How can documents speak about practices? Practice tracing,the Wikileaks cables,and diplomatic culture

Abstract

Practice theorists favor interviews and participant observations in their study. Using insights from anthropological works on bureaucratic texts, in this article we develop methodological tools to complement these interpretive methods of data collection. We suggest a way to trace practices by systematically looking through both the content of documents and their form. We probe this approach with an analysis of 408 diplomatic cables sent by the US Embassy in Dar es Salaam, Tanzania in 2005–2009 and subsequently released by Wikileaks. We draw on these documents to tell two related stories about diplomatic practices: the first about epistemic practices and how the cables privilege certain voices and types of knowledge over others, and the second about diplomatic culture, where the cables serve as evidence of the powerful socialization processes that diplomats are subject to. This contributes to International Relations (IR) with a new approach for systematically analyzing written documents to uncover international practices.

Keywords

diplomacy documents interpretivism methodology practice theory Wikileaks

1 Introduction

The idea that organizations in global politics can exert powerful socializing effects on their members is not new in the field of IR (Checkel, 1997, 2007; Flockhart, 2006; Johnston, 2001, 2008). Documents play a central role in this process, particularly in bureaucratic organizations like Ministries of Foreign Affairs. Documents serve as material touchstones of a common culture, and anchor shared understandings and norms in a durable way that reinforces ongoing socialization processes. Indeed, arguably ‘we cannot understand government without understanding documents’ (Freeman and Maybin, 2011: 156).

However, the recent turn towards practice theory has been slow to embrace the study of documents, instead giving methodological pride of place to interviews and participant observation. We contend that there is a need for the development of new methodological tools that allows for the tracing of practices through document analysis. We show herein how we can analyze written documents to investigate culture, habits, and practices, taking US diplomatic cables as our example. Diplomats serving abroad convey their views in codified ways that resonate with how insiders of the diplomatic fields express themselves. They speak to an audience of peers that have been socialized to expect specific forms for the written communications between them. They deem some types of knowledge as legitimate and worthy of inclusion in the official record, while ignoring other types of equally available knowledge. We unravel these shared ways of thinking about and acting on the world by both looking through the cables and analyzing the content of the texts through which the collective institutional voice speaks, as well as at the format of the cables, focusing on their specialized vocabulary, their use of pronouns, and their distribution.

This study remains interpretive in nature. Even if our empirical investigation is based on a systematic analysis of a sample of 408 cables issued by the US Embassy in Dar es Salaam, Tanzania between January 1, 2005 and December 31, 2009, our interpretation of these cables is informed by a 13-month-long participant observation in Dar es Salaam in 2008 and 2010. Our analysis of the cables is guided by our knowledge of the political context in Tanzania, as well as our direct observations of diplomats’ ways of doing things, including US diplomats, in Dar es Salaam. We followed an inductive path: before beginning the analysis of the cables, our fieldwork gave us indications about which cables we should analyze, and what to look for in these cables; during the analysis, the fieldwork informed our interpretations of the results.¹ In this way, we contribute to the literature advocating the use of ‘neo-positivist methods combined with an interpretivist epistemology’ (Hopf, 2017: 1) and quantitative tools to ‘effectively complement discourse analyses and other interpretive methods’ (Bayram and Ta, 2020: 4).

Our analysis is divided into five additional sections. We begin in Section 2 by reviewing the current dominant approach to documents within the practice turn in IR. Then, in Section 3, we develop our approach for uncovering the practices latent in documents, focusing on two related techniques: looking through documents to infer the material, bureaucratic, and epistemic conditions that permitted their creation; and looking at documents to establish how they serve as sites for powerful socialization processes. We present our empirical analysis in the next two sections: in Section 4, we discuss how US diplomatic cables are the result of a socially situated and politically-oriented perspective that privileges elite ‘insider’ knowledge over non-elite, historical, and sociological forms of knowledge. Then, in Section 5, we note several aspects of the cables’ form and content which generate ‘we’ feelings and encourage American diplomats to identify more closely with national and organizational interests than with their own personal views. We conclude in Section 6 by discussing what our findings tell about (American) diplomatic culture.

2 International practice theory and the study of documents

Practices are ‘what practitioners do’: they are patterned, socially embedded, meaningful arrangements of actions done to/in the material world (Adler and Pouliot, 2011: 4). Actors develop cultural habits, unconsciously known and unspoken ways of seeing the world and acting upon it. These can easily be beyond explanation from a given actor’s perspective. As Pouliot (2008: 258) puts it, they make ‘what is to be done appear “self-evident” or commonsensical.’ Yet, patterned, here, does not mean precise repetition per se, but rather malleable yet structured iteration (Bueger and Gadinger, 2018). Indeed, with unending variation in the situations agents end up in, precise repetition may not be possible (Cornut, 2018: 720–721).

The question about how we can study practices has fueled protracted debates among IR scholars (Adler-Nissen, 2014; Andersen and Neumann, 2012; Bueger, 2014; Bueger and Gadinger, 2018; Cornut, 2015b; Kusterman, 2016; Pouliot, 2007, 2012, 2014; Schindler and Wille, 2019; Walter, 2019). We contribute to this literature by looking at how the analysis of written documents—diplomatic cables, in our example—can uncover practices, habits, culture, and socializing processes with all the consequences they entail in the international realm. After all, ‘artefacts and practices entail each other, they are mutually constitutive: practices generate artefacts, which in turn structure practices. The artefact serves as an embodiment of practice, which makes that practice knowable by others, repeatable over time’ (Freeman and Maybin, 2011: 165).

Diplomatic writings are an excellent place to apply the techniques of looking at and through documents. Large sets of documents exist, usually written in very similar formats. Moreover, many scholars and practitioners have identified national diplomatic styles, as well as a broader international diplomatic culture, undergirded by a strong set of norms, practices, and shared knowledge. Thus, it is not surprising that diplomacy has attracted a great deal of attention from international practice theorists (Adler-Nissen, 2014; Adler-Nissen and Pouliot, 2014; Cooper and Cornut, 2019; Pouliot and Cornut, 2015; Sending et al., 2015), while scholars of diplomacy have focused extensively on diplomatic culture (Constantinou, 2013; Cross, 2007; Dittmer and McConnell, 2015; Sharp, 2004; Wiseman, 2011; Wiseman and Sharp, 2007).

Two specific International Practice Theory (IPT) studies illustrate the tremendous potential in analyzing diplomatic documents. Neumann (2012) looks at how the textual practices undergirding speech writing reproduce a common diplomatic culture, while Adler-Nissen and Drieschova (2019) show how negotiation practices are enabled and structured by the circulation of drafts of resolutions and declarations. For these authors, as in this study, ultimately it is the socializing functions of diplomatic documents that are of chief concern. These documents allowed social norms and procedures to remain anchored while they are drafted, disseminated, and read. In these analyses, processes of socialization inform specific practices of textual production, which in turn shape textual outputs, which then serve as a reference point for the socialization process, bringing the circle full close.

However, these two studies give little information about how texts may be looked at systematically and interpreted; their accounts—based primarily on ethnographical work and interviews—are more about documents, rather than analyses based on them. Indeed, many IPT scholars similarly prefer participant observations and interviews, emphasizing the need to conduct empirical research directly within the sites where practices are performed (Adler-Nissen, 2014; Adler-Nissen and Pouliot, 2014; Pouliot, 2010, 2016). It can even be said that the practice turn is a reaction against what Neumann (2002: 628) labeled ‘armchair analysis,’ that is, ‘text-based analyses of global politics that are not complemented by different kinds of contextual data from the field, data that may illuminate how foreign policy and global politics are experienced as lived practices.’ For Bueger and Gadinger (2018: 3), IPT scholars ‘rather than trying to be “objective” and “distant” observers [have] to engage with their objects of investigation.’ For Navari (2011: 622), ‘a focus on discourse or text alone threatens to leave practice behind.’ As a consequence, there are few discussions in the IPT literature about how we can uncover practices using documents, severely constraining IR scholars.

On the other end of the methodological spectrum is the growing move among some political scientists, particularly in the United States, to employ quantitative text analysis (QTA) methods to study large corpuses of documents. Benefiting from the ‘big data’ explosion, equipped with newly developed computational techniques, and utilizing neo-positivist methods, QTA is on the rise in political science and IR (Connelly et al., 2020; Grimmer and Stewart, 2013). Even in the field of diplomatic studies, scholars are trying their hands at using large-N methods to study how states communicate with one another, for instance during multilateral negotiations (Bayram and Ta, 2020) or in diplomatic crises (Katagiri and Min, 2019). Particularly relevant to our article is forthcoming analysis of 163,958 WikiLeaks cables which uses matching techniques and supervised machine learning methods to unpack which lexical and meta-data features of cables best predict the classification level they ultimately received (Gill and Spirling, forthcoming).

Such large-N approaches may yield valuable insights into diplomatic practice—in the latter case, that a sizeable proportion of US diplomatic cables received higher classification levels than their actual contents seemingly warranted, probably due to the habits of American diplomats. But QTA methods typically make a mistake highlighted by Freeman and Maybin (2011: 159), who note that textual and discourse analyses ‘concerned with the content or meaning of documents [. . .] often [. . .] strip the document of the practices surrounding it.’ On the contrary, we consider that analyses of written documents ‘are a means of examining communicative practice so as to uncover signs of social identities, institutions, and norms as well as the means by which these social formations are established, negotiated, enacted, and changed through communicative practice’ (Bazerman and Prior, 2003: 3). Our approach takes documents as artefacts of practices, emerging from the ways of doing things of those who write them. For us, their form and content give access to international practitioners’ mental habits, and the sources of their organizational cohesiveness. Seen in this light, diplomatic cables are material artifacts which help underpin a constant, collective process of socialization and reify diplomatic culture.

The approach to analyzing documents we outline below is not in opposition to other ways of studying international practices. Our contention is that complementing participant observations and interviews with a systematic study of documents can help uncover socialization effects that participants or interviewees may not be consciously aware of. On the one hand, participant observations provide knowledge that inform our looking-through document analysis. Insights gleaned from the field gave us crucial information about how cables were produced and what consequences they had, making them usable sites for the reconstruction of international practices. On the other hand, the systematic analysis of documents allows for wide-ranging and diachronic analyses of socialization processes in several places, while participant observations and interviews are necessarily more limited in time and space. Documents are easier to access, circulate quickly and are more permanent than the transitory experience of practitioners on which participant observations and interviews are based, providing a simpler and less costly path to the study of practices (Kuus, 2013; Lie, 2013). Overall, the combination of methods we suggest fits particularly well with an approach that (1) emphasizes common ways of doing things, and (2) draws upon a medium-sized corpus of documents, which can be analyzed both qualitatively and quantitatively.

3 Incorporating documents into IPT: a method for looking at and through documents to uncover practices

We contend that documents can be used to uncover practice and complement participant observations or interviews, and we devise an approach to do so. Specifically, we propose two related techniques that work best for medium-sized collections of bureaucratic writings: looking through documents to reveal the conditions that enabled their production, and looking at documents to uncover the social effects they generate.

Our approach to document analysis is deeply informed by the tools social scientists outside IR have developed in order to study bureaucratic documents. As Adler-Nissen and Drieschova (2019: 6) note, ‘Beyond international relations theory, there is a large body of literature on the production of documents and their role within organizations and bureaucracies [. . .] but researchers have rarely established connection between these studies and diplomatic scholarship.’ All knowledge production practices are shaped by the place, time, and manner in which they take place. Building on works by sociologists, anthropologists, linguists, and historians that unpack the knowledge production activities of practitioners (Bazerman and Prior, 2003; Bonelli and Ragazzi, 2014; Ferguson, 1994; Freeman and Maybin, 2011; Gusterson, 1998; Harper, 1997; Hull, 2012a, 2012b; Kafka, 2009; Pennebaker, 2013; Riles, 1998), we contend here that we can unravel these shared ways of thinking about and acting on the world by looking through the cables and analyzing the content of the texts through which the collective institutional voice speaks, as well as at the format of the cables, their specialized vocabulary, their use of pronouns, and their distribution.

Looking at and through documents are complementary techniques, differing in terms of their temporal orientation. Looking through documents is backwards looking, attempting to recover information from the documents that helps situate the conditions prior to when the documents were produced. Scholars interested in understanding practices infer the material, bureaucratic, and epistemic conditions that permitted the documents to be created in the first place. The goal is to uncover the mental and habitual landscapes of the author(s) by seeing what practices have caused elements in the documents to be either present or absent. This tells us what the agents writing the documents habitually do, providing information about the mental dispositions, cultural and social norms, and common practices associated with their particular field. The kinds of questions asked when looking through documents include: What can be inferred about the ways of doing things of the author(s) of the documents from their content? What practices were necessary for the production of the documents? What types of knowledge did the author(s) of the document rely upon, and what does this tell us about their mental habits? What might have been included in the documents but was left out? Why might that be?

Looking at documents, on the other hand, helps better understand the social and political effects the documents are having on their intended readership after their production. Scholars should place themselves in the mindset of an intended, habitual reader of the documents and use structural, syntactic, and semantic signposts in the documents to establish what the probably unconscious socialization effects of the documents are likely to be. They should pay particular attention to the documents’ form, organization, specialized vocabulary, dissemination, and structure. In the words of Bazerman and Prior (2003: 3), the ‘focus [is] on what texts do and how texts mean rather than what they mean.’ Searching for traces of internalized ways of doing things and their mechanisms of diffusion, the kinds of questions we might ask when looking at documents include: How do the documents convey authority and legitimacy, defining how both writers and readers should do things? How does the format of the documents create insider/outsider distinctions, reinforcing social boundaries? How do the documents signal their intended readership and delimitate fields of practice? What practices do the documents tacitly call for?

We use US diplomatic cables below to illustrate how our method might look in practice. The cables are taken from the Public Library of US Diplomacy dataset, which is maintained by WikiLeaks and has proven controversial since its emergence in the 2010–2011 Cablegate incident. While some scholars have refrained from using the dataset, we agree with the growing calls for political scientists and diplomatic historians alike to make greater use of WikiLeaks cables (Michael, 2015). The authenticity of the cables has been corroborated in a variety of ways since their release nearly a decade ago, and while ethical concerns about using the documents persist, we side with those who believe the benefits for enriching scholarly analyses outweigh the risks (cf. Ottosen, 2012). Methodologically speaking, while the dataset is vast, with over 250,000 US diplomatic cables written between 2003 and 2010, it is not comprehensive, accounting only for around 5% of all American cable traffic during the years in question, which were selected in a largely ‘as if random’ fashion by the dataset’s original compiler, Chelsea Manning (Gill and Spirling, 2015). Given that we derive important insights into the practices of American diplomats from analyzing a considerably smaller sub-sample of cables (N=408), the non-comprehensiveness of the WikiLeaks dataset is not an issue for the methodological tools we outline below. Contrary to other recent empirical articles which draw upon the Wikileaks cables (e.g., Hunt, 2019), we are primarily interested in using the dataset to spark a methodological discussion. Accordingly, we ask two main questions in this article: (1) How particular epistemic practices undergird the production of the cables; and (2) How the stylized form of the cables can shed light on their social functions.

Our sample consists of 408 cables issued by American diplomats serving at the US Embassy in Dar es Salaam, Tanzania between January 1, 2005 and December 31, 2009. The sample contains an even mix of different levels of classification: 39% of the sampled cables were labelled as ‘Unclassified’; 25% received the slightly stronger designation ‘For Official Use Only’; a further 32% were classified as ‘Confidential’; and 4% were designated as ‘Secret’ (none received the highest classification of ‘Top Secret’). We selected these cables for practical reasons. First, one of the authors lived in Dar es Salaam from January to July 2008, and again from January to June 2010.² This participant observation gave us crucial knowledge of both diplomatic processes and the social and political situation in Tanzania, informing our understanding of the practices associated with diplomats’ knowledge production that the cables help uncover. Moreover, the available sample from the US mission in Tanzania is large enough and covers a sufficiently long period of time that we can establish clear patterns and standard operating procedures, while holding mostly constant the cables’ authors and intended audience (other US diplomats).

Our analysis is not generalizable as such. Our methodological approach starts with contextual analysis, and ethnographic insights used below illustrate how previous observations of diplomatic practices and their context inform our empirical account. We cannot extrapolate from our findings patterns of behavior that we expect to see replicated in other contexts. In Negotiating Across Cultures, Cohen (1991) challenges the view of a single international diplomatic culture and highlights the importance and need to focus on national diplomatic cultures. Similarly, we do not claim that we uncovered practices common to all embassies around the world. Different practices from diplomats working in other contexts would not invalidate our analysis. Quite the contrary, we contend that practices are adapted to the contexts of practitioners, and we expect different contexts to be associated with different practices.

This being said, interpretive practice tracing starts but does not stop with contextual knowledge as ‘social causality is to be established locally, but with an eye to producing analytically general insights’ (Pouliot, 2014: 237). As the conclusion discusses, our analysis says something about other (US) embassies around the world, as well as (American) diplomatic culture. This is particularly likely to be true as there is a concerted effort in the State Department to create a unitary bureaucratic culture, reinforced by specific practices like the forced rotation of diplomats every three years.

In using these cables, our goal is not to use them to draw specific conclusions about either American or Tanzanian policy or politics. Rather, we concentrate on the broader stories the cables can tell us in the aggregate about the conditions in which US diplomats in Tanzania work, socialize, and write. In doing so, we build upon growing scholarly interest in the practices diplomats use to gather and construct knowledge (Bicchi, 2014; Cornut, 2015a; Mérand et al., 2011; Neumann, 2012). We take cables as the traces—both causes and consequences—of the forging of a common diplomatic culture and an entry point into the epistemic practices of American diplomats posted in Tanzania during the period under analysis.

4 Looking through diplomatic cables: the epistemic practices of American diplomats

Scholars using documents to uncover practices should first look backwards to the conditions under which those artifacts were created. In this section, we are particularly interested in looking through our sample of Wikileaks cables to understand what types of knowledge American diplomats posted in Tanzania privilege, and which kinds of knowledge they normally downplay. This in turn gives us access to the practices associated with their knowledge production.

During our participant observation in Dar es Salaam, we identified six broad classes of knowledge that diplomats in Tanzania can draw upon when reporting back to headquarters about the country: (1) Tanzanian elite knowledge (officials working at the Tanzanian Ministry of Foreign Affairs (MOFA), Tanzanian officials and bureaucrats in other ministries, business leaders, politicians, army officers); (2) Non-Tanzanian elite knowledge (non-US diplomats based in Dar es Salaam, the local representatives of large international organizations, such as the World Bank or United Nations); (3) Tanzanian non-elite knowledge (ordinary Tanzanians, the representatives of Tanzanian civil society organizations); (4) Media reports (Tanzanian and non-Tanzanian); (5) Other kinds of knowledge, such as historical or sociological knowledge;³ (6) Direct knowledge derived from personal observation of an event or object.⁴

In order to empirically ascertain what types of knowledge American diplomats relied upon when drafting the cables, a team of coders manually coded each of the cables in the sample to see if it contained at least one occurrence of each of our possible sources of knowledge. Table 1 presents the results of our analysis, expressed as the percentage of cables in the overall sample which contained at least one occurrence of that type of knowledge.

Table 1.

Sources of knowledge in a sample of American diplomatic cables.

Source of knowledge		% of sampled cables drawingupon source of knowledge
Tanzanian elites		76%
	Tanzanian Ministry of Foreign Affairs official(s)	35%
	Other Tanzanian state official(s)⁵	40%
	Other Tanzanian elite(s)(especially political, business, or academic)	15%
Foreign elites ⁶		22%
	Inter-governmental organizations, non-governmental organizations based outside of Tanzania, and multilateral donors	12%
	Foreign diplomat(s) and staff of bilateral donor agencies⁷	13%
Tanzanian non-elites		10%
	Tanzanian CSOs	6%
	Other Tanzanian non-elites	5%
Media reports		10%
	Tanzanian media report	9%
	Foreign media report	1%
Other types of knowledge
	Historical (knowledge that references events prior to 1990)	13%
	Sociological (knowledge of slow-moving social and demographic shifts)	10%
	A field visit by US Embassy personnel(outside of Dar es Salaam)	6%

(N = 408 cables)

The data reveal that the single category of knowledge most cited by US diplomats in Tanzania is Tanzanian elite knowledge. American diplomats referred to knowledge gleaned from Tanzanian state officials in 69% of the cables sampled. The most frequently consulted ministry was the Tanzanian Ministry of Foreign Affairs (referenced in 35% of the cables sampled), with frequent references to officials housed in the Central Bank, the Office of the President of Tanzania, the Ministry of Health, and the Ministry of Planning as well. Elite Tanzanians without direct ties to the state were mentioned as a source of knowledge in 15% of the cables. The second largest set of interlocutors for American diplomats in Tanzania was foreign elites (referred to in 22% of the cables sampled), affiliated with another embassy (in 13% of the cables) or with an international organization (in 12% of cables).

Overall, these findings are perhaps not surprising. American diplomats serving abroad are typically elites—both back in the States, where they are typically well educated and have a high-status job, but even more so when serving abroad in the developing world, where their purchasing power allows them to frequently hire staff and maintain a very comfortable lifestyle. Accordingly, it is unsurprising that elites turn first and foremost to other elites when seeking knowledge.

Non-elite Tanzanians were cited far less as a source of knowledge than either Tanzanian or foreign elites, appearing in only 10% of the cables. When non-elite Tanzanians were consulted, it was often as representatives of civil society organizations (6% of cables sampled). Despite much rhetoric in Washington about the importance of civil society at the time, the numbers suggest that US diplomats do not consider ordinary Tanzanians as particularly important sources of knowledge about their country.

Media reports were not often directly cited as a source of knowledge in the cables, appearing in only 10% of the cables sampled; foreign media reports appear in substantially fewer cables (1%) than local media reports (9%). These figures are low, and we hypothesize that American diplomats may consider what newspapers publish to be public knowledge that does not necessarily require direct attribution. Being perceived as overly reliant on media reports is also unlikely to win American diplomats any credit back in Washington, since they are posted abroad specifically in order to cultivate unique channels of information (Ross, 2007).

Historical knowledge is somewhat absent from the WikiLeaks cables, with only 13% of the cables containing any references to events that predate 1990. This contrasts with how academics tend to approach knowledge-production. Most Africanists, for instance, would arguably consider that it is nearly impossible to understand present-day Tanzania without having some knowledge of the formative presidency of Julius Nyerere and his ujamaa philosophy (see Ibhawoh and Dibua, 2003). In a similar vein, while it is impossible to find a major academic book about Tanzania that does not invoke the concept and legacies of colonialism at some point, only 10 cables in the entire sample explicitly refer to colonialism or Tanzania’s colonial past.

Sociological knowledge also seems little employed by American cable writers, with 10% of the cables referring to large-scale demographic, economic, or cultural shifts. A close reading of the cables reveals that most US diplomats serving abroad implicitly draw upon a ‘Big Man’ understanding of politics, where political activity and social change are primarily the result of contestation between individual leaders instead of sociological transformations. This assumption is coherent with American diplomats relying upon elites so much. Yet socio-political processes are also often the result of large-scale macro trends that are not reducible to the actions of individuals, which seems to imply that the State Department may be missing the forest by focusing on just a handful of trees.

Finally, 6% of the sampled cables involved knowledge obtained firsthand by a US diplomat through direct observation. Typically these involved a junior diplomat undertaking a field visit, often related to election monitoring, and often on the troubled island of Zanzibar, something that a more senior diplomat could not do as easily for practical, security, and status reasons. This shows how hierarchy within the embassy influences what diplomats can and should do, which in turn has consequences on the knowledge they produce.

In sum, the cables show that US diplomats in Tanzania relied heavily on a single type of knowledge—Tanzanian elite insider knowledge—while marginalizing several other forms of available knowledge. US diplomats are biasedly and selectively learning about Tanzanian society and politics, often limiting their ability to comprehensively understand the country in which they are based. Compounding this epistemological isolation, it is worth remembering that the typical American diplomat is socially, culturally, racially, linguistically, and infrastructurally separated from the country they are tasked with understanding. Being physically confined in embassies-cum-fortresses make diplomats’ reporting task more difficult; as does the State Department’s requirement that diplomats rotate posts every two to three years, which sacrifices depth of insight and historical understanding in pursuit of supposed objectivity and impartiality (Ross, 2007).

5 Looking at diplomatic cables: socialization via writing

Looking at documents focuses on the effects that routine exposure to the documents is likely to produce. In this section we focus on how the diplomatic cables sent by American diplomats produced powerful socialization effects in both their readers but also their writers, as a single institutional voice drowned out individual opinions and attitudes.

To uncover socialization processes looking at cables, one place to begin is with the highly standardized format of the latter. Without exception, all contemporary American cables take the same shape: first, a header (which includes the cable’s title, reference number, intended distribution list, classification level, and subject tags); second, a one-paragraph summary of the cable; third, the body of the text, broken up into paragraphs containing between two and nine sentences (although typically the length is about four sentences); occasionally a COMMENT section where the cable’s author can provide more speculative annotation of the preceding material (expressing skepticism, for instance, about the official line of the Tanzanian government); and finally, a typed signature, typically of the ambassador or the most senior diplomat at the post, which serves as a sort of Embassy imprimatur. This systematic formatting makes it easier to review and find information in a cable quickly, but also limits cable authors to writing in a linear, chronological, point-by-point manner, as opposed to freer, looser, or more narrative styles.⁸

In addition to structure, other aspects of the cables are both evidence of and likely contributors to intense socialization. There is the shared usage of technical language—a vocabulary of démarches and chargés d’affaires that outsiders would likely be unfamiliar with—and above all a proliferation of unexplained acronyms that make certain cables difficult to decipher. When one of us worked in a foreign embassy in Dar es Salaam, on his first day he was taken aside by a sympathetic peer and handed a manually typed-up list of acronyms. It was clear that this list was passed around the Embassy as a crib sheet for newcomers seeking to make sense of the sea of acronyms used daily in the Embassy, which included Tanzanian ministerial designations, bilateral development initiatives, multi-donor coordination committees, Tanzanian political parties, and others.

To a certain extent, developing and utilizing a specialized vocabulary is a characteristic of all professions, since it facilitates an exclusionary process which confers higher status to insiders (Beaufort, 2009; MacDonald, 1995). But the effect is even more pronounced in US diplomatic culture, given the socially-privileged position of most American diplomats and the emphasis on secrecy rife within the State Department (Gill and Spirling, forthcoming). Thus, the seemingly indecipherable content of the majority of the cables (on top of the actual cryptographic algorithms that are used to eventually transmit the data over telecommunications lines) helps construct a sense of separation and secrecy between the Embassy staff and the outside world (cf., Gusterson, 1998: Chapter 4).

Another interesting syntactic element in the cables has to do with the use of personal pronouns. In only 1% of the sampled cables do American diplomats directly refer to themselves by using first-person singular pronouns and adjectives (‘I’ or ‘me’ or ‘my’), excluding examples that occurred in direct reported speech. Instead, a document’s real authorship is masked by employing more indirect, collective subjects. For instance, an indefinite ‘we’ that refers to the views of the State Department or the US government (or even the United States in toto) appears in 34% of the cables. Alternatively, a number of circumlocutions are also employed by the cable writers to refer to themselves, including ‘Mission,’ ‘Post,’ ‘POLOFF’ [political officer], and ‘USG’ [United States government].

And yet, as linguists and psychologists are increasingly realizing, the use of pronouns and other identifying markers is not an innocent process (Pennebaker, 2013). There is power at play in decisions to use—or not use—‘I.’ In the cables, the deliberate effort to avoid the first-person singular represents an effort to cultivate a writing style that not only removes any sense of personal authorship, but also fosters the sense that the Embassy is a unitary collective and has a sole view about the events in question. This erases traces of the bureaucratic infighting around documents (Barnett, 1997; Neumann, 2012), even if our participant-observation made it clear that there were often frequent tussles about individual documents or even phrases. When one of us would have to write aide-mémoires following inter-embassy coordination meetings of Western donor countries, the initial draft was always sent to the Ambassador for careful review and rewriting before being disseminated to meeting participants.

The US diplomatic cables reinforce this preference for collective pronouns and circumlocutions by typically utilizing the passive voice instead of the active voice. As scholars have noted, in English the passive voice makes it difficult to locate agency, suggesting that outcomes arise organically instead of as a result of deliberate actions. Its common occurrence in US government documents seems deliberate at times (Osborn, 2010: Chapter 8), and its use in the WikiLeaks cables suggests that toeing the party line is considered more important than permitting more idiosyncratic writing styles (and opinions).

A final aspect of the cables’ form that can seem puzzling at first to a neophyte is the distribution list. Even the most mundane and routine of cables often have surprisingly long distribution lists—the mean number of designated receiving offices for the sample was 6.87. For scholars, distribution lists provide information about membership, functions, hierarchies, and responsibilities in the field under study. Diplomats’ competence is to know who should receive specific cables and who should not, and a suitable distribution list is evidence of this. This not only requires knowledge of who is doing what within specific diplomatic networks; it also sets and reinforces boundaries by stating who should and should not access knowledge generated by peers, as well as who should be interested by this new knowledge. Including some and excluding others, distribution lists are traces as well as instruments of selectivity and exclusivity in diplomatic circles (and bureaucratic environments more generally). The networks through which cables circulate draw boundaries; in this sense, distribution lists of these documents reify social networks.

When making these types of inferences, however, scholars should keep in mind that a ‘certain circle of recipients, designated by codes in the cable’s header, will have the cable in their digital inbox immediately,’ but that ‘the message is also translated into a record in a digital database. Everyone at the State Department with the appropriate access rights can search this database and retrieve cables’ (Wille, 2016: 174). Recipients named in the distribution list may also choose to circulate it to other members of the field. In other words, the distribution list, even if long, only reveals the direct intended readership and is a trace of parts of the socialization processes associated with knowledge circulation, while the actual audience could be smaller (when recipients do not read the cables) or larger (when for instance a recipient chose to transfer the cables).

6 Conclusion

The findings presented above have general validity for those interested in understanding international practices and who might turn to written documents to do so. The stories we have focused on here unravel the standardizing effects and socialization processes that are both the result of the cables and their cause. We confirmed that US diplomatic practices tend to be similar to the practices that Cornut (2015a) observed in European and Canadian embassies.

Our study is interesting not just for what we found, but also for how we found it. Our systematic analysis of diplomatic cables—their form as well as their content—was informed by 13 months of participant observation in embassies at Dar es Salaam. Compared with Cornut (2015a), who only used interviews to reconstruct the practices associated with diplomatic knowledge production in Cairo and did not look at the cables themselves, we were able to reconstruct (US diplomatic) practices, culture, and habits in novel ways. This illustrates how methodical document analysis associated with fieldwork suggests a distinctive approach to practice, one that combines experience-distant and experience-near knowledge about social and international life. Such an approach can be applied to the form and content of other documents—internal documents of international organizations, governments, and NGOs, resolutions, reports, treaties, transcript of discourses, minutes of meetings—to unravel other international practices.

There are of course limitations to a study of the epistemic and writing practices of diplomats that notably relies on a few hundred cables from only one post over a period of only five years. Some of our findings are specific to documents written in English—the usage of pronouns will likely differ in other languages, with different socio-political effects. Others of our findings are influenced by the particular features of the Tanzanian environment during the period in question. For instance, had US relations with Tanzania been less friendly during the period in question, it is probable that American diplomats would have been less able to rely on insider, Tanzanian elite knowledge when writing their reports—at least not governmental ones. Similarly, the fact that relatively few cables referenced Tanzanian media accounts may be linked to the fact that many Tanzanian media outlets were unreliable during the period in question, with clear political axes to grind, such that they occasionally printed blatant falsehoods. Consider, however, that we were able to observe that Tanzania did have a vibrant civil society during the late-2000s, and yet it was still very rarely a source of knowledge for American diplomats in Tanzania.

The semi-public, formal nature of the cables analyzed here sheds important light on how the State Department operates, but only partially so. The cables are simply one layer of State Department writings out of many, beginning with thoughts quickly jotted down on notepads at one end of the spectrum and extending upwards to blue-ribbon policy memos and bird’s-eye-view annual reports. There is no reason to believe that any single one of these layers contains more epistemic truth than others; instead, the researcher’s task should be to: identify the function that the layer serves in the context of the overall organization; locate how socialization processes operate within the layer studied and assess their effects in practice; and eventually make comparisons across the layers when the available material permits doing so.

Despite these contextual limitations, we believe that the practices uncovered here provide general analytical insights about (US) diplomatic culture. They are crucially important to understand the character and impact of US foreign policy in particular. Consider two examples: how the writing practices of US diplomats makes it more difficult to express dissent, and how the epistemic practices of US diplomats have caused them to fail to anticipate grassroots protests.

Some commentators have suggested that having a common diplomatic culture can facilitate internal communication and help generate a positive esprit de corps in a polity’s diplomatic service (Cross, 2011). While this is likely true, there is also a dark side to a bureaucratic culture that seeks to excise all markers of individual authorship, namely that the socialization processes at play in the State Department can lead to insularism and groupthink. According to one recent history of the US State Department (Gurman, 2012), this has been a recurring problem for the organization throughout the 20th century, with individual diplomats unable to express their heterodox views on, for instance, the US’s China policy in the 1950s, or the US’s military escalation in Vietnam in the 1960s, leading to dramatic foreign policy failures. Even since the institution of an Official Dissent channel within the State Department in 1971, the pressure for diplomats to toe the official policy line has led to the State Department backing ill-advised foreign policies like the 2003 Iraq War. Is there not after all always a danger when ‘institutional voice is achieved by drowning out the voices of specific diplomats’ (Neumann, 2012: 87)?

Separately, we have shown that American diplomats tend to discount non-elite knowledge in favor of interacting with local elites. As Bull put it some time ago, diplomatic knowledge ‘derives from day-to-day personal dealings with the leading political strata in the country to which a diploma[t] is accredited, sometimes to the detriment of his understanding of society at large in that country’ (Bull, 1977: 175). These epistemic practices may help explain why the State Department has a history of failing to anticipate significant grassroots mobilizations and popular protests, such as the fall of the Berlin Wall, the Green Revolution in Iran, or the Arab Spring. If American diplomats began engaging broader cross-sections of society, could this tendency be alleviated? Looking at and through documents should help IR scholars identify similar blind spots, helping diplomats everywhere better practice their craft.

Footnotes

Acknowledgements

The authors are grateful to Aakanksha Lohia, Achyut Mishra, Arudra Ravindran, Nagul N, Rajesh Ghosh, and Corin Tentchoff for excellent research assistance on this project. We also appreciate the helpful comments received on earlier drafts of this article from Patrick James, Deniz Kuru, Geoffrey Wiseman, J. Ann Tickner, Wayne Sandholtz, Andrew Lakoff, Iver Neumann, Joey Huddleston, Giacomo Chiozza, Theo Mazumdar, Kristen Guth, Krishanu Karmakar, and Darya Berezhnova.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Jérémie Cornut

Nicolas de Zamaróczy

Notes

Author biographies

Jérémie Cornut is associate professor in the Department of Political Science at Simon Fraser University, Vancouver, Canada. He holds a PhD in political science. His teaching and research focus on diplomacy, practice theory, international relations theory, and Canadian foreign policy. He is currently working on digital diplomacy in practice.

Nicolas de Zamaróczy is a scholar of International Relations with interests in diplomacy, regional integration, feminist IR, text as data, and digital games. He received his Ph.D. from the University of Southern California.

References

Adler

Pouliot

(2011) International practices. International Theory 3(1): 1–36.

Adler-Nissen

(2014) Opting Out of the European Union: Diplomacy, Sovereignty and European Integration. Cambridge: Cambridge University Press.

Adler-Nissen

Drieschova

(2019) Track-change diplomacy: Technology, affordances and the practice of international negotiations. International Studies Quarterly 63: 531–545.

Adler-Nissen

Pouliot

(2014) Power in practice: Negotiating the international intervention in Libya. European Journal of International Relations 20(4): 889–911.

Andersen

Neumann

(2012) Practices as models: A methodology with an illustration concerning Wampum diplomacy. Millennium: Journal of International Studies 40(3): 457–481.

Barnett

(1997) The UN Security Council, indifference, and genocide in Rwanda. Cultural Anthropology 12(4): 551–578.

Bayram

(2020) Measuring common knowledge: Latent semantic analysis, linguistic synchrony, and common knowledge in international relations. International Relations 34(2): 180–203.

Bazerman

Prior

(2003) Introduction. In: Bazerman

Prior

(eds) What Writing Does and How It Does It. An Introduction to Analyzing Texts and Textual Practices. New York: Routledge, pp. 1–10.

Beaufort

(2009) Writing in the professions. In: Bazerman

(ed.) Handbook of Research on Writing. History, Society, School, Individual, Text. New York: Lawrence Erlbaum Associates, pp. 269–288.

10.

Bicchi

(2014) Information exchanges, diplomatic networks and the construction of european knowledge in European Union Foreign Policy. Cooperation and Conflict 49(2): 239–259.

11.

Bonelli

Ragazzi

(2014) Low-tech security: Files, notes, and memos as technologies of anticipation. Security Dialogue 45(5): 476–493.

12.

Bueger

(2014) Pathways to practice. Praxiography and international politics. European Political Science Review 6(3): 383–406.

13.

Bueger

Gadinger

(2018) International Practice Theory. Basingstoke: Palgrave Macmillan.

14.

Bull

(1977) The Anarchical Society: A Study of Order in World Politics. Basingstoke: Palgrave Macmillan.

15.

Checkel

(1997) International norms and domestic politics: Bridging the rationalist-constructivist divide. European Journal of International Relations 3(4): 473–495.

16.

Checkel

(ed.) (2007) International Institutions and Socialization in Europe. Cambridge: Cambridge University Press.

17.

Cohen

(1991) Negotiating across Culture: International Communication in an Interdependent World. Washington. DC: United States Institute of Peace.

18.

Connelly

Hicks

Jervis

, et al. (2020) Diplomatic documents data for international relations: The freedom of information archive database. Conflict Management and Peace Science. Epub ahead of print 30 June. DOI: 10.1177/0738894220930326.

19.

Constantinou

(2013) Between statecraft and humanism: Diplomacy and its forms of knowledge. International Studies Review 15(2): 141–162.

20.

Cooper

Cornut

(2019) The changing practices of frontline diplomacy: New directions for inquiry. Review of International Studies 45(2): 300–319.

21.

Cornut

(2015a) To be a diplomat abroad: Diplomatic practice at embassies. Cooperation and Conflict 50(3): 385–401.

22.

Cornut

(2015b) The practice turn in International Relations Theory. In: Denemark

Marlin-Bennett

(eds) The International Studies Encyclopedia. New Jersey: Blackwell Publishing, Oxford Reference Online.

23.

Cornut

(2018) Diplomacy, agency and the logic of improvisation and virtuosity in practice. European Journal of International Relations 24(3): 712–736.

24.

Cross

(2007) The European Diplomatic Corps: Diplomats and International Cooperation from Westphalia to Maastricht. Basingstoke: Palgrave Macmillan.

25.

Cross

(2011) Security Integration in Europe: How Knowledge-Based Networks Are Transforming the European Union. Ann Arbor, MI: University of Michigan Press.

26.

Dittmer

McConnell

(eds) (2015) Diplomatic Cultures and International Politics. Translations, Spaces and Alternatives. New York: Routledge.

27.

Ferguson

(1994) The Anti-Politics Machine: ‘Development,’ Depoliticization, and Bureaucratic Power in Lesotho. Minneapolis, MN: University of Minnesota Press.

28.

Flockhart

(2006) ‘Complex socialization’: A framework for the study of state socialization. European Journal of International Relations 12(1): 89–118.

29.

Freeman

Maybin

(2011) Documents, practices and policy. Evidence and Policy: A Journal of Research, Debate and Practice 7(2): 155–170.

30.

Gill

Spirling

(2015) Estimating the severity of the WikiLeaks U.S. diplomatic cables discolsure. Political Analysis 23(2): 299–305.

31.

Gill

Spirling

(forthcoming) Dimensions of diplomacy: Understanding private information in international relations using the WikiLeaks cable disclosure.

32.

Grimmer

Stewart

(2013) Text as data: The promise and pitfalls of automatic content analysis methods for political texts. Political Analysis 21(3): 267–297.

33.

Gurman

(2012) The Dissent Papers: The Voice of Diplomats in the Cold War and beyond. New York: Columbia University Press.

34.

Gusterson

(1998) Nuclear Rites: A Weapon’s Laboratory at the End of the Cold War. Berkeley, CA: University of California Press.

35.

Harper

(1997) Inside the IMF: An Ethnography of Documents, Technology and Organisational Action. New York: Routledge.

36.

Hopf

(2017) There are no panaceas in social theory. International Studies Quarterly Post, 18 June.

37.

Hull

(2012a) Government of Paper: The Materiality of Bureaucracy in Urban Pakistan. Berkeley, CA: University of California Press.

38.

Hull

(2012b) Documents and Bureaucracy. Annual Review of Anthropology 41: 251–267.

39.

Hunt

(2019) The WikiLeaks cables: How the United States exploits the world, in detail, from an internal perspective, 2001-2010. Diplomacy and Statecraft 30(1): 70–98.

40.

Ibhawoh

Dibua

(2003) Deconstructing Ujamaa: The legacy of Julius Nyerere in the quest for social and economic development in Africa. African Journal of Political Science 8(1): 59–83.

41.

Johnston

(2001) Treating international institutions as social environments. International Studies Quarterly 45(4): 487–515.

42.

Johnston

(2008) Social States: China in International Institutions, 1980-2000. Princeton, NJ: Princeton University Press.

43.

Kafka

(2009) Paperwork: The state of the discipline. Book History 12: 340–353.

44.

Katagiri

Min

(2019) The credibility of public and private signals: A document-based approach. American Political Science Review 113(1): 156–172.

45.

King

Keohane

Verba

(1994) Designing Social Inquiry: Scientific Inference in Qualitative Research. Princeton, NJ: Princeton University Press.

46.

Kustermans

(2016) Parsing the practice turn: Practice, practical knowledge, practices. Millennium 44(2): 175–196.

47.

Kuus

(2013) Foreign policy and ethnography: A sceptical intervention. Geopolitics 18(1): 115–131.

48.

Lie

(2013) Challenging anthropology: Anthropological reflections on the ethnographic turn in international relations. Millennium 41(2): 201–220.

49.

MacDonald

(1995) The Sociology of the Professions. London: SAGE.

50.

Mérand

Hofmann

Irondelle

(2011) Governance and state power: A network analysis of European security. Journal of Common Market Studies 49(1): 121–147.

51.

Michael

(2015) Who’s afraid of WikiLeaks? Missed opportunities in political science research. Review of Policy Research 32(2): 175–199.

52.

Navari

(2011) The concept of practice in the English school. European Journal of International Relations 17(4): 611–630.

53.

Neumann

(2002) Returning practice to the linguistic turn: The case of diplomacy. Millennium: Journal of International Studies 31(3): 627–651.

54.

Neumann

(2012) At Home with the Diplomats: Ethnography of a European Foreign Ministry. Ithaca, NY: Cornell University Press.

55.

Osborn

(2010) Anarchy and Apocalypse: Essays on Faith, Violence, and Theodicy. Eugene, OR: Cascade Books.

56.

Ottosen

(2012) WikiLeaks: Ethical minefield or a democratic revolution in journalism? Journalism Studies 13(5–6): 836–846.

57.

Pennebaker

(2013) The Secret Life of Pronouns: What Our Words Say about Us. New York: Bloomsbury Press.

58.

Pouliot

(2007) ‘Sobjectivism’: Toward a constructivist methodology. International Studies Quarterly 51(2): 359–384.

59.

Pouliot

(2008) The logic of practicality: A theory of practice of security communities. International Organization 62(2): 257–288.

60.

Pouliot

(2010) International Security in Practice: The Politics of Nato-Russia Diplomacy. Cambridge: Cambridge University Press.

61.

Pouliot

(2012) Methodology: Putting practice theory into practice. In: Adler-Nissen

(ed.) Bourdieu in International Relations: Rethinking the Key Concepts in IR. London: Routledge, pp. 46–58.

62.

Pouliot

(2014) Practice tracing. In: Bennett

Checkel

(eds) Process Tracing: From Analytic Metaphor to Best Practices. Cambridge: Cambridge University Press, pp. 237–259.

63.

Pouliot

(2016) International Pecking Orders: The Politics and Practice of Multilateral Diplomacy. Cambridge and New York: Cambridge University Press.

64.

Pouliot

Cornut

(2015) Practice theory and the study of diplomacy: A research agenda. Cooperation and Conflict 50(3): 297–315.

65.

Riles

(1998) Infinity within the brackets. American Ethnologist 25(3): 378–398.

66.

Ross

(2007) Independent Diplomat: Dispatches from an Unaccountable Elite. Ithaca, NY: Cornell University Press.

67.

Schindler

Wille

(2019) How can we criticize international practices? International Studies Quarterly 63(4): 1014–1024.

68.

Sending

Pouliot

Neumanneds

(eds) (2015) Diplomacy and the Making of World Politics. New York and Cambridge: Cambridge University Press.

69.

Sharp

(2004) The idea of diplomatic culture and its sources source. In: Slavik

(ed.) Intercultural Communication and Diplomacy. Malta: DiploFoundation, pp. 361–379.

70.

Walter

(2019) The road (not) taken? How the indexicality of practice could make or break the ‘New Constructivism. European Journal of International Relations 25(2): 538–561.

71.

Wille

(2016) Diplomatic cable. In: Salter

(ed.) Making Things International 2: Catalysts and Reactions. Minneapolis, MN: University of Minnesota Press, pp. 166–178.

72.

Wiseman

(2011) Distinctive characteristics of American diplomacy. The Hague Journal of Diplomacy 6(3–4): 235–259.

73.

Wiseman

Sharp

(eds) (2007) The Diplomatic Corps as an Institution of International Society. New York: Palgrave Macmillan.