Conventionalism as an Adequate Basis for Policy-Relevant IR Theory
| DOI | 10.1177/1354066108100056 |
| Author | Fred Chernoff |
| Published date | 01 March 2009 |
| Date | 01 March 2009 |
| Subject Matter | Articles |
Conventionalism as an Adequate Basis
for Policy-Relevant IR Theory
FRED CHERNOFF
Colgate University, USA
This article considers three factual observations about the history of the
study of International Relations and examines how well several different
metatheories of IR can account for them. The three facts are, first, that
there has been persisting disagreement between supporters of contend-
ing theoretical approaches; second, that there have been occasional
cases in which opposing scholars have converged on certain conclusions;
and third, that the field of IR was intended by its founders to have some
bearing on policy and some capacity to help change the world. The art-
icle contrasts several well-known philosophical principles on which
metatheories have been based. The article concludes that all three chal-
lenges can be met by only one such metatheory, which I term ‘causal
conventionalism’, based in part on principles developed a century ago
by Pierre Duhem.
KEY WORDS ♦conventionalism ♦Duhem ♦democratic peace
♦International Relations ♦metatheory
1. Introduction — Agreement, Disagreement and Policy Value
In the past 20 years there has been growing interest among IR theorists in
questions of metatheory.1Much of the substance of what they have pro-
duced, however, undermines the possibility that IR could have any empirical
relevance to the world of policy-making, which is contrary to the aims of the
post-World War I founders of the discipline of IR. This article identifies three
facts from the history of IR theory and argues that the need to account for
them constitutes three challenges that must be met by any adequate
metatheory of IR.
European Journal of International Relations Copyright © 2009
SAGE Publications and ECPR-European Consortium for Political Research, Vol. 15(1): 157–194
[DOI: 10.1177/1354066108100056]
European Journal of International Relations 15(1)
158
One challenge is the need to account for the well-documented, centuries-
long debates between political realists, liberals, idealists, Marxists and others.
The fact that scholars from different theoretical traditions have not through
the centuries reached common answers to some of the most vigorously
debated questions is a striking feature of the history of most social science
disciplines, including IR. Any adequate metatheory of IR must be able to
account for it.
Second, there have also been areas of agreement, though these have been
rare and concern much narrower questions. For example, in the early 1980s
an intensive debate developed on the question of whether democracies acted
differently from non-democracies with regard to war. Since that time a great
many scholars entered the debate, realists generally arguing that domestic
‘regime-type’ does not have a significant impact on any behaviour, with many
liberals arguing that regime-type does make a difference. Most scholars in
both groups, after examining data-based studies of the ‘monadic hypothesis’,
that democracies fight fewer wars than non-democracies, and of the ‘dyadic
hypothesis’, that democracies are more peaceful toward other democracies
than any other sorts of pairs (dyads) of states are towards each other, came to
accept the former as false and the latter as true. Any metatheory that purpor ts
to account for the field of IR theory should be able to explain the cases of
approach to consensus, like the ‘democratic peace’ case and also why conver-
gences have been so rare.
A third fact about the field of IR is that it was consciously developed after
World War I in order to assist nations and leaders in maintaining the hard-
won peace and in avoiding a recurrence of the devastation of 1914–18. The
early scholars in the field believed that careful study would show what means
should be employed to prevent or minimize war. They clearly believed that
systematic enquiry could show empirical connections between particular pol-
icies and future outcomes, which requires some form of prediction
(as described below). The third challenge then is to explain how IR can have
empirical value for policy and justify belief in theory-based prediction. It
should be noted that while realists opposed the idea that these idealist goals
were attainable, they shared with idealists the view that only the systematic,
serious study of IR has the potential to guide the development of good pol-
icies; realists identified the goals of the latter in more modest and state-centric
terms.
The claim that a metatheory must account for this third fact is a different
sort of challenge from the first two in that someone might argue that, upon
deeper analysis, IR is really not capable of providing the sort of empirical sup-
port (which requires prediction) for policy-makers that theorists initially
claimed. If it were to be argued successfully that IR lacks the ability to offer
any empirical guidance that can be applied to the real world, most IR
Chernoff: Conventionalism and Foreign Policy Relevance
159
theorists would need to reorient their understanding of the field in a signifi-
cant way. They may even re-think the value of devoting their professional
lives to a field that cannot offer any possibility of making the world better
(using whatever measure of ‘better’ they deem appropriate). After four well-
known metatheoretical principles are considered below — the last of which
questions the grounds for including policy-making as a legitimate criterion
of metatheory — the article turns to that question, namely, whether the third
challenge is well founded by looking at empirical evidence and arguments for
predictiveness. The article seeks to show that the view I have termed ‘causal
conventionalism’ is the only approach that can deal with all three challenges
identified here. The metatheory defended in the following builds upon sev-
eral arguments developed in other publications, which for the sake of brevity,
will be referred to but not repeated here.2
2. Prediction and Policy-making
This article argues that if a theory is able to offer empirical guidance to
policy-makers it must be capable of prediction. The argument may be sum-
marized as follows: (1) policy-makers must make decisions; (2) policy deci-
sions require expectations about the future – a certain sort of justified belief
about future events, which, broadly defined, constitute, ‘predictions’; (3)
predictions or expectations require beliefs about patterns of behaviour, that
is, law-like generalizations; (4) law-like generalizations are derived from or
justified by theories;3therefore (5) prediction-generating theories (among
other things) are necessary for rational policy-making.4
For example, President Bush and Prime Minister Blair apparently believed
that Afghanistan hosted al-Qaeda terrorist training facilities and that Iraqi
President Saddam Hussein possessed chemical and biological weapons,
which might fall into terrorist hands (see Waldman, 2004). In their view,
Saddam Hussein’s cooperation with terrorists entailed that his weapons con-
stituted potential threats to the US, UK and other Western states. Bush
believed several conditional propositions about the future, such as, P1: ‘If al-
Qaeda and its support-system are not physically destroyed, then al-Qaeda
will likely continue to launch attacks on American citizens and American
interests.’
P1 deals with the future; it says what will likely happen at a time later
than the moment of utterance. Second, P1 is implicitly based on a set of
connected beliefs that are rational and arise from past observations. The
beliefs may include ‘terrorist organizations that are based on extremist ide-
ology, that have advocated and committed violence, and that have access to
weapons, training facilities and organization opportunities, will continue to
strike at those whom they regard as their enemies unless stopped by internal
To continue reading
Request your trial