Does grand theory shape officials’ speech?
| Author | Heather-Leigh Kathryn Ba,Timothy McKeown |
| Published date | 01 December 2021 |
| Date | 01 December 2021 |
| DOI | http://doi.org/10.1177/13540661211012060 |
E
JR
I
https://doi.org/10.1177/13540661211012060
European Journal of
International Relations
2021, Vol. 27(4) 1218 –1248
© The Author(s) 2021
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/13540661211012060
journals.sagepub.com/home/ejt
Does grand theory shape
officials’ speech?
Heather-Leigh Kathryn Ba
University of Missouri, Columbia, USA
Timothy McKeown
University of North Carolina, Chapel Hill, USA
Abstract
In what situations is the speech of foreign policy officials a reflection of speech
expounding grand theories of International Relations? Using a linear support vector
classification algorithm, we analyze all the published volumes of Foreign Relations of the
United States as a single corpus, examining the use of five frames taken from a mix
of academic theory and day-to-day practice: realism, geopolitics, liberalism, Cold War
politics, and a pragmatic frame. We also investigate how the use of these frames varies
across subject matter and in ways that are consistent with some simple hypotheses that
connect frame use to situational factors.
Keywords
Realism, liberalism, diplomacy, foundational theory, foreign policy, theory and practice
Introduction
Do general theories of international politics such as realism and idealism have any con-
nection to the everyday speech of foreign policy decision-makers? Recent considerations
of the relationship between “grand theory” and foreign policy (Eriksson, 2014; Jahn,
2017; Zambernardi, 2016) have suggested that the connection is to be found in the way
in which these theories shape the speech of these officials. Although these IR grand theo-
ries are highly abstract and non-operational, they provide a language for communicating
about International Relations that foregrounds some concepts and relationships while
Corresponding author:
Heather-Leigh Kathryn Ba, University of Missouri, Professional Building 113, Columbia, MO 65211-6080,
USA.
Email: bah@missouri.edu
1012060EJT0010.1177/13540661211012060European Journal of International RelationsBa and McKeown
research-article2021
Article
Ba and McKeown 1219
backgrounding others. They are the substrate upon which discussions of more specific
policy measures are supposedly constructed.
Bueger’s (2014) review of this work situates the emergence of a substantial empirical
research interest in such questions in the 1990s, but also notes much earlier efforts, such
as Mitrany’s (1943) work on functionalism.1 The accumulation of case studies, elite
interviews, participant observation reports, and surveys illuminates particular episodes,
but coverage is dominated by contemporary cases, and these are the very situations
where government restrictions on the release of security classified material are most
likely to impede an examination of how officials talk among themselves, far from the
public eye.
Because much grand theory is quite old (Seabury, 1965), its influence on the speech
of diplomats could have been manifest any time within the last 200 years. Claims that
grand theories have passed out of fashion within the academy (Mearsheimer and Walt,
2013), even if correct, do not provide any basis for drawing a parallel conclusion about
their relationship to officials’ speech, past or present. Theorizing about factors that affect
the relation between grand theory and official speech is quite recent, and neither theory
nor previous empirical research provides a well-established account of the nature of their
connection. It would be risky to assume that a given body of theory is read and used by
officials in the same way as was done a century ago, but the question of how theories
have been put to use can only be resolved by developing ways of measuring similarities
and differences in that usage. We thus begin with an agnostic view of the stability of the
relationship between these theories and how officials talk and write, and treat this for
now as an empirical matter.
We seek to address these challenges with a research strategy that examines the stated
views of foreign policy decision-makers over as wide a range of international contexts as
possible, while simultaneously examining the influence and applicability of IR theory to
the making of foreign policy. We use the State Department’s definitions of the situation
(Snyder et al., 1954)—in this case, the agency’s categorization of the foreign policy and
national security work product into distinct blocks of text that are labeled and separated
from other blocks—to divide a corpus of documents covering United States foreign pol-
icy from 1861 to 1985 into some 5668 partitions. (The Foreign Relations of the United
States (FRUS) categorization of both country and subject files is broadly similar to the
way that foreign affairs and national security documents are typically organized in presi-
dential libraries and in the National Archives). For each segment of this corpus, we
employ support vector classification (SVC) to examine its text and categorize it in terms
of one of four frames that exemplify a particular way to discuss foreign policy: realism/
neorealism (hereafter simply realism), liberalism, Cold War politics, and a pragmatic
frame based on a journalistic record of international events. (A sixth frame, constructiv-
ism, is not used because its presence before the post-Cold War era was so slight that a
meaningful corpus of constructivist theorizing for that period is not available). Third, we
use multinomial logit regression models to test some simple hypotheses about how the
subject of the partitions in the corpus, or the broad contextual characteristics of
International Relations at the time the partitions were written, might account for which
frame was used. Doing this over as large a domain as possible provides us with some
ability to assess the stability and generality of these relationships.
To continue reading
Request your trial