Can we draw a sharp distinction between regimes that are democratic and those that are not?
If so, what are the criteria? If not, why not?
'[Scholars] use the term in whatever way best fits their own research purposes, funding needs,
and advertising strategies, while the usage of the same key term maintains the illusion of a
common theoretical enterprise, a common purpose, a common language, a common
“dependent variable.” (Schedler, Andreas, 2009)
Arguably, to be able to draw a sharp distinction between regimes that are democratic and
those which are not, a single lucid definition of ‘democratic regime’ as compared to for
example a dictatorship which arguably does not amongst political scientific scholarship exist.
In fact, it is the case that swathes of scholars resort to differing requirements for the
classification ‘democratic regime’ which exists in contention amongst Polity IV’s, Freedom
House’s, Democracy and Dictatorship (DD) Measure and Selectorate qualifications.
Approaching each of these structures with the tools of validity, reliability and replicability it
should become more evident what the shortcomings of each categorisation are and thus the
current inability of ‘we’ the political scientific community to come to a consensus on what
sharply distinguishes the democratic regime from other regimes.
The basic conceptualization of democracy in (DD) has two founding components:
governmental offices and contestation (Cheibub, Gandhi, Vreeland, 2010). Contestation is
then divided into three other premisses: prior to the election the outcome remains undecided
(ex ante uncertainty); after the election the result is irreversible (ex post irreversibility); these
elections following the first two premisses occur at repeated and known intervals (Przewoski,
1991). In operationalizing their conception of democracy the makers of (DD) have the
conditions: the chief executive and the legislature are both elected; multiple parties compete
in elections and following these elections an alternation of power has taken place. Out of
these conditions the last may be the hardest to understand but the argument presented by the
authors is supposedly that until we observe that the chief executives are willing to relinquish
their power, we are uncertain if those in power are willing to follow democratic convention
and hence whether the regime is democratic or dictatorial. Peculiarly, this means countries
such as Japan which most would hardly contend to be undemocratic are seemingly dictatorial
till proven otherwise by an alternation of power from 1955 to 1993 where the Liberal
Democratic Party remained in power. Another point of contention is the dichotomous
approach to dictatorships and democracies wherein the authors of (DD) refuse to
acknowledge a hybrid system unlike Polity IV and Freedom House which do allow room for
hybrid regimes. Further confusion may come from the choice of a minimalist approach over
the substantive approach which although has advantages in isolating causal processes it could
also be seen as a failure on the part of (DD) to places emphasis on specific cornerstone
institutions which uphold democracy such as the judicial system. Another problem comes
from the question of its validity as scholars argue that the (DD) measure can be seen as
defective as it ignores Dahls component of inclusion imperative to democracies.
Following other scholars who take democracy and freedom to be synonymous the Freedom
House Measure can be used to categorize democracies. There are two main components
which branch out into sub-components in the measure, these are civil rights and political
rights. Civil rights is scored out of 60 and then converted into a 1 to 7 score whereas pollical
rights is measured out of 40 and given a 1 to 7 score. Freedom House takes the substantive
approach and says that whilst minimal and thus institutional factors are necessary, they alone
are ultimately insufficient. However, the more holistic approach which encapsulates
outcomes does have a trade-off in terms of practical use of the measure as in measuring
causes it is circular to use a measure which encapsulates effects. Furthermore, the Freedom
House measure is generally unreliable as scholars are asked to place their own value
judgement to quantify how independent media are or on reasonable self-determination
meaning two people on the same nation can feasibly get differing results. On top of it being
unreliable, the validity can be brought into question due to the effect of aggregation meaning
there are many ways to get the same Freedom House score, which then draws into question
whether all components should necessarily be weighted equally such as academic freedom
and equal electoral campaigning opportunities. These result in a measure which in its
unreliability and invalidity is depreciated by being generally irreplicable as it provides no
clear coding rules to explain why necessarily a country has a given score. Compounding in a
measure of democracy which seems totalising and yet defunct to the political scientist.
Polity IV follows Dahl’s, as does Freedom House, approach to democracy as a continuum
employing an interval measure hence, they employ some form numerical scoring and
aggregation. The Polity IV score from -10 to 10 accounts: for the competitiveness of
executive recruitment; the openness of executive recruitment; the constraints on the
executive; the regulation of political participation; the competitiveness of political
participation. However, unlike Freedom House, Polity IV takes a minimalist approach which
makes it seemingly more practical in isolating causal processes whilst not capturing all of
what we mean often when we use the term democracy. A concern in the continuum concern
might be that it supposes a midpoint between dictatorship and democracy which is hotly
debated as existing (Bollen and Jackman, 1989) or as absurd as ‘there is a natural zero point’
(Alvarez et al. 1996, 21). Moreover, another issue arises as the minimal, interval and cogently
coded manner of Polity IV – which gives it replicability – may mean political scientist are
prone to overlook the arbitrary nature of the numerical threshold they create for a country to
be considered democratic; this conundrum has resulted in researchers being ‘overly sanguine
about the properties of the Polity democracy scale applied in statistical work’ ( Terrier and
Jackman 2008, 201).
The Selectorate theory posits that the differentiating factor between democracies and non-
democracies is their winning coalitions and selectorate size. (Bueno de Mesquita, Smith,
Siverson, and Morrow, 2003) The proposal is that with a winning coalition of above 0.75 and
selectorate above 0.95 you can classify that state as a democracy and often they take its
effects of proof of validity for example: countries that share this W/S often tend to have an
increased provision of public goods which sit in trend with that of generally agreed upon
democratic regimes. The reason given for this to be the case is that when there is a large
selectorate the chief executive has many it needs to please and when combined with a large
winning coalition they has to ensure people do not turn to their political rival. Essentially
when the selectorate is large and the loyalty norm is low the chief executive acts more
vigilantly and judiciously in appeasing and being responsible to the many citizens they must
please through the provision of public goods. In an indirect manner it manages to encapsulate
Dahl’s requirement for inclusion as the authors of selectorate theory establish a clear
dichotomy between the selectorate and disenfranchised and place that as a key component in
the behaviour of the state. Likewise, the winning coalition which is reflective of the
opportunity for an alternative political actor to assume power indirectly accounts for Dahl’s
point on contestation. Subtly, it could be argued that though the selectorate argument
accounts for the number of people allowed of vote it doesn’t account for which type of
people, resultantly, though in practice the places with the most selectorates will tend to be
those with universal suffrage, they could also be assigned the vote based on gender or
property rates and as long as the number stay the same they would pass in behaviour for
democratic regimes. The importance behind this point is that it means the principals that
underpin the democracy in its essence as captured by the right to vote may be undemocratic
yet by this selectorate metric appear democratic.
Ultimately, the means by which we would make the sharp distinction of democratic and
undemocratic are flawed due to the trade-off between minimal and substantive evaluations.
This has spearheaded the outlook that political scientists ‘use the term in whatever way best
fits their own research purposes[…]’ .To answer the question, I think if pressure is put on the
word ‘we’ then, evident in the variation in approach, it is not the case that we can make clear
and sharp distinction because many of the choices for democratic measurements are highly
contestable like continuum versus dichotomy, moreover, even if it were not the case that the
terms were contestable, all bar selectorate theory draw issues either with validity, reliability
or replicability.
Bibliography:
Clark, William Roberts, Matt Golder, and Sona Nadenichek Golder (2012f), “Chap- ters 5
and 10: Democracy and Dictatorship”, in Principles of Comparative Politics, 2nd ed.,
London: CQ Press.
Schedler, Andreas (1998), “What Is Democratic Consolidation?”, Journal of Democ-
racy, 9, 2, pp. 91–107, DOI: 10.1353/jod.1998.0030.
Alvarez, M., Cheibub, J.A., Limongi, F. et al. Classifying political regimes. St Comp Int
Dev 31, 3–36 (1996). https://doi.org/10.1007/BF02719326
Bruce Bueno de Mesquita, James D. Morrow, Randolph M. Siverson and Alastair Smith
British Journal of Political Science
Vol 32, No.4 (Oct 2002), pp. 559-590 (32 pages)
Published By: Cambridge University Press
Kenneth A. Bollen and Robert W. Jackman
Democracy, Stability, and Dichotomies
American Sociological Review, Vol. 54, No. 4 (Aug., 1989), pp. 612-621 (10 pages)
Published By: American Sociological Association
