Barbara Geddes, “What do we know about democratization after twenty years?” Annual Review of Political Science 2 (1999), pp. 115-144.

_Summary_: synthesizes the results of studies of late-20th century democratization from the 1980s and 1990s. Strong evidence supports the claims that democracy is more likely in more developed countries and that regime transitions of all kinds are more likely during economic downturns – but very few of the other arguments advanced in the transitions literature appear to be generally true. This study proposes a theoretical model, rooted in characteristics of different types of authoritarian regimes, to explain may of the differences in democratization experience across cases in different regions.


_Key point_: argues that military regimes differ from single-party and personalist regimes because most officers value the unity and capacity of the military institution more than they value holding office (!?) – which leads them to cling less tightly to power than do other kinds of authoritarianism and, in fact, often initiate transitions.


_Method_: statistical analysis of a data set that includes 163 authoritarian regimes.





*most military transitions being with internal disagreements and splits. Transitions from military rule are usually well underway before protests swell.

*most personalist regimes, however, maintain their grip on power as long as possible. As a result, they are more likely to be overthrown by popular uprising or rebellion.

*most military regimes end in negotiation, which accounts for the emphasis on bargaining and the advantages of moderation.

*most personalist regimes, however, end in coups, many of them accompanied by widespread violence.

*leaders of personalist regimes also negotiate when under pressure from leaders or faced with widespread public protest, but the proportion who renege on the deals they make has been very high.

*single-party regimes under pressure from donors and popular opposition are more inclined to negotiation than are personalist regimes.

*like officers, single-party cadres can expect life as they know it to continue after liberalization or even regime change. If they cannot avoid regime change, they are better off in a democracy than in some other form of authoritarianism.

*previously hegemonic parties have remained important in political life wherever countries have fully democratized, but they have been outlawed and repressed in several that did not.


Richard Anderson et. al. (eds.). Postcommunism and the Theory of Democracy. (2001).

From Book Reviews:”


_Main Argument:_

They each challenge the current received wisdom in accounts of democratization, using evidence from the post communist region. The authors are developing an alternative to the old “preconditions” school of explaining the emergence of democracy, which saw economic development, economic equality, modernization, class structure (including a strong bourgeoisie), ethnic homogeneity, or a strong dose of Protestantism as preconditions of democracy. They argue that we should expand the range of elements of the democratization process we consider. They claim that dispersion of political power is important for democracy. They support an understanding of democratization, which is not deterministic, but allows for human agency.


_Method:_ Four substantive sections, each written by one of the listed author. Neoinstitutionalist perspective. Common thread is that the fragmentation of elites is of fundamental importance for democratization.



== Notes ==



_Philip Roeder: The Rejection of Authoritarianism_

–     An examination of why the old authoritarian regimes were rejected in the first place

–     He argues that elements of the old ruling classes were more likely to turn toward a democratic solution at the point when elite fragmentation prevented the old authoritarian “constitution” from working effectively

–     Institutional pluralism within the authoritarian states was the precondition for a democratic breakout system, he identifies a number of types of authoritarian regimes (autocracies, oligarchies, and exclusive republics); the most successful democratizers are states in which elites are forced to broaden their base of support beyond the old “selectorate”


_M. Steven Fish: The Dynamics of Democratic Erosion_

–     Comparative analysis of all twenty-eight postcommunist countries, Fish finds that the key independent variable is excessively strong executive power

–     Considers those postcommunist regimes that first democratized, and then de-democratized (at least partially)

–     ”Hyper-presidentialism” is the best explanation for backsliding


_Richard Anderson: Origins of Russian Democracy_

–     Focuses on shifts in linguistic discourse that begin to undermine the integrity of the formerly hegemonic ruling blocs

–     The importance of a strong, distinct, elite identity for the preservation of authoritarian patterns of rule, and of the breakdown of that distinctive identity (and the social boundary it maintains between the elite and the population) for the beginnings of democracy

–     The willingness of the members of the nomenklatura elite to begin to compete with one another for external popular support, and the willingness of the population to take sides in this competition

–     Democracy emerges as the elite begins to speak a form of Russian measurably closer to that of the population at large


_Stephen Hanson: Defining Democratic Consolidation_

–     Looks at the problem of the consolidation of democracy in general

–     He argues that democracy can be considered relatively secure only when a sufficiently large proportion of the political class and its associated officialdom have internalized the democratic regulation of their own behaviour

–     He asserts that democratic consolidation cannot be seen in isolation from regime consolidation –> Russia’s flawed democratic consolidation is conditioned by geopolitical and sociological factors

–     Proposes a middle way between these two alternatives: A democracy is consolidated when at least the “enforcers of democratic institutions” are seriously committed to democracy



_Critique:_ Although Fish’s approach, especially when combined with Roeder’s, provides a rich empirical and theoretical basis to take the debate forward, the attempt to hold other variables constant, including, one might stress, the role of contingent factors like personalities and other agency issues (including weak oppositions), together with structural factors such as international conjunctures, proximity to the European Union, ethnic composition, religious tensions, historical memories of statehood and the like, renders the argument rather less than convincing.

Kathleen Collins, The Logic of Clan Politics in Central Asia: Its Impact on Regime Transformation (2006), Chapter 2.

_Summary: _ Collins identifies and articulates the ‘clan’ as an important actor, and investigates the impact they have upon regime trajectories in modern states. Clans are distinct from other groups in that they rely on _kinship_ as the core foundation of relations and identity, and a _network_, which is the organizing principle of the unit.  These two factors supply a context for rational action to occur (limits). Clans serve several roles, from bolstering the effectiveness of socioeconomic interactions to reinforcing internal identity norms.  Importantly, clans cannot be understood in terms of ethnicity or ideology – they narrowly pursue their own economic and political interests and form sub-ethnic, and sub-national units.  They also cannot be confused with tribes, clientelism or corruption (which are institutions, not identities) or mafias (which do not require a kinship component).  Clans still persist in modernity due to their internal properties (identity is slow to change), but also because of their interaction with external factors in certain regions. They are more likely to persist in modernity when there is: i) late state formation (colonialism); ii) late formation of a national identity, and; iii) the absence of a market economy.  On each of these formations, the clan supplies mechanisms for overcoming the problems which the above seek to address, thus they persist and undermine modern states.


Collins makes the following propositions concerning the politics of clans

* 1. Clans may persist under strong colonial states under certain conditions, and may gain power under weak, declining ones

* 2. Clan pacts respond to threats and foster regime durability

* 3. Elites, ideology, and formal institution have only a short-term effect

* 4. Under transitional uncertainty clan politics emerges, pervading formal regimes and weakening regime durability in the longer term


Clans ————> Formal Regime ———-> Informal Regime of —–> Declining Regime Durability

(Kin Patronage)     (Democracy or Autocracy)    Clan Politics

(Asset Stripping)

(Crowding Out)

Theda Skocpol, “Bringing the state back in: strategies of analysis in current research,” in P. Evans, D. Rueschemeyer and T. Skocpol, eds., Bringing the State Back In (1985).

Summary: Skocpol advocates “bringing the state back in” to a central place in explanations of politics and policy formation, arguing that society-based, class-based (Marxism) and economic (Keynesianism) explanations of political behaviour are incomplete. Rather, Skocpol argues that comparative-historical research that focuses on state autonomy and capacity to affect policy change will help build a new theoretical understanding of states in relation to social structures and individual-level political activity.

Method: This piece is a call to action for political scientists to re-examine the role of and the importance of the state, therefore, it is more of a descriptive literature review that suggests ways for scholars to bring about formal theory in the future.

Important Insight: Skocpol (re)introduces the concept of the bureaucratic structure in her definition of the “state”, comprised of individual actors with various interests and levels of expertise. Assigning specific roles and preferences to the bureaucracy (both as a whole and on a micro-level) had been absent from political science’s assessment of the state and the policy-making community.

Critique: Skocpol’s description of the state is somewhat underspecified. It is unclear whether she is referring to the legislative or policy-making structure, the bureaucracy, interest groups or a combination of several actors.

  • Initial focus on society-centred explanations; government not taken as a separate actor
  • Weber: administrative, legal, extractive and coercive organisations are the core of any state [7]
  • States stand at the intersections of domestic socio-political orders and transnational relations [8]
  • Skocpol looks at state autonomy arguments and capacities of states as actors trying to realise policy goals  states impact content and workings of politics [8]
  • Heclo: civil service administrators have made more important contributions to government policy than political parties and interest groups [11]
  • State autonomy is not a fixed structure  crises may precipitate actions [14]
  • Autonomous actions will take forms that reinforce authority, political longevity and social control of state organisation [15]

George Hendrik von Wright, Explanation and Understanding (2004 [1971]), Chapter 1.

Summary: scientific inquiry (seen broadly) presents two main aspects: ascertaining and discovering facts (descriptive science); and the construction of hypotheses and theories (theoretical science). Theory-building serves two main purposes: prediction and explanation. (1) Two main traditions can be distinguished in the history of ideas: Aristotelian (teleological and finalistic) and Galilean (causal and mechanistic). (2) This paper looks at recent (mid-19th century onwards) developments in these two traditions with regard to methodology. (3)

Mark Okrent, “Hermeneutics, transcendental philosophy and social science”, Inquiry (1984), pp. 23-49.

Summary: argues that neither transcendental or hermeneutic arguments (based on Heideggerian or Wittgensteinian premises) can be used successfully to show an impt or essential difference b/w natural and social science. It does this by examining arguments proposed by Winch and Dreyfus, showing how they are fallacious and misconstrue the import of the premises upon which they are based, and generalizing these objections to the transcendental and hermeneutic styles of argument in the field as such. (23) In contrast to these positions, he argues that there is a need to make distinctions between nature and spirit (i.e. that we can predict what will come out of someone’s mouth without knowing what they mean), the inability of fact to ground justification, the assertion that ‘meaning’ does not preclude human behaviour, and that no substantive factual, ontological, or methodological claims can be based on these distinctions.

  • the transcendental argument (e.g. Winch): based on the argument that one must distinguish two different types of questions in regard to man: on the one hand, it is possible to ask what specific qualities man has, or what particular predicates are true of human beings, or what indvl stimuli precede which indvl motions. On the other hand, one may ask not about particular characteristics of human beings, but rather about what it means to be a human being in general, or what web of concepts are used or must be used in describing and discussing human beings and their activities. This second type of question does not concern which specific things are true of human beings and their behaviour, it concerns what characteristics could be ascribed to persons, and what it would mean to ascribe a property to a person. The first group of questions is ontic, the second ontological – and the latter is prior to the former, but it is also pre-eminently philosophical rather than empirical. These issues are settled by a priori conceptual analysis rather than by empirical research.
    • As such, the methods or concepts of natural science are not applicable to the study of man. (23-24)
      • because there could be no knowledge w/o social rules or shared practices, human behaviour can only be known as following social rules or in terms of the shared practices of the person studied. That is, necessary conditions for the possibility of knowledge cannot be used as the premises for a deduction of a specific method which must be used to study human beings, a method which would be different from that used to study non-human objects. (33)
        • But to argue in this way is to confuse the conditions under which knowledge could be justified with ontologically or transcendentally necessary characteristics of objects known.
  • the hermeneutic argument (e.g. Dreyfus): based on a distinction b/w those investigations which use a method for the sake of developing an explanation and those discourses which interpret for the sake of developing an understanding. On this view, social science has as its aim the understanding of a variety of discourses which articulate a variety of worlds. Following early Heidegger, to understand something is to have that thing related in a particular way – the thing can be understood in that it stands in a context of purposes (i.e. a hammer is understood insofar as we see that it can be used to drive nails). Interpretation, for Heidegger, is the explication of such understanding, the explicit grasp of the object as a hammer. Language, on the other hand, is seen as the primary articulation of the general context of such purposes and possibilities, i.e. that which breaks up the totality of purposes into specifiable meanings and objects. As such, language provides the original interpretation of the world, where ‘world’ is understood as a related structure of purposes, uses, and possibilities, all oriented toward some possibility of human being, that for the sake of which everything in the world is as it is (as possibility for use). To understand the language of an other, then, is to situate the manner of organizing a world of purposes, projects, and possibilities, in short an understanding of an other, in the world of the investigator.
    • As such, the investigation is an investigation of the understanding and self-interpretation of the world of the other which is carried out for the sake of some purposes or possibility for being of the other, in the world of the investigator. (25)
      • For this position the social sciences are radically distinct from the sciences. Their ‘object’ (what is investigated) is understanding, not things, their aim is understanding, not explanation, and their method is hermeneutic, not empirical… Insofar as their objects are human beings and their activities, their aim must be hermeneutic understanding and not scientific explanation.
        • practical (viz. understandings within a community) rather than theoretical wholes allow for the possibility of understanding, and practical wholes are irreducible to theoretical wholes… The social sciences study human behaviour, not present-at-hand things, and as such the practical skills and understanding which grant the possibility of social scientific investigation are also characteristic of the object studied by that investigation, because the practice of social science itself is an example of human behaviour… Contrast this with natural science which is a practice for showing physical properties independently of their involvements in practical concerns (39, 41, 40.)

Alasdair MacIntyre, “Is a science of comparative politics possible?” in Alasdair MacIntyre, Against the Self-Images of the Age (1971), pp. 260-279.

  • Obstacles stand in the way of a general science of political action
  • Concerned with the ability to use legitimate comparative methods to test genuine law-like cross-cultural generalizations

Possible to identify political attitudes independent of institutions and parties?

  • The notion of political cultures is secondary to and parasitic upon the notion of political practice
  • Political attitudes are implausible candidates for constructing causal generalizations. What of institutions and practices?

Where environment and culture is radically different, the phenomenon is viewed so differently by those who participate in it that it is an entirely different phenomenon

  • 1) Difficulties in constructing true and warranted cross-cultural generalizations about political institutions
  • 2) This shortcoming does not undervalue the importance of the work
  • 3) This is not a new problem in political theory

We cannot identify institutions in different cultures as ‘the same’

  • A science of comparative politics necessitates a series of comparative histories.
  • Comparative history can provide us with Machiavellian maxims rather than Hobbesian laws. True? Can we formulate law-like generalizations?
  • The political agent cannot rely on law-governed regularities in his activities
  • Insistence on political science being value-free implies that justice play no part in political life
  • Predictions are not possible

Bryan D. Jones, Politics and the Architecture of Choice: Bounded rationality and governance (2001), Chapters 1-6.

Summary: makes the following arguments: (1) Human behaviour is mostly adaptive and goal oriented. (2) Because of biological limits on cognitive capacities, however, humans are disproportionate information processors. They tend to react to new information by neglect or overestimation. (3) The formal organizations created by humans aid in adaption by overcoming inherited limitations in adaptive abilities. (4) Nevertheless, some of our limitations in adaptability will show through in even the most rational of institutions. (5) As a consequence these institutions will not react proportionately to incoming information, and outputs from the most rational of institutions will be disjointed and episodic. (23)

  • systems of multiple decision makers help to deal with certain errors in rationality, largely through adaption, and despite individual irrationality, legislatures, committees, and other institutions appear to work reasonably well – as such, the nature of the environment must matter in the conduct of decision making.
    • They do have a few adaptive inefficiencies, however: (52)
      • lags in matching: adaptation cannot be instantaneous, and it is difficult to know whether an observed environment-decision match is ‘b/w equilibria’ or whether the match has reached equilibrium and is hence optimal.
      • multiple equilibria: there may be a number of local equilibria – choices that are not perfect, but are better than any other ‘nearby’ choice – and hence the decision making process may not reach global adaptation.
      • path dependency: adaptation may have occurred relative to past environments, which preclude certain present choices.
      • networks of diffusion: information is contingent on ‘who talks to whom.’ This means that certain decisions within a network may be suboptimal at the same time that others are optimal.
      • mutual adaptation and coevolution: decision makers mould their environments, and hence affect the cues that they receive.

Alexander George and Andrew Bennett, Case Studies and Theory Development (2005).

Key point: contributes to the methodological dialogue by focussing on the comparative advantages of case study methods and on these method’s ability to contribute to the development of theories that can accommodate various forms of complex causality. In particular, they advocate for process-tracing.

Process-tracing may be used to test whether the residual differences b/w two similar cases were causal or spurious in producing a difference in these cases’ outcomes. Or the intensive study of one deviant case may provide significant theoretical insights. Process-tracing can perform a heuristic function as well, generating new variables or hypotheses on the basis of sequences of events observed inductively in case studies.

George Tsebelis, Nested Games (1991), Introduction, Chapter 1.


Summary: All actors are rational. If they appear not to be, it’s because we’ve made a mistake in observing what their possible options are, or what their optimal action is. The observer tends to focus on a particular game (i.e. a politician’s single vote) while failing to see that the actor is likely involved in nested games. The actor could be competing in concurrent or nested games, or acting to achieve institutional change to affect the prospects of future games. All hail rationality.


  • Why do Actors choose the ‘non-rational’ path? Why do they opt for apparently suboptimal choices?
  • Considering rationality within a more complex context – what seems irrational in the short game is logical in the long run
  • Rationality is equated with self-optimization
  • Takes into account ‘nested games’ and ‘simultaneous games’
  • General argument: Democracies have built in situations where games are not played in isolation and, therefore, where choices may appear to be suboptimal

Nested Games: The Logic of Apparently Suboptimal Choice

  • Assumption that human activity is goal oriented and instrumental and actors try to maximize their goal achievement
  • This is a pure rational choice approach to comparative politics

Two cases where an actor chooses suboptimally (Tsebelis ignores these cases as unimportant)”

  • he cannot choose rationally
  • he makes a mistake

Two cases where the observer may not recognize the optimal course of actions:

  • Observer makes a mistake about what the optimal action is
  • Observer has a limited idea of the set of possible actions
  • Apparently sub-optimal actions are frequently causes of disagreement between actor and observer
  • The observer tends to focus on one game, while a choice can be seen as optimal when the whole network of games is considered
  • Games in multiple arenas
  • Games about the rules of the games (institutional design)
  • For games in multiple arenas, any of the actor’s moves may not be optimal with respect to the entire network of arenas in which the actor is involved
  • Institutional changes can be explained as conscious planning by the actors involved in order to increase the number of alternatives, thereby enlarging strategy space
  • In the presence of adequate information, if actors do not choose what appears to be optimizing strategies, it is because they are involved in nested games: games in multiple arenas, or insitutional design
  • (seems a bit deterministic…)