where some monitoring bodies have dubious agendas and the governments who intend to cheat can ‘forum shop’. Even highly respectable organizations can be genuinely torn between assessing an election as fair and serving other, sometimes laudable, purposes that such an assessment might help secure, such as a smoother path towards democracy. Conflict-prone environments where a damning indictment of an election might trigger mass violence offer a second illustration. Kelley is critical of the self-restraint that monitors often seem to show towards making severe criticisms but even this might be spun in a more positive light, on the basis of her finding (Kelley Reference Carothers2012: 162) that no evidence exists of monitors causing nefarious politicians who wish to escape censure to shift from overt cheating to less detectable irregularities. Just as monitors express their own verdicts on an election's quality, so independent analysts such as Kelley can still reach their own very different but authoritative assessments. This is an achievement of sorts. A particular finding of Kelley's that is worth noting is that funding of monitoring may not be the problem: throwing more resources at creating bigger missions would not reduce the biases.

On Kelley's second question, the results from comparing monitored with unmonitored elections provide some grounds for believing that monitoring can improve election quality, by reducing the incentives to cheat. She finds monitored elections in multiparty states that are not fully established democracies are ‘both likelier to be seen as representative and to produce a turnover in power’ (Kelley 2012: 167). But several domestic and international factors largely beyond the power of monitors to influence have a bearing on actual effectiveness. The moral for monitors who want to be effective is to target places where the conditions are most favourable. If the capacity to stage elections is weak, then democracy support should prioritize capacity-building over election assessment.

Before reaching its 83 pages of methodological appendices, Monitoring Democracy summarizes seven dilemmas for international actors hoping to promote democracy by election monitoring and outlines 10 policy recommendations in response (while leaving unanswered the ‘election fixation dilemma’ that conflates democracy with elections. Of course this fallacy is not one that any of the writers in these books commits; indeed, according to Susan Hyde (Reference Bermeo2011: 269), ‘as far as I can tell, no one makes this argument in print’. Richard Youngs (Reference Gallie2012: 107) says no one seriously involved in democracy support today would make this mistake either). By helping identify circumstances where monitors should stay at home, Kelley in fact draws attention to the limits of what this form of international democracy support can do for peoples who may be in most desperate need. Avoiding societies where violence will be a factor, for example, could mean forgoing opportunities to exert a beneficial influence, and is not always a feasible option. In what comes across as reinforcement of the finding made in the other books that elections might, but often do not, contribute to a transition to democracy, Kelley's country case studies show that even in the presence of international monitors the domestic struggles for power among rival politicians can very easily descend into serious electoral abuse. Certainly, she demonstrates that democratization is not entirely a domestic process, but even if election monitoring still is the ‘flagship’ of democracy promotion that she says it is, then it seems at most able to reinforce democratic trends when favourable conditions are present but cannot completely transform the democratic outlook. In fact, there are many aspects of what international election monitoring does or could do better that Kelley acknowledges are left for future research (Reference Carothers2012: 179–80), one example being the idea that the most valuable contribution it can make is to create a strong local monitoring capability.Footnote ³ As a corrective, the bigger picture that contains a more varied range of international engagement providing support to democratic development than just on elections must be consulted.

At this point Steven Levitsky and Lucan Way's Competitive Authoritarianism (Reference Chirot2010) can be introduced. It is a veritable tour de force in the literature on democratization, and speaks to the concerns of democracy promotion too. Building on their previous articles, the authors compare the effects of international linkage and leverage with key domestic political variables to explain the various trajectories that competitive authoritarian regimes have taken in the post-Cold War era. Competitive authoritarian regimes are a subset of hybrid regimes defined by their incorporation of elections that are real and competitive but so unfair as to fall short of democracy. ‘Attention to the slope of the playing field thus highlights how regimes may be undemocratic even in the absence of overt fraud or civil-liberties violations’ (Levitsky and Way Reference Chirot2010: 6). Levitsky and Way's study of the 33 such regimes that they detected in 1995 (over a dozen survived in 2010) finds that Western leverage, which refers to vulnerability to external democratizing pressure, has limited potential to effect sustained transition to democracy. In contrast, linkage to the West, meaning ties (economic, political, diplomatic, social and organizational) and cross-border flows, is far more potent. The findings are compelling: 28 out of the 35 cases are explained and only one case is wholly anomalous. Of particular value to democracy support's interest in identifying the most promising countries to focus on are Levitsky and Way's conclusions about what made some competitive authoritarian regimes relatively unstable. Their evidence pinpoints weak state and governing party organization. However, it is where organizational power is high that transitions are more likely to bring sustained democratization, even if it is more difficult to engineer.

Unfortunately, the authors do not at first say exactly where the different mechanisms of international influence they claim to distinguish in the literature – diffusion; direct democracy promotion, which lumps diplomatic persuasion in with threats and force (although their defining properties can be very different); multilateral political conditionality; democracy assistance; and transnational advocacy network – sit in relation to their own neat dichotomy of leverage and linkage. Readers are left to infer that direct promotion, for example, and the use of democratic conditionality are bound up with leverage, whereas democracy assistance – civil society aid for instance – is a (comparatively minor) dimension of linkage. At first glance the large part of democracy promotion that makes up assistance can then draw some comfort from the book's findings. However, there are problems.

First, the analytical dichotomy of linkage and leverage looks too straightforward: as the authors admit, ‘linkage and leverage may overlap, and when both are high, they can be difficult to disentangle’ (Levitsky and Way Reference Chirot2010: 50); in some situations linkage ‘also may be viewed as a form of leverage’ (Levitsky and Way 2010: 50); the greater the linkage the greater the possibilities for (and likelihood of) leverage. High linkage appears as an external shaper of ‘democratizing pressure’ (‘diffuse and indirect but nevertheless considerable’) even where leverage is low (Levitsky and Way 2010: 53); linkage magnifies the domestic impact of external pressure by increasing the chances that it will trigger broad domestic opposition to the regime (Levitsky and Way 2010: 51). In low-linkage countries, international democratizing pressure is considered weaker. So, although democracy assistance is distinguished from promotion, it is certainly not detached from pressure.

Second, Levitsky and Way's account of what leads competitive authoritarian regimes to democratize or alternatively persist, or become more authoritarian, is a structuralist one that casts aside explanations relying solely on economic crises, or institutionalist or leadership-centred arguments. By themselves multiparty elections, for example, cannot explain transition, although international efforts to make them fairer must surely be advantageous. Moreover, linkage is mainly rooted in long-term structural factors that emphasize geography and historical processes not amenable to short-term manipulation (Levitsky and Way 2010: 83). The scope for democracy support specifically to influence the trajectory of competitive authoritarian regimes shrinks. But Levitsky and Way appear reluctant to draw this conclusion: ‘Western linkage-building efforts have a significant medium-term democratizing effect’ (2010: 353). In practice, however, such linkage could comprise mainly an increase in economic ties and flows of people across borders of the kind that China (an authoritarian rather than competitive authoritarian regime) seems very comfortable with, and rightly so if we accept Teorell's conclusions about the political consequences of development.

Finally, Levitsky and Way's analysis is restricted to a particular regime type where most examples have not democratized and it studies a distinctive world historical moment that recent global shifts in the balance of power are now bringing to an end. So, even if democracy support considered as a (relatively modest) contributor to linkage which (when combined with democracy promotion understood as one mechanism for maximizing leverage – not the most potent dimension of international influence) really did help move some regimes along the continuum running from authoritarian to competitive authoritarian to democracy, there are doubts about policy guidance relevant to the future. The lesson for democracy support seems to be to invest in increasing linkage over many years rather than pursue agency-oriented solutions for a quick effect. But Western leverage will decline as the time frame lengthens and the odds against precipitating the breakthrough that would enable international pressure to secure sustained democratization look likely to increase.

Whereas Levitsky and Way (Reference Chirot2010) first and foremost examine the democratizing potential of a particular type of political regime and leave democracy support to ponder any policy implications, The Conceptual Politics of Democracy Promotion (Hobson and Kurki Reference Burnell and Gerrits2012) invites democracy promotion to interrogate the idea(s) of democracy embodied in what it tries to do. Christopher Hobson and Milja Kurki's book breaks new ground in the democracy promotion literature. The first thing to note is that although the editors offer a perspective that has ‘strong affinities with critical theory and post-positivist approaches’ (Hobson and Kurki 2012: 5), they take pains to maintain that the book's purpose is not to undermine the desire to see democracy spread or the aspiration to help that happen. They ‘caution against equating a conceptual politics approach with critical theory, even if neo-Gramscians have done most of the work in this area to date. Rather, the framework outlined in this volume is consciously much more open and pluralist, and one that has potential to be adopted by positivist and post-positivist scholars’ (Hobson and Kurki 2012: 217). The benefits of doing so could work in favour of promoting democracy more effectively. What democracy means – or can mean – and not sympathy for promotion is what is placed in the firing line, although of course different ideas about what democracy is must inevitably have implications for how it comes about – an explanation of democratization – and how it might be supported and should not be promoted.

There is a tradition of saying that democracy promoters operate with restricted ideas of democracy that go no further than a vision of liberal democracy embodying political (party) pluralism, competitive elections and a raft of political rights and civil liberties secured by the rule of law, while in practice often having to settle for something less (‘electoral democracy’). The legacy of elitist or largely proceduralist conceptions of democracy that go back to Joseph Schumpeter and Robert Dahl is mainstream. This is what Hobson and Kurki challenge (Reference Burnell and Gerrits2012: 4), telling readers, ‘It is the aim of this volume to take seriously’ W.B. Gallie's suggestion (Reference Burnell1956) that democracy is best understood as an ‘essentially contested concept’, meaning that different interpretations of what democracy means are acceptable, and no version should be considered the only right version (this is not quite the same as saying that democracy legitimately has multiple meanings). It must follow that democratization is contested too, but in this context meaning not competing causal theories of the type Teorell (Reference Carothers2010) tests against the data but meaning that what democratization aspires to realize can be variable too. A strategy for democracy promotion, then, must be informed by how ideas of democracy come to be defined and used and by whom and, if necessary, how these can be challenged or replaced, as much as rest on a credible social scientific explanation of democratization.

A second point about Hobson and Kurki's intention is that they honour their own commitment to pluralism in regard to what a conceptual politics perspective can offer, by providing space for a counter-critique of sorts. This is the chapter by Richard Youngs (2012), who argues trenchantly that the real problem with democracy promotion is not too much emphasis on liberalism (including neoliberal economics, which is targeted elsewhere in the collection), but too little commitment to promoting democracy whether liberal or otherwise (a point for which Youngs provides ample illustration from Europe in the form of country case studies depicting the failings of European democracy promotion, in his The European Union and Democracy Promotion (2010a)). Unlike the other contributing authors to The Conceptual Politics of Democracy Promotion, whose backgrounds lie in academia, Youngs is an academic with one foot in the world of policy-oriented think tanks, which lends an evidence-based authority to his counter-critique of theorizing the conceptual politics democracy promotion.

Hobson and Kurki are largely successful then in their declared aim that the book should not read like critical theory – and more particularly theory with neo-Gramscian pedigree – talking to theory seemingly for its own sake and without providing clues to what policymakers can take away. The strength of the book's commitment to pluralism, however, might also be considered a source of weakness. Little guidance is given about what the limits of democracy are and where the theoretical alternatives to liberal democracy overstep the bounds of democracy. They say pluralism does not mean moral or analytic relativism – recognizing a ‘plurality of potential kinds of democracy . . . does not force us to accept that all democratic forms are equally democratic or democratic in the same way’ (Hobson and Kurki Reference Burnell and Gerrits2012: 221). But a confident sense of democracy's minimum necessary constituents does not emerge from the book. Social democracy as found most conspicuously in Scandinavia comes through as a favoured alternative in chapters by Sheri Berman and Heikki Patomäki (although presumably social democracy too must be considered a contested concept), and Jonas Wolff's chapter portrays some kind of democracy currently under construction in Bolivia. But the book would have been strengthened by providing greater illumination on the full range of theoretical alternatives, on where they might be appropriate and where they could plausibly develop if democracy promoters took the trouble to offer the right kinds of support. Hobson and Kurki acknowledge that ‘major empirical examples of alternative models of democracy, they are admittedly few and far between’ (2012: 219). Democracy promotion could be left with the onus of offering practical support for certain ideas of democracy that are not properly tried and tested at home or are considered too risky there and possibly entail a foreign involvement in other societies that is (even) more intrusive than usual because they require engagement across a broader range of social, political and other affairs. Saying that ‘one needs to distinguish between adopting a critical theory approach to considering liberalism's role and being critical of it’ (Hobson and Kurki 2012: 218) leaves behind some doubts about how much abandonment of liberalism, and which aspects precisely, the conceptual politics of democracy promotion is comfortable with. If democracy really is an essentially contested – that is to say disputed – concept and not essentially contestable (so internally contradictory as to be incoherent or fundamentally confused), then democracy promotion actors, whose greater knowledge can be assumed to lie on the support side of democracy support, might feel entitled to expect more specific guidance on which trade-offs lie within legitimate invocations of the idea of democracy and which ones fall outside.

None of the above detracts from what is the very real achievement of The Conceptual Politics of Democracy Promotion in offering ‘a different way of thinking about democracy promotion’ (Hobson and Kurki 2012: 222) that should prompt further work on mapping out alternatives. One of its strengths lies in the different chapters’ cross-referencing back to the editors’ framing agenda. For example, in his chapter, Laurence Whitehead argues in support of biological metaphors for characterizing democracy promotion understood not as the mechanical transfer of some political design principles (the practice that Carothers objected to back in 1997) but ‘rather as a cooperative process of “cultivating” or even “nurturing” locally pre-existing democratic potentialities’ (Whitehead Reference Burnell and Gerrits2012: 21). This does not look a million miles away from the way previous writers have conceived the ideal relationship between international support and democracy conceived as the product of struggle for change originating from within society and (sometimes) from the grassroots below. Democracy promotion has paid lip service to this for many years. Whitehead's conviction (2012: 33) that democracy support must be founded on protecting and sustaining democratic potentialities at home, as an ethical argument and a matter of credibility and intellectual coherence, also looks unexceptional now in the light of the excesses conducted by governments in the West in the name of the war on terror – notwithstanding the reflection that Brown's The Dynamics of Democratization (Reference Bermeo2011: 315) offers that the well-established critical discourse on the quality of democracy in established democracies has yet to exert much influence on studies of democratization. In the West, certain policy and institutional responses currently being made to the serious public financial problems might be thought to be inflicting even more damage on democracy there, giving added force to Whitehead's point. Alas, democracy promotion's terms of reference do not extend that far; on the contrary, insiders say the weak economic outlook will cause its resourcing to be cut.

Piki Ish-Shalom's chapter (2012: 52) argues for a ‘participatory and deliberative understanding of democracy and democratization’ – something that must mean not one but several different possibilities. He notes that there are ‘criteria of reasonableness’ that impose limits on legitimate alternative conceptualizations to liberal democracy but does not identify them. Once again, the advice offered to ‘policy-oriented scholars’, which is to invest in helping to ‘construct a civil society of informed, involved, and participating citizens’ (Ish-Shalom 2012: 44) looks pretty unremarkable even if, once again, this is something the democracy promoters would probably say they find much easier to endorse than put into practice – especially in Putin's Russia, for example, where (external support for) autonomous civil society is increasingly repressed. In another chapter, Heikki Patomäki extends Sheri Berman's argument for social democracy to the global level, which dovetails well with Beate Jahn's contribution too. Richard Youngs's chapter (2012), already noted, completes the first half of the book, on ‘Orientations’. Remarkably, Youngs's chief claim (2012: 100) – that democracy promotion is simply not doing enough to further core liberal norms around the world in a way that would allow local variations in and choices over democratic reform to flourish – seems to take us full circle back to the gist of Whitehead's argument for supporting and reinforcing locally rooted democratic impulses. But for Youngs (2012: 106) this also seems to mean that more should be done to help societies prevent the benefits of (essential) neoliberal economic policy reforms being captured by narrow – often autocratic – politico-economic elites. Where other contributors see a tension between liberalism's political freedoms and economic freedoms or its attachment to private property, Youngs cautions against underestimating the true extent of popular demand for liberal democracy, especially if competing democratic conceptions mean less space for a variety of different local choices.

In their chapter among the six chapters in the book's second half, on ‘Cases’, Valerie Bunce and Sharon Wolchik's pilot research into how recipients and providers of democracy support in post-communist countries define democracy finds that social and economic benefits hardly rate, compared to freedom and more institutional factors. An additional finding, however, is that recipients are more likely to highlight external obstacles to democratic development, including problems with the democratizing strategy, goals and the nature of democracy assistance, which the authors say ‘is very closely related to conceptions of democracy’ (Bunce and Wolchik 2012: 165). The findings are puzzling; as the authors observe, further research is needed. Wolff's account depicting the conflict between majoritarian support and procedural correctness in the Bolivian case notes the reservations of US and German ‘donors’, but advises democracy promotion not to aim for a particular end. Instead, outsiders should help the people decide their political future in a peaceful, constructive and inclusive way. The anticipated response from democracy promoters: of course, but if only it were so simple; were this really possible then democracy support would be largely superfluous; and, finally, what do we do next if in our judgement the people who decide make politically illiberal or anti-liberal, not extra-liberal choices?

All things considered, it would not seem that democracy support and assistance have too much to fear from having the conceptual politics subjected to critical inquiry. But the injunction to embrace more pluralistic conceptions falls short of presenting a satisfying account of suitable concrete, feasible and democratically compatible alternatives to liberal democracy, with the partial exception of social democracy (which might not travel well from its northern European roots). This is a limitation. If an effective strategy to promote democracy must rest on a sound theory of democratization, then a theory of democratization needs a (workable sense of) dependent variable, however intermediate and open-ended or multifarious that variable is conceived to be. Only then is it possible to either agree or disagree with statements such as that which Youngs makes (Reference Gallie2010b: 12): that much which has been labelled democracy policy ‘has in practice generally been aimed at governance changes rather than democratization’. For, as Youngs rightly says (2010b: 13), the policy-making logics of support for increased governance capacity and democratization ‘are not necessarily mutually supportive’.

Whereas one chapter in each of The Conceptual Politics of Democracy Promotion (Hobson and Kurki Reference Burnell and Gerrits2012) and The Dynamics of Democratization (Brown Reference Bermeo2011) speaks exclusively to US policy, The European Union and Democracy Promotion (Youngs Reference Gallie2010a) offers a critical assessment of the world's other main source of democracy support, apart from the United Nations. Democracy promotion by European governments and the Organization for Security and Cooperation in Europe as well as the European Commission fall within this book's terms of reference. Although the book also contains cases from several regions, the exclusion of Latin America and China, for example, belies the subtitle of A Critical Global Assessment. The book predates the great soul-searching in Europe about strategy towards its southern neighbours in particular that was prompted by the ‘Arab spring’ in 2011. However, far from being rendered obsolete, the book prefigures some of the critical reflection that European democracy promoters now know they must engage in as a result of discovering how badly they had misjudged politics in North Africa especially, and how misguided were Europe's former policies and assumptions centred on stability-seeking cooperation with the region's authoritarian rulers. So there are some grounds here for quipping ‘if only independent scholarly analysis had been heeded sooner . . .’

At the same time, several chapters leave a strong impression that it would be foolish to exaggerate the power of what even a revised European strategy for promoting democracy could achieve. This is not simply because of the presence of competing or conflicting external policy goals, but because, as Youngs (Reference Gallie2010b: 11) surmises from the country case studies, the ‘domestic political structures of each country play the most potent role in explaining variation in European policies’. This is not out of line with the reasoning of Valerie Bunce and Sharon Wolchik (Reference Bermeo2011: 305) that ‘democracy is too complex, too much shaped by local conditions, and too much the product of multiple influences to move in some magical and holistic way from one country to others. It is advisable, therefore, to focus on the diffusion of specific innovations that create democratic openings or that contribute in some concrete way to democratic development.’ But chapters as varied as on Romania, the Gulf region and Central Asia seem to confirm ‘the importance of reverse direction causality: namely how domestic political dynamics explain the nature of external actors’ policies’ (Youngs Reference Gallie2010b: 11). The operation of ‘reverse direction causality’ presents a challenge to the notion of basing strategies for democracy support first and foremost on a sound theory of democratization, not least in countries where the political dynamics show little or no respect for democracy. Although it is true that a better understanding of the conditions abroad under which Europe's external policies can affect the domestic political dynamics is needed (Youngs 2010b: 2), this alone cannot guarantee a more effective democracy promotion so long as European policy calculations vary and must be ‘mediated through domestic political structures inside Europe’ too (Youngs 2010b: 11).

A different way of expressing Youngs's main point here is that democracy support will have more purchase where/when the tide is running in favour of democratic change (‘international influences are influential only in so far as they resonate with a receptive local environment’, say Valerie Bunce and Sharon Wolchik apropos post-communist countries; Reference Bermeo2011: 305) – not an original thought but one that gives some reason to believe that democracy support can now make a more positive contribution to democratic development in a region such as North Africa than it did previously. This is just about consistent with Brown's overall verdict on democratic diffusion (Reference Bermeo2011: 2): international action can indeed shape possibilities for democratization and the path it takes but does so in varied, unanticipated and sometimes long-term ways that ‘will likely frustrate conscious democracy promotion polices’. So, can democracy be exported?

Marc Plattner in his Introduction to Is Democracy Exportable? (Reference Barany and Moser2009: 11) replies to the question by telling us that the ‘prevailing tone of most of this book is one of relatively benevolent skepticism’. Most of the 14 contributing authors ‘seem to harbour more or less profound doubts regarding the ability of democracy promotion to accomplish its aims’ (Plattner 2009: 12). A noteworthy exception is the statistical study of the United States Agency for International Development's (USAID) assistance to democracy and governance (the two are joined at the hip) in 165 countries by Mitchell Seligson, Steven Finkel and Aníbal Pérez-Liñán. They found modestly positive returns over the period 1990–2003, but it is fair to say that since the findings were first made public a degree of scepticism has been expressed about their value, particularly in Europe (where no comparable study exists). This is partly due to old controversies pitting qualitative against quantitative methods of assessment. It perhaps also signals that the study fell victim – however unfairly – to an impression that such is the political process by which government budgets must receive Congressional approval that US democracy support has to show that it achieves measureable results, or impact, which is what this USAID-commissioned research delivered.

Nancy Bermeo's chapter in Is Democracy Exportable? has the status of Conclusion and offers a considerable contrast. Although noting that ‘the promotion of democracy as an idea has been triumphant’ (Bermeo 2009: 249) (even if less true of liberal democracy now), she comments that its popularity owes something to the fact that it means different things to different people. This is not something that Hobson and Kurki would necessarily see as inherently problematic, but for Bermeo it looks more suspicious. The label democracy has also been (mis)appropriated by non-democrats. Furthermore, the idea's appeal ‘has probably resulted less from deliberate attempts at promotion by governments than from how democracy itself has been seen (and imagined) to work in actual states’ (Bermeo 2009: 251) – appraisals that, as already explained, are less rosy now than previously. The case Bermeo mounts against (international pressures to install) neoliberal economic reforms which undercut the very institutions that democracy assistance seeks to strengthenFootnote ⁴ and feed democracy-damaging social inequality is overtly in line with the conceptual politics of democracy promotion as explored in Hobson and Kurki's book (Reference Burnell and Gerrits2012).

Compared to Hobson and Kurki's book, Is Democracy Exportable? (Barany and Moser Reference Barany and Moser2009) actually comes somewhat closer to implying that democracy promotion should not even be attempted. Several reasons are provided, including conflict with a liberal respect for the self-determination of nations (Thomas Pangle); the risk of encouraging a vibrant civil society that may be ‘neither an indicator of nor a precursor to healthy democracy’ (Sheri Berman 2009: 51); the presence of a ‘prickly’ nationalism that makes foreign democracy promotion a wasted effort, if not counter-productive (M. Steven Fish 2009); and democratization's tendency to expose problems of difference and intolerance (Adam Seligman). Surprisingly, 12 years further on from Carothers's observation that strategy for promoting democracy neglects lessons from domestic political experience in the West, we find Daniel Chirot (2009: 99) claiming that it is still the case that ‘many of those who propose to spread democratic reforms around the world, particularly Americans, have forgotten the history of how democracy evolved in those Western countries where it originated’. The point Chirot wants to make takes a more sinister twist when he goes on to say the smooth transition to democracy in much of eastern and central Europe after 1989 would have been much harder ‘had most of each state's major minorities not been massacred or expelled in the twentieth century’ (Chirot 2009: 106). The moral for spreading what Chirot calls ‘tolerant democracy’ is to wait until the right conditions or preconditions are in place: ‘patience, generations of hard work, and a willingness to accept very incomplete and different versions of that democracy for a long preparatory period’ (Chirot 2009: 109). The critical stance that Hobson and Kurki (Reference Burnell and Gerrits2012) take towards promotion looks more upbeat by comparison – less because of any difference about the practicalities of promotion and more because of greater optimism about the potential for developing acceptable varieties of democracy.

To conclude this review, the literature on democracy promotion is now showing signs of coming of age. It has evolved from early debates on the official policy rationales or motives behind it, through empirical studies evaluating the performance of democracy assistance (still a major field of inquiry) to the kind of democracy that democracy supporters should try to help spread. The logic of the argument that an effective strategy for promoting democracy is served by anchoring it in a sound understanding of democratization is as valid now as it ever was. Knowledge of the determinants of democratization has moved forward but we are left with the sense that the process is highly complex, influenced by multiple factors, often long drawn out and vulnerable to interruptions and retreat. There is still much to find out about the part played by reciprocal interaction between international factors, including democracy support and domestic factors. The books in this review share in varying degrees the general consensus that democracy support can make a difference. But promotion is constrained to work in, with and on circumstances that by and large are not of its own making and which can exert their own influence on democracy promotion (policy).

Although only a fraction of these books’ 2,500 pages has been touched on here, and much rich material has not been reported, what should be clear by now is that academic literature even where sharing some affinities with critical theory stands ready to offer suggestions for improving democracy support. These vary from changes at the micro level such as in election monitoring through being more flexible about what democracy means to identification of where among the different types of authoritarian regime democracy support might have the most traction, and the kinds of support to select. Democracy promotion is beginning to be embedded in new literature on democratization even if democratization theory has yet to be fully embedded in democracy support. This looks like advance.

As for the democracy promoters, they must make their own judgements. They are certainly not freed from having to make choices of their own. For, while analysis of the sort exemplified in these books can take us further down the road of being able to distinguish the more challenging from the more promising cases for democratic progress and for democracy support, questions about whether to concentrate their limited resources on the former or on the latter remain for them to determine. On no account should democracy promotion's power to achieve its stated ends be exaggerated relative to the larger set of international influence conceived much more broadly and relative to domestic political factors especially. Meanwhile, exactly how the large, diverse ‘network’ of democracy promotion institutions now makes decisions on democracy support, and more particularly the role that independent scholarly inquiry into democratization and democracy promotion actually plays in the policy process, remain outstanding as areas that further new research beyond these books could do yet more to address.