Selling Intervention

How the American Press Influenced the Outbreak of the Spanish-American War

Term Paper (Advanced seminar), 2010

29 Pages, Grade: 1.5



1. Introduction

2. Media and Politics: Theoretical Linkages
2.1 General Propositions
2.2 Construction of Reality
2.3 Information and Bias
2.4 Frames and Attributes
2.5 Narratives and Symbols
2.6 Public Opinion
2.7 Problem Recognition and Agenda-Setting
2.8 Political Entrepreneurs
2.9 Decision and Behavior
2.10 Synthesis and Specification

3. Research Design
3.1 Case-Study Design and Case Selection
3.2 Research Question, Variables and Hypotheses
3.3 Operationalization and Method

4. “The Newspaper War?” – An Empirical Analysis
4.1 William Randolph Hearst and the New York Journal
4.2 The Humanitarian Crisis on Cuba: Major Events and Press Coverage
4.2.1 The Weyler Case
4.2.2 The Cisneros Case
4.2.3 The de Lôme Letter
4.2.4 The Maine Incident
4.2.5 Senator Proctor
4.2.6 President McKinley and Congress
4.2.7 The War Outbreak

5. Conclusions

6. References


Humanitarian interventions are multi-dimensional affairs. One of these many dimen­sions is the relationship of media and politics – perhaps so much, that such interven­tions may be called ‘media-driven’. The following scrutiny focuses on this matter in detail with the aim of elucidating the numerous aspects of this relationship. I have se­lected a historical case – the humanitarian intervention of the USA in Cuba in 1898 (better known as The Spanish-American War or The Splendid Little War) – which has sometimes been called ‘The Newspaper War’, because of the large influence of a sen­sationalist press (yellow journalism) and its exaggerated accounts of the crisis on Cu­ba under Spanish colonial rule.

Indeed, the role of the American press in this war has been widely discussed.1 Be­cause of this, it is already quite clear that the media was a necessary, but not sufficient condition for the intervention to occur. Thus, I find it redundant to assess deterministic antecedent conditions or measure mere causal effects any further. However, it is still of interest how exactly the relationships between media, public opinion and politics developed to the point where intervention became inevitable and what modern theo­ries there are to explain these historical facts.2 So, with this example, I hope to provide useful insights into the mechanism of how media may impact political decisions to in­tervene – a real world problem still present today. My research question therefore cen­ters on the nature of the links between media, public opinion and politics: How did the American press and public opinion influence the political decision-making process that led to the American intervention in Cuba in 1898? In other words, what was the causal mechanism and process at work?

In order to answer this question, first of all, I develop a theoretical framework, which incorporates various dimensions and elements of the relationship between media and politics. Secondly, the methodical framework is laid out and (explanatory) hypotheses are formulated. Thirdly, my empirical scrutiny focuses on the most prominent empirical example, William Randolph Hearst’s New York Journal, which provoked public pres­sure like no other newspaper. Finally, I finish with a thorough conclusion.

Even though my endeavor may be a substantial contribution to the controversial dis­pute about the topic, it is clearly constrained by limited access to primary sources of that time and the fact that only one aspect of the causation of this case is highlighted.


There are numerous theories to model the relationship between media and politics.3 In the following exploration, the most relevant concepts for my investigation are outlined and developed into a coherent framework. Firstly, general implications are considered. Secondly, linked aspects of the social construction theory are concisely introduced. Thirdly, I reflect shortly on information and bias and then the central concepts of news framing and narratives are established. Next, relations with public opinion are expli­cated. Most prominently, I scrutinize aspects of problem recognition and agenda-setting. Concepts of attention, behavior and decision are added. Finally, a succinct specification and synthesis end this discussion.

Before I begin, though, here are some preliminary thoughts: As Blatter and Junk (2008) have suggested, modern humanitarian interventions can be perceived as multi­level processes. In the selected historical case, this can be narrowed down to a simp­ler two-level game with the media and public opinion on the one side, and the political elites on the other.4 The basic approach is information- and media-centered; however, the so-called ‘CNN-Effect’ is not considered, due to its obvious inapplicability to the chosen case.


First of all, there are classic models of institutional settings for the media. In the USA, this is the libertarian tradition – a commercial media system, in which the press is free of governmental control (see Johnson-Cartee 2005: 82 and Oates 2008: 5f.). Enter­taining news rather than serious coverage is quite common. Opinions may be ex­pressed freely and confrontation with the political elites is allowed and sometimes even deliberately sought.

Journalists can be viewed as political actors and newspapers as political institutions, when defining the term broadly and untraditionally (see Cook 2005). However, con­trary to what journalists are taught and what has become known as the ‘fourth estate premise’, the media certainly is not the fourth branch of government (see Johnson-Cartee 2005: 77f. and Cook 2005).

I conceptualize a bottom-up perspective, in which media may ‘upload’ political ideas onto the governmental agenda (see 2.7).5 The focus is on macro-effects of the media on society and, ultimately, on the political sphere. Johnson-Cartee (2005: 13f.) defines certain conditions, under which the media have “real and powerful effects”. These are: (a) ubiquity (the information is widely spread); (b) consonance (similar pictures of the world and unanimous argumentation) and (c) cumulation (repetition of similar messag­es), all of which combined constitute a high degree of uniformity. The result is the at­traction and direction of public attention, eventually structuring definitions of reality. In that way, the media outputs can also serve as instrumental prioritization and problem-detection devices for politicians (see Jones & Baumgartner 2007: 281).


Johnson-Cartee (2005: 4) posits the existence of mediated and constructed reality by contact with information: “[...] political knowledge is constructed through the mass me­dia.” Furthermore, she states that “even lands far away [...] suddenly become real as we attend to the mass media. And we behave toward what we have learned as if it were real (real in the sense of personal validation of existence).” The conclusion is that “political reality is formed by mass communication reports which are talked about, al­tered, and interpreted by citizens in a society. The totality of this process constitutes reality” (Kraus and Davis 1976: 211, cit. in Johnson-Cartee 2005: 14).


We all depend on media sources for information. Jones and Baumgartner (2007: 274) argue that such information is “critical to all decision making and to politics”, even though it is “often misleading”. They add that information is often processed dispropor­tionately, for instance when “decision makers over-rely on a single source rather than numerous sources” (ibid.: 280). Moreover, there are constraints by public opinion or extremely visible events publicized in the media. Therefore, disproportionate informa­tion processing is a characteristic of political systems (see ibid.).

The news may reflect ideological biases. Street (2008: 16ff.) conceives bias as favor­ing a particular point of view or interpretation or sympathizing with only one cause and makes a clear distinction between opinion and facts. The verity that media select facts, perhaps even systematically, may already distort depiction of events. Propaganda bias would be the utmost extreme version of this, since, in that case, a report is made with the deliberate intention of making the case for a particular point of view without expli­citly stating this (see ibid.).


Entman (2000: 12) defines framing 6 as selecting, highlighting and associating ele­ments of reality to tell a coherent story. Or as McCombs (2004: 87) puts it: “To frame is to select some aspects of a perceived reality and make them more salient in a com­municating text, in such a way as to promote a particular problem definition, causal interpretation, moral evaluation and/or treatment recommendation for the item de­scribed” (emphasis in original). This means that journalists fundamentally affect how news readers understand the events, thus producing a framing effect. How may this happen? Johnson-Cartee (2005: 24) proposes that frames affect opinions simply by making certain considerations seem more important than others and thereby impacting the final attitude. Of course, the framing power depends on how much importance the particular actor using it has and, thus, how dominant the frame is.

Additionally, frames can be associated with exemplars, catchphrases and/or visual images (ibid. 169), producing so-called condensational symbols. Exemplars are “dra­matized accounts of real events, [...] which are then used to represent abstract forces, issues or entities [...]” (ibid.). Catchphrases (taglines or slogans; often headlines) are “attempted summary statements about the principal subject” (ibid. 165/170). Visual images are routine visual characterizations used in illustrating the issue’s prime frame. Such condensational symbols can evoke moral values and general societal and cul­tural self-images to generate enthusiasm for particular policy options (see ibid. 171).

Frames can also be seen as attributes describing an object in question (see McCombs 2004: 88). Nevertheless, not all attributes are frames: attributes range from a micro- to macro-continuum, and frames are at the macro end of this continuum as the dominant perspective.7

Johnson-Cartee (ibid. 29) makes a further distinction between low-threshold issues and high-threshold issues. The former signifies an issue about which people’s know­ledge is influenced by their own experiences (e.g. the price of gasoline), whereas the latter implies no firsthand experience in people’s lives (e.g. foreign policy). It is as­sumed that media have the greatest influence on high-threshold issues, because people have no independent means to verify the news accounts.


Johnson-Cartee (2005: 16) conceives politics as a symbolic world. Our political world is not a photocopy of the objective world, but rather a created world of symbols, often mass-mediated symbols. This may have real consequences for our opinion and beha­vior, partly because we often simply acknowledge such symbols.

News can also be seen as narratives that provide elaborate explanations, using sym­bols as raw materials, organizing ideas and embedding facts. Narrative here may be defined as an “account of connected events in order of happening” (ibid. 148). Story formats are efficient structures to reduce complexity to a minimum and for collapsing a long time frame into a short interesting summary (see ibid. 157). The logic of the con­crete (people, situations, plans, and motives) is clearly prominent, while background information is marginal (see ibid. 266). That way, political narratives can establish the antagonistic forces of evil and good. They assign responsibility as to the origin of a problem and to who may solve it. Johnson-Cartee (ibid. 149) even goes further declar­ing that “narration is the means by which societies ultimately govern themselves, for these shared stories establish commonalities, promote goodness, and discourage wickedness. Narratives provide societies with their moral reasoning” (ibid. 149). Final­ly, news accounts may be “mythical narratives” (ibid. 186) that do not necessarily re­flect an objective reality.8


Also, there is the concept of public opinion, which signifies the preferences of the ma­jority of individuals on a specific issue. The media selects and aggregates the expecta­tions of the public or depicts political events and conveys them to the public. Accor­dingly, the public may perceive a situation in a wrong way. This is called “pluralistic ignorance” (ibid. 36), which occurs because of erroneous environmental messages (false, inadequate and/or misleading information).

Generally, McCombs (2004: 85) argues that “attributes that are prominent in the mass media are prominent in the public mind”. Therefore, it is only logical that “policy-makers often take news reports as indicative of public opinion” (Johnson-Cartee 2005: 80). In other words, the media can invoke a certain perception of public opinion (puta­tive public opinion; cf. Entman 2000: 19ff.). Public opinion therefore poses a constraint on political processes and functions as a catalyst between the media and the political elites’ actions.9


An agenda can be defined as a list of issues that attract serious attention (cf. Kingdon 1984: 205). There are various types of agendas, namely the media agenda, the public agenda and the governmental agenda. There is a correspondence10 of the media agenda and the public agenda (see McCombs 2004: 134). However, as Jones and Baumgartner (2007: 211) state, the media and public agenda are “exogenous to the policymaking process”. Still, the public agenda, once set or reflected by the media agenda, influences the policy agenda of elite decision makers. Even more, the media agenda seems to have direct and sometimes strong influence on the governmental agenda (see Johnson-Cartee 2005: 17 and Jones & Baumgartner 2007: 232-273).

Agenda-setting is the “process by which information is prioritized for action, and atten­tion allocated to some problems rather than others” (Jones & Baumgartner 2007: viii/ix). For an agenda priority shift to occur, a problem must first be recognized. In oth­er words, “problem recognition is critical to agenda setting” (Kingdon 1984: 210). Moreover, “the recognition and definition of problems affect outcomes significantly” (ibid.), partly because problem definition may be coupled to a policy proposal. Blatter and Junk (2008: 4) model the concept of Windows of Recognition, which signifies “times when [a problem] receives so much attention in the domestic public discourses that it is appropriate for the government to give it a high priority on the governmental agenda”. They propose that such recognition is determined “by the ‘salience’ of an issue, the attention that the issue gets within the media discourse and, in conse­quence, the priority it gets on the political agenda of the government”.11 Or, in other words, “[...] in agenda models, information primarily causes heightened attention to an issue, [...]. That is, information is never neutral in the policy process, and that is why it is so fundamental” (Jones 1994: 23, cit. in Blatter & Junk 2008: 14).

Furthermore, McCombs (2004: 36) puts forward that “agenda-setting is a robust and widespread effect of mass communication, an effect that results from specific content in the mass media.” The mere salience of objects is considered first-level agenda-setting, whereas framing is second-level agenda-setting (attribute agenda-setting12). Thus, how issues are framed or connected to certain attributes has also a significant impact on the priority of objects on an agenda (cf. Blatter & Junk 2008: 16). To set an issue on an agenda successfully, the following framing features may foster the process: clarity, urgency, simplicity, and great symbolic meaning (cf. Blum & Schubert 2009: 112). McCombs (2004: 128) confirms that there is evidence about the effects of both substantive attributes and affective tone in media messages, which demonstrates that both first- and second-level agenda-setting have considerable consequences.


Political entrepreneurs can be seen as pushing for one kind of problem definition and often dramatizing its importance (see Kingdon 1984: 214f.). That way, they may over­come structural constraints in the process of a policy decision (see Blatter & Junk 2008: 15). By generating links and framing information flows, public pressure can be increased by such actors.


The Thomas Theorem posits that if men define situations as real, they will become real in their consequences. This is central in the outlined context. Johnson-Cartee (2005: 35) explains that “if people have unwittingly created ‘false’ images about the attitudes, beliefs, and behaviors of others, they will act upon those images as if they were real, and as a result, these images will ultimately have real consequences.” McCombs (2004: 120) conceives behavior as being “governed by cognition”13. As there are many cognitive (and affective) elements in attribute agenda-setting, there is a clear and direct connection of attribute salience, alternative specification, and beha­vioral decisions.

To sum up, here are the various stages of a decision: recognition of a problem, assessment of the various dimensions of the problem, finding potential solutions, and making a choice (see Jones & Baumgartner 2007: 275). Stressing my point conclusively, all of these stages are affected by mediated reality in a substantial way.


1 An early explanation centered on the yellow press as the primary catalyst. More sophisticated analysis subsequently discredited this argument. Alternative explanations have shown a combination of factors at work, among them economic and humanitarian motivations as well as domestic tensions and the political agenda of various national leaders. With rare exception, the role of the press has faded into the back­ground, even in studies by modern media historians. Useful as it is to question the myth that the yellow press caused the war on its own, shoving newspapers out of the picture altogether nevertheless creates a distortion. It is speculated that the press really was a factor in the decision to go to war – it did not start the war, but based on what scholars now know about media effects, it is argued that it shaped public opinion to create an enabling environment for declaring war.

2 Van Evera (1997: 75) states: „Political scientists seldom do case-explaining case studies, partly because they define [this task] as the domain of historians; however, historians often explain cases in a softer way than political scientists would. Their explanations are left vague, [...]. As a result their explanations are hard to interpret and evaluate. This leaves wide latitude for political scientists to contribute to discussion of historical explanation.” Also, McCombs (2004: 34) affirms that “the knowledge that we have gained in re­cent times about the agenda-setting role of the mass media [can be and] has been used [...] to organize our understanding of the historical past.”

3 I use general theories so as to incorporate as many aspects as possible. The specification follows in chapter 2.10 and in the empirical part. Blatter & Junk (2008: 5-11) provide a helpful state of the art review more explicitly focused on humanitarian interventions.

4 The international level need not be discussed here, since the intervention considered was executed unilaterally and no institutionalized international political arena existed in 1898.

5 This is a mere model, which is not true in reality, as there is not only the power of the media, but also the power over the media as a top-down dimension (see Street 2008: 4).

6 Framing is part of the social construction theory and it accounts for the way in which political communi­cators utilize and construct political meanings within a society (see Johnson-Cartee 2005: 28).

7 There are many more conceptual differentiations, such as priming or compelling arguments, establishing additional links (see McCombs 2004: 92, 124).

8 One must bear in mind that history also is a construction of narratives that does not necessarily reflect the objective truth. Instead, it is indeed multi-perspective.

9 Western (2007: 15) writes that there are several accounts that suggest a heightened monitoring of the public’s attitudes toward foreign policy options by political elites.

10 Blatter and Junk (2008: 17) stress that correspondence does not mean causality.

11 They also add a concept of a threshold, which is linked to normative debates on an issue. This is not considered here, because I do not want to measure causal effects as being below or above this threshold, but to explain mere causal mechanisms. The concept of Windows of Opportunity is also skipped, since it cannot be applied to the selected case (due to unilateral action and an anarchic international political are­na, the Window of Opportunity was always open and therefore did not play a role).

12 “Attribute agenda-setting explicitly merges agenda-setting theory with the concept of framing” (McCombs 2004: 88).

13 It must be added that human cognition is boundedly rational.

Excerpt out of 29 pages


Selling Intervention
How the American Press Influenced the Outbreak of the Spanish-American War
University of Luzern  (Politikwissenschaftliches Seminar)
Hauptseminar: The Politics of Humanitarian Intervention
Catalog Number
ISBN (eBook)
ISBN (Book)
File size
1695 KB
Eine Kausalprozess-Analyse über den Einfluss der Medien auf die humanitäre Intervention der USA in Kuba im Jahr 1898.
Humanitarian Intervention, Media, Inhaltsanalyse, Process Tracing, Spanish-American-War, USA, Political Communication, American Press
Quote paper
Samuel Schmid (Author), 2010, Selling Intervention, Munich, GRIN Verlag,


  • No comments yet.
Read the ebook
Title: Selling Intervention

Upload papers

Your term paper / thesis:

- Publication as eBook and book
- High royalties for the sales
- Completely free - with ISBN
- It only takes five minutes
- Every paper finds readers

Publish now - it's free