Propaganda

Propaganda is deliberate, organized communication intended to shape perceptions, beliefs, and behavior in ways that advance the purposes of its sponsors. It is distinguished less by a particular medium or by the presence of falsehood than by its purpose, selective presentation, and relationship to power. Propaganda may employ accurate information, misleading interpretations, fabricated claims, or combinations of these. Its boundaries with persuasion, advertising, public relations, and education depend on the definition used and the circumstances of communication. (csmeyns.github.io)

Propaganda has been important to religious movements, political mobilization, warfare, and the maintenance of political authority. Its reach expanded through printing, mass literacy, film, and broadcasting; digital networks subsequently introduced new forms of automated distribution and audience targeting. Its significance lies both in the messages it circulates and in the institutions and communication systems through which those messages acquire visibility and credibility. The subject therefore encompasses historical campaigns, persuasive techniques, audience responses, and questions concerning evidence, accountability, and freedom of expression. (sagepub.com)

Meaning, Etymology, and Usage

Propaganda is purposeful communication organized to shape an audience’s perceptions, judgments, or conduct in accordance with the communicator’s objectives. Definitions differ over whether deception, concealed intentions, or manipulation are essential features. A widely used communication-centered definition, developed by Garth S. Jowett and Victoria O’Donnell, emphasizes deliberate, systematic efforts to influence understanding and behavior for the propagandist’s intended ends. In this usage, propaganda is identified by the organization and purpose of communication, not simply by its political subject matter or the presence of false statements. Its objectives may include maintaining existing loyalties as well as changing opinions or inducing action. (onlinelibrary.wiley.com)

The word can designate an individual artifact, a coordinated campaign, or an enduring institutional system. Calling a poster “propaganda” identifies it as a vehicle of influence; describing a propaganda campaign refers to the planned relationship among messages, audiences, distribution, and objectives. A propaganda system extends beyond particular campaigns to the organizations and routines that commission, finance, produce, and circulate communication. These levels should not be conflated: an isolated slogan does not by itself demonstrate a coordinated campaign, while a campaign’s purpose may become apparent only when its component messages are examined together. Jowett and O’Donnell’s approach accordingly treats intention, organization, and the relationship between message and context as central analytical questions. (uk.sagepub.com)

The term derives from the Latin title Sacra Congregatio de Propaganda Fide, conventionally translated as the Sacred Congregation for the Propagation of the Faith. Pope Gregory XV established this body of the Catholic Church on January 6, 1622; the constitution Inscrutabili Divinae, issued on June 22 that year, set out its missionary responsibilities. Its task was to coordinate missionary activity and the extension of Catholic belief. In the Latin phrase, propaganda is a gerundive agreeing with fide, expressing the faith’s being propagated or spread, rather than a name for deceptive communication. English subsequently borrowed the institutional designation and broadened it to movements and activities devoted to spreading doctrines or practices. This history dates the word’s institutional origin, not the beginning of organized persuasion: earlier royal, religious, and civic communication must be examined in its own historical settings. (archiviostoricopropaganda.va)

Descriptive and pejorative meanings coexist. In descriptive scholarship, “propaganda” can identify organized influence without deciding beforehand that its claims are false or its objectives objectionable. In ordinary English, however, the label frequently implies dishonesty, distortion, or manipulation, and can itself function as a means of discrediting an opponent’s communication. Translation complicates the distinction. Chinese xuanchuan (宣传) covers publicity and the dissemination or promotion of ideas as well as activities English speakers call propaganda; in official Chinese usage it can have neutral or positive connotations. Translating every occurrence as “propaganda” may therefore import an English-language condemnation absent from the original, whereas “publicity” may understate a specifically political function. The appropriate rendering depends on the activity and context, not merely on dictionary equivalence. (csmeyns.github.io)

Disagreement, political commitment, emotional language, and factual error are consequently insufficient, individually, to identify propaganda. An openly committed argument is not classified solely by the position it supports; an emotional appeal requires examination of its communicative purpose and setting. Likewise, an erroneous statement does not establish a systematic influence effort, while accurate information can be selected and arranged to serve one. Under the purposeful-communication definition, the relevant questions concern who is communicating, which audience is addressed, what response is sought, and how the message participates in an organized effort to secure that response. (uk.sagepub.com)

Boundaries with Related Forms of Communication

Propaganda overlaps with persuasion but is not synonymous with every attempt to influence an audience. Garth S. Jowett and Victoria O’Donnell distinguish persuasion that accommodates the interests of participants from propaganda organized to secure a response serving the propagandist’s objectives. This is an analytical distinction, not a division into entirely separate media or techniques: both may use evidence, emotional appeals, repetition, and credible spokespeople. Advertising and political campaigning identify activities or institutional settings rather than settle their classification. An advertisement promoting a product or a candidate’s speech may belong to a propaganda campaign, but its persuasive purpose alone does not establish that classification. (csmeyns.github.io)

Public relations similarly includes practices that can overlap with propaganda without being reducible to it. In 2012, the Public Relations Society of America adopted a definition emphasizing strategic communication and mutually beneficial relationships between organizations and their publics. That professional formulation allows for consultation and organizational adaptation, rather than only changing an audience’s attitudes. Public diplomacy is distinguished principally by its engagement with foreign publics. The United States Department of State’s archived 2021–2025 mission statement explicitly combined informing and influencing foreign audiences with strengthening relationships. Neither the public-relations label nor acknowledged government sponsorship determines whether a particular operation is propagandistic; its objectives, presentation, and treatment of audience participation remain relevant. (prsa.org)

The boundary with information and education concerns how evidence and inquiry are organized, not whether communication produces an effect. Instruction necessarily selects material, and factual information can change attitudes. The relevant contrast is between enabling understanding—including examination of sources and competing explanations—and arranging communication so that a sponsor’s preferred conclusion is effectively predetermined. Openness to questioning, correction, and contrary evidence therefore matters more than an educational label. Disclosure of sponsorship helps audiences interpret a message but does not make it non-propagandistic: an openly identified organization can still systematically select information to secure its desired response. (csmeyns.github.io)

Factual accuracy and strategic purpose are separate dimensions. In their 2017 Council of Europe report, Claire Wardle and Hossein Derakhshan distinguished misinformation, shared without harmful intent, from disinformation, knowingly false information shared to cause harm. Their framework also recognized “mal-information”: genuine information used harmfully, including disclosure of private material. These categories do not map directly onto propaganda. An accidental falsehood need not constitute an organized influence effort, while a campaign can use accurate statements selectively. The report further distinguishes creators, producers, distributors, and interpreters, whose intentions may differ; someone circulating a campaign’s false claim may believe it sincerely. (rm.coe.int)

Rumor concerns the circulation of claims whose verification remains unsettled, rather than necessarily identifying a strategic sponsor. Satire depends on a communicative context in which exaggeration, invention, or irony is recognizable. Neither category establishes propagandistic intent by itself. Wardle and Derakhshan note, however, that rumors circulate within information disorder and that satire can mislead when its contextual signals disappear. Consequently, a satirical item’s original purpose and its subsequent deployment as purported factual evidence require separate analysis. (rm.coe.int)

Journalism can transmit propaganda when externally prepared material acquires the appearance of independently gathered news. In testimony dated May 12, 2005, the United States Government Accountability Office examined government-produced video news packages designed for television broadcasts. It identified concealed agency authorship—not simply favorable content—as the central problem in the cases examined. This illustrates how the sponsor’s influence operation and a newsroom’s act of dissemination can be distinct. It does not establish that every partisan report or editorial is propaganda: the relevant evidence concerns provenance, editorial independence, attribution, and organized purpose, rather than agreement or disagreement with the position expressed. (gao.gov)

Classification by Source, Audience, and Purpose

Source attribution provides one widely used classification. White propaganda openly and correctly identifies its sponsor; gray propaganda leaves sponsorship unidentified, uncertain, or incompletely disclosed; black propaganda falsely presents a message as originating from someone other than its actual sponsor. These categories concern the relationship between actual and apparent authorship, rather than simply whether a message is public or secret. Institutional definitions nevertheless differ. A 1951 United States information-policy instruction classified concealed sponsorship attributed to a nonhostile source as gray, while reserving black for material attributed to Soviet, Communist, or other hostile sources. Thus, even the familiar three-color scheme has historically specific variants. (history.state.gov)

Attribution and factual accuracy are separate questions. An acknowledged sponsor can circulate false claims, while a falsely attributed publication can contain accurate reporting selected to establish credibility. Garth Jowett and Victoria O’Donnell associate white propaganda with generally accurate information, but emphasize its selective presentation in the sponsor’s interest. “White” therefore does not mean impartial, and “black” need not mean that every substantive statement is fabricated: the deception may lie in the purported identity of the speaker. A campaign’s factual claims and its representations of sponsorship require separate examination. (csmeyns.github.io)

Audience classifications distinguish domestic from foreign recipients, civilians from military personnel, and supporters from opponents or uncommitted groups. These distinctions identify different intended responses: reinforcing supporters, recruiting the undecided, weakening opponents’ morale, or encouraging inaction rather than conversion. They do not necessarily identify separate communication systems; one broadcaster can address several audiences through different programs. During the Second World War, British operations associated with Sefton Delmer included Atlantiksender, directed at U-boat crews, and Soldatensender, directed at German army and Luftwaffe personnel. These were foreign military audiences, unlike the domestic civilian audiences addressed by American war-loan publicity. (csmeyns.github.io)

Purpose and duration supply further distinctions. In Propaganda: The Formation of Men’s Attitudes, published in English in 1965, Jacques Ellul contrasted agitation with integration. Agitation seeks to activate discontent, intensify antagonism, and move people toward disruptive or exceptional action; it can serve an insurgent movement or a government mobilizing against an enemy. Integration seeks sustained adaptation to an established social order, connecting accepted beliefs with habitual participation. In Ellul’s account, a revolutionary organization may employ agitation before gaining power and integration afterward. These are attributed theoretical categories, not a universally accepted division between revolutionary and governmental communication. (theanarchistlibrary.org)

Short-term mobilization and long-term socialization overlap with this distinction without being identical to it. An urgent appeal to buy a war bond specifies an immediate action, whereas repeated narratives of national belonging may cultivate dispositions useful across successive campaigns. Integration can require active participation rather than mere passivity, and agitation can draw upon loyalties established over many years. Ellul’s distinction consequently concerns the intended relationship between people and a social order, not just a message’s emotional intensity or the length of time it circulates. (theanarchistlibrary.org)

Organization of dissemination ranges from centralized direction to participatory circulation. Ellul distinguished vertical communication from leaders to audiences from horizontal processes organized through groups and discussion; horizontal participation, in his account, did not necessarily exclude leadership or supervision. A concrete combination appeared in the American Committee on Public Information’s Four Minute Men program, documented in 1917: more than 75,000 local participants delivered short, vetted speeches promoting measures including Liberty Loan drives. Analytically, this combined acknowledged sponsorship, domestic civilian targeting, immediate mobilization, centralized message supervision, and local participation. The classifications therefore describe overlapping dimensions of the same campaign, rather than mutually exclusive types. (theanarchistlibrary.org)

Early Persuasive Practices and the Expansion of Print

Royal monuments and inscriptions offered selective accounts of power long before mass communication. In ancient Egypt, the thirteenth-century BCE account of Ramesses II’s battle at Kadesh presented the king as the central agent of Egyptian victory; its preservation in papyrus as well as monumental inscriptions illustrates the movement of royal narratives between media. In Mesopotamia, Sennacherib’s palace reliefs at Nineveh depicted the conquest of Lachish, including prisoners, captured goods, and the king reviewing the spoils. Their palace setting distinguished their audience from an unrestricted public: these were statements of military authority encountered within the royal establishment. Such objects support comparisons with propaganda, but their ritual functions and restricted settings caution against treating them as equivalents of modern mass campaigns. (britishmuseum.org)

In the Roman Empire, imperial victories were represented through both public ceremony and circulating objects. Vespasian’s coinage of 71 CE celebrated the suppression of the Jewish revolt: one bronze sestertius paired his portrait with a captive and a mourning personification of Judaea beside a palm tree. The Judaea Capta issues also commemorated the Roman triumph over Jerusalem. Unlike a fixed palace relief, a coin combined an official image with an object used in transactions, repeatedly associating the ruler’s likeness with conquest. Its survival demonstrates the production and circulation of that association, not how consistently ancient users noticed or accepted it. (britishmuseum.org)

Before mass printing, patronage and manuscript production could support identifiable political purposes without addressing large audiences. The eleventh-century Encomium Emmae Reginae was composed by a Flemish monk at Queen Emma’s request; a surviving manuscript depicts the author presenting it to her while her sons Harthacnut and Edward look on. Here the commissioner and the work’s laudatory purpose are explicit. Religious preaching could likewise be organized toward a particular action: scholarship on crusade recruitment identifies sermons, papal letters, and public ritual as interconnected instruments, including Urban II’s preaching at Clermont in 1095. These commissioned narratives and recruitment practices should be distinguished from devotional manuscripts or sermons whose religious content alone does not establish a strategic campaign. (searcharchives.bl.uk)

Reformation controversies joined theological argument to reproducible images and vernacular publication. The Passional Christi und Antichristi, published at Wittenberg in 1521, contained 26 woodcuts designed by Lucas Cranach, contrasting Christ’s conduct with that of the pope, identified polemically as Antichrist. Paired scenes made comparison itself the argumentative mechanism; an edition printed at Erfurt later that year documents reproduction beyond its initial place of publication. Catholic responses also used print rather than simply opposing it. Henry VIII’s Defence of the Seven Sacraments answered Martin Luther in 1521, while the English College at Rheims produced an English New Testament, printed by John Fogny in 1582. Its title page explicitly advertised explanations of difficult passages and the correction of purported translation errors, combining vernacular access with confessional interpretation. (editions.mml.ox.ac.uk)

Revolutionary print depended on networks of authors, printers, sellers, and newspaper editors rather than necessarily on government direction. Thomas Paine’s Common Sense, first published anonymously on January 10, 1776, argued for American independence. Reprinting carried it between colonial towns: a surviving Providence edition names John Carter as printer and seller, while newspapers reproduced the argument and published hostile responses. Nineteenth-century nationalism similarly combined political prose with cultural production. Thomas Davis, Charles Gavan Duffy, and John Blake Dillon founded Dublin’s Nation in 1842; Davis used essays and ballads, including “A Nation Once Again” in 1844, to cultivate Irish nationality among its readers. (loc.gov)

The widening market for print connected political persuasion with literacy, commercial distribution, and imperial display. Britain’s parliamentary debates over newspaper taxation in 1855 explicitly linked reading instruction with access to affordable news, while exposing disagreements over taxation, postal costs, and poorer readers. Exhibition publicity connected that commercial press to public spectacle: the Illustrated London News reported and pictured preparations for the Great Exhibition of 1851. Organized under a Royal Commission with Prince Albert’s support, the exhibition presented manufacturing and imperial possessions to more than six million visitors, with Britain and its territories occupying the building’s western half. Objects, spatial arrangement, ceremony, and illustrated reporting together presented an officially sponsored image of industrial and imperial achievement. (hansard.parliament.uk)

The First World War and Mass Mobilization

British wartime publicity combined government direction with the authority and distribution networks of private publishing. From 1914, the War Propaganda Bureau at Wellington House, headed by Charles Masterman, enlisted established writers and circulated material through commercial publishers, including Hodder & Stoughton and Methuen. This arrangement could make officially sponsored arguments appear to originate independently. France developed a parallel combination of official coordination and civilian participation: the Maison de la Presse, established in January 1916, brought together journalists, linguists, diplomats, and military personnel, while associations, newspapers, and artists also produced patriotic material. French news management rested partly on a separate apparatus established in August 1914, whose censors scrutinized newspaper proofs before publication. Publicity production and restrictions on journalism were connected but distinct activities. (ssoar.info)

In Germany, military authorities exercised substantial influence over news distribution and publication. The Kriegspresseamt, established in October 1915, became a center for military press management and propaganda, operating through briefings and relations with an existing commercial press rather than replacing newspapers with a single state publication. The German and French systems thus gave military control of information a prominent institutional place, whereas Wellington House initially emphasized discreet sponsorship and cooperation with literary intermediaries. None of these arrangements meant that every patriotic newspaper article or privately published book resulted from government instructions. (unvis.it)

The United States established the Committee on Public Information (CPI) by executive order on April 13, 1917, with journalist George Creel as chairman. Its responsibilities included releasing government news, sustaining morale, overseas publicity, and administering voluntary press censorship. It published the Official Bulletin, supplied news releases, and organized photographic and film activities. Its Four Minute Men connected national messaging to local speakers and commercial cinemas: approximately 75,000 volunteers participated, addressing audiences during short intervals and subsequently in churches, union halls, and other venues. Campaign bulletins supplied subjects and arguments rather than requiring every speaker to recite an identical text. Their sequence made mobilization concrete: selective-service registration in May 1917, food conservation in July, Liberty Loan subscriptions, and a shipbuilding campaign in January–February 1918. Recruitment, household economies, industrial labor, and lending to the government became coordinated forms of civilian participation. (archives.gov)

Posters translated these objectives into recognizable figures and immediate actions. James Montgomery Flagg’s 1917 army recruiting poster depicted Uncle Sam pointing at the viewer, with space for a local recruiting address; its nationally recognizable image could therefore direct readers to a specific enlistment office. British posters similarly connected recruitment to the defense of civilians: the Parliamentary Recruiting Committee’s 1914 “Remember Belgium” poster placed an advancing soldier before a burning settlement. Food-saving publicity addressed another audience and behavior: the Smithsonian’s collections document cooperation among the CPI’s Division of Pictorial Publicity, the United States Food Administration, and artist Adolph Treidler on a poster urging conservation. (loc.gov)

Atrocity narratives required distinctions between documented violence, embellished reporting, and fabrication. German troops executed civilians and burned settlements during the 1914 invasion of Belgium; these events were not inventions merely because Allied publicity exploited them. Nevertheless, individual sensational allegations required separate verification. In April 1917, British newspapers circulated the “corpse factory” story, alleging that Germany processed soldiers’ bodies for industrial materials—a notorious false atrocity narrative whose circulation should not automatically be equated with proof of government authorship. Visual records also mixed documentation and reconstruction: The Battle of the Somme (1916) preserved authentic battlefield footage but included staged “over the top” sequences filmed because close-quarter attacks could not be recorded directly. (collection.nam.ac.uk)

The wartime integration of news, entertainment, public speaking, and economic appeals furnished both material and political context for subsequent propaganda research. Postwar distrust centered particularly on the belief that official publicity and atrocity stories had misled civilian populations. Harold Lasswell’s Propaganda Technique in the World War (1927) examined the conflict’s organized persuasion, while Arthur Ponsonby’s Falsehood in War-Time (1928) exemplified the critical reaction to wartime claims. These responses made the relationship between factual reporting, selective presentation, and organized influence a central problem of interwar analysis. (propagandaarchive.org)

Interwar Regimes and the Second World War

In Fascist Italy, publicity administration developed into the Ministry of Press and Propaganda in 1935, renamed the Ministry of Popular Culture in May 1937. Cultural supervision and message production were institutionally connected: the surviving theatrical censorship archive contains approximately 18,000 files on theatrical and radio works approved or rejected between 1931 and 1944. Scripts submitted for authorization made cultural gatekeeping an administrative process, distinct from—but complementary to—the commissioning and circulation of persuasive material. (search.acs.beniculturali.it) In Nazi Germany, the Reich Ministry of Public Enlightenment and Propaganda was established in March 1933 under Joseph Goebbels. Its jurisdiction encompassed broadcasting, cinema, theatre, and much of the press, while Goebbels simultaneously headed the Nazi Party propaganda apparatus. Authority was extensive but not undivided: press control also involved Max Amann and Otto Dietrich. (encyclopedia.ushmm.org)

These institutions served specifically Nazi racial objectives. Coordinated antisemitic campaigns helped prepare public acceptance of discriminatory measures in 1935 and 1938; wartime messages presented Jews as dangerous enemies and sought acquiescence in their removal. Fritz Hippler’s The Eternal Jew (1940) used the apparent authority of documentary film to disseminate dehumanizing representations. Propaganda contributed to the Holocaust both by legitimizing persecution and by concealing killing: camp officials forced prisoners to send reassuring postcards, while the staged Red Cross visit to Theresienstadt on June 23, 1944, misrepresented conditions through renovated surroundings and instructed prisoner behavior. These were connections between persuasion, deception, and coercion, not evidence that communication alone produced genocide. (encyclopedia.ushmm.org)

The Soviet Union operated through Communist Party and state cultural controls rather than a direct equivalent of Goebbels’s ministry. An April 4, 1928, memorandum from Glavlit, the censorship agency, proposed prohibiting publications considered hostile to proletarian dictatorship or damaging to Soviet prestige. Party leadership also intervened in literary organization and publishing. Wartime communication could adapt earlier ideological priorities: after the German invasion in 1941, Stalin revived the Russian Orthodox Church’s public role to strengthen patriotic support. (loc.gov) In imperial Japan, the Cabinet Intelligence Bureau, established on December 5, 1940, consolidated information functions previously divided among cabinet, foreign-affairs, army, and navy offices. Its predecessor had launched the photographic weekly Shashin Shūhō on February 16, 1938; the journal reportedly reached a circulation of 200,000 by March 1941. This cabinet-centered coordination differed from both Nazi party–ministry overlap and Soviet party supervision. (jacar.go.jp)

Allied agencies likewise connected government objectives with established media organizations. Franklin D. Roosevelt created the United States Office of War Information on June 13, 1942. At its request, five American newsreel companies formed the privately organized but fully government-funded United Newsreel Corporation. Its United News series, produced principally for overseas audiences and reportedly in sixteen languages, combined selected battlefield and home-front footage with commentary advancing the Allied cause. Government sponsorship therefore did not necessarily mean production exclusively by civil servants. (archives.gov)

British arrangements distinguished domestic publicity from political warfare abroad. The Political Warfare Executive, formed in August 1941, combined personnel and functions from the Ministry of Information, European sections of the BBC, and the propaganda branch of the Special Operations Executive. Broadcasting, printed publications, and airborne leaflets addressed different audiences across enemy and occupied territory. (nationalarchives.gov.uk) At home, “Dig for Victory” promoted food production, while “Careless Talk Costs Lives” connected everyday conversation with military security; an archived government radio broadcast from 1940–1943 explained how civilians could obstruct enemy agents. (nationalarchives.gov.uk) Military leaflets sought more immediate behavioral responses: the National Army Museum preserves an Allied safe-conduct leaflet used in Burma around 1944 to encourage Japanese soldiers to surrender. (collection.nam.ac.uk)

Wartime persuasion must nevertheless be separated analytically from restrictions and force. A campaign supplied preferred interpretations or requested action; censorship prevented material from circulating; coercion imposed conduct through threats or punishment. Italian script authorization and German exclusion of Jewish journalists demonstrate how persuasive institutions could also restrict who was permitted to speak. (movio.beniculturali.it) Military deception similarly had a distinct operational purpose, although concealed sponsorship and fabricated appearances could connect it with propaganda. British archival records accordingly include both attributed publicity and covert political-warfare operations, which cannot be treated as institutionally identical merely because both used broadcasting or print. (nationalarchives.gov.uk)

The Cold War, Decolonization, and Revolutionary Campaigns

During the Cold War, international broadcasting combined news, commentary, and cultural programming for audiences beyond national borders. Soviet Radio Moscow broadcast domestically and abroad in multiple languages; its output also supplied Western monitoring services with evidence of Soviet policy, notably during the October 1962 Cuban missile crisis. American-sponsored Radio Free Europe and Radio Liberty addressed Eastern European and Soviet audiences respectively, employing émigré broadcasters rather than simply retransmitting American domestic programs. Their apparently private institutional structure concealed substantial government financing: a July 1971 Senate report stated that secret Central Intelligence Agency budgets had almost entirely covered their costs for approximately twenty years, including more than $30 million in the preceding fiscal year. These arrangements distinguished covert sponsorship from openly identified state broadcasting without making sponsorship alone a measure of factual accuracy. (iwm.org.uk)

Cultural diplomacy sought political influence through performances, exhibitions, and exchanges rather than continuous political exhortation. Between 1954 and 1959, approximately 140 American groups of performers and athletes visited more than ninety countries. Dizzy Gillespie’s interracial band toured the Middle East, Yugoslavia, and Greece in 1956 under State Department sponsorship; Benny Goodman performed in six Soviet cities in 1962 under a bilateral cultural-exchange agreement. Such programs presented American cultural achievement while leaving performers some room to contest official representations, including racial discrimination. Soviet visual publicity likewise made achievement politically representative. The State Archive of the Russian Federation preserves A. Kruchina’s poster celebrating Yuri Gagarin, dated April 17, 1961—five days after his spaceflight—and identifies it as part of mass poster production both centrally and in the Soviet republics. Its rapid publication documents an organized commemorative response, not how viewers interpreted it. (wwws.loc.gov)

In China, revolutionary communication joined centrally promoted campaigns to established publishing and retail networks. Poster House’s exhibition documentation records the circulation of posters through government institutions and the sale of inexpensive smaller versions through Xinhua Bookshops for homes and classrooms. Guarantee the Grain Harvest (c. 1961) addressed agricultural production, while Gao Rufa’s To Everyone According to Their Work; The One Who Works More Earns More (1962) publicized labor incentives. During the Cultural Revolution of 1966–1976, local publishers participated in poster production, including images promoting Mao Zedong’s authority. The same exhibition catalogue identifies Mao Says the Hope Is on Your Shoulders (c. 1967), Long Live Chairman Mao! (1971), and a Dazhai-model agricultural poster credited to the Jin Xian Worker-Peasants’ Propaganda Team and Yan Chengfu (c. 1976). These records establish campaign themes, production credits, and distribution mechanisms, rather than uniform public belief. (posterhouse.org)

In Cuba, international solidarity generated a distinct poster-distribution system. OSPAAAL—the Organization of Solidarity with the People of Asia, Africa and Latin America—was founded following the 1966 Tricontinental Conference in Havana to connect liberation movements and activists internationally. Posters circulated through its Tricontinental magazine and typically carried Spanish, English, French, and Arabic text. The Victoria and Albert Museum records Alfredo Rostgaard’s Create, Two, Three...Many Vietnams (1967) and Daysi García’s Angola Day of Solidarity (1969). Their multilingual captions and geographically specific commemorations addressed transnational political networks rather than only Cuban citizens; their preservation establishes the intended appeal, not its persuasive success. (vam.ac.uk)

Anti-colonial broadcasting also developed institutions whose purposes exceeded alignment with either superpower. Radio Republik Indonesia, established on September 11, 1945, supported resistance to Dutch colonial restoration. Archival research on its Surakarta station traces a subsequent shift toward national-cultural construction during the 1950s and 1960s. Under broadcasting-bureau policy, Javanese-language and regional performance traditions—including gamelan, wayang wong, kethoprak, and shadow puppetry—appeared alongside national music and singing competitions. National integration therefore involved selecting and broadcasting regional culture, not simply replacing it with a single standardized political message. (mpn.komdigi.go.id)

Postcolonial Tanzania connected broadcasting to Julius Nyerere’s promotion of Swahili culture through state-supported musical production. Radio Tanzania’s surviving recordings include government-sponsored dance music, moral dramas, and educational songs. Sponsorship operated through multiple institutions: the immigration department supported Uhamiaji Jazz, the ruling party’s youth organization supported Vijana Jazz, and the national workers’ union owned NUTA Jazz. These arrangements embedded political and civic communication within entertainment and organizational patronage, rather than restricting it to official speeches or imported Cold War material. (tanzaniaheritageproject.org)

Sponsors, Institutions, and Communication Systems

Propaganda production can distribute authority among a sponsor, an operating institution, media producers, and local disseminators. A government may establish objectives and supply funds, while an agency commissions publications or broadcasting; armed forces may maintain specialist units integrated into military planning. The United States Army’s 2017 description of its Psychological Operations Branch, for example, placed military information support operations within commanders’ operational objectives. Political parties can supply additional organizational channels: in Nazi Germany, propaganda directives issued at daily Berlin press conferences passed through Nazi Party offices to regional and local newspapers. These arrangements distinguish the institution setting a campaign’s purpose from those producing and circulating particular messages. (api.army.mil)

Religious organizations, businesses, advocacy organizations, and informal movements provide other institutional settings, although organizational identity alone does not establish that a particular communication is propaganda. The Catholic Church combined missionary administration with linguistic and publishing infrastructure: the Congregation de Propaganda Fide’s printing house issued grammars, religious controversy, and devotional works for eastern missions; its 1639 catalogue listed publications produced since 1628. Corporate communication likewise extends beyond product advertisements. The University of California, San Francisco’s tobacco-document archive preserves internal records concerning advertising, research, and political activities, allowing financial sponsorship to be investigated separately from the public identity of a spokesperson or organization. (archiviostoricopropaganda.va)

Commissioning, financing, editorial supervision, and distribution are distinct operations. A commission specifies the intended audience and product; financing supplies staff, printing, airtime, or transport; supervision determines who approves copy; distribution connects finished material to audiences. These functions need not belong to one bureaucracy. The United States Committee on Public Information, established on April 13, 1917, maintained separate activities involving publications, films, photographs, and public speaking. Its Four Minute Men network connected central organization with local delivery. Surviving records include bulletins and correspondence alongside speakers’ report cards recording venue, date, attendance, and topic. Thus, unpaid participation could be institutionally organized: volunteering did not necessarily mean that speakers selected their subjects independently. (archives.gov)

Intermediary organizations can combine official resources with private fundraising and civic participation. Records of a 1952 meeting about the Crusade for Freedom specified cooperation among the National Committee for a Free Europe, the Central Intelligence Agency, and the Department of State, including consultation over what local groups should be told. The discussion reported approximately $2 million in gross fundraising during the preceding campaign and proposed activities such as letters and essay contests. A January 27, 1969 memorandum separately described Radio Free Europe and Radio Liberty as CIA covert-action projects costing approximately $32 million annually. These records demonstrate why private donations, voluntary participation, and concealed state financing must be reconstructed as separate relationships rather than treated as mutually exclusive explanations. (history.state.gov)

Institutional influence can also operate without campaign-specific instructions. Ownership gives proprietors powers over appointments and budgets; advertising dependence can make particular audiences commercially valuable; regular access to official briefings and prepared material can reduce reporting costs. Edward Herman and Noam Chomsky identified ownership, advertising, and reliance on government and business sources among the institutional “filters” in their 1988 account of mass media. These are proposed mechanisms of selection and incentive, not evidence that every published item was centrally commissioned. They differ analytically from documented editorial commands, such as the Nazi ministry’s instructions specifying which stories newspapers could publish and how they should present them. (chomsky.info)

Coordination therefore requires evidence beyond agreement in vocabulary or political position. Funding records, instructions, correspondence, reporting requirements, and shared operational arrangements can establish relationships that message similarity alone cannot. Informal participants may independently adopt a movement’s language, while professional journalistic routines or reliance on the same source can produce similar coverage without common supervision. A parallel distinction appears in Facebook’s October 2019 explanation of coordinated inauthentic behavior: its stated focus was groups misleading audiences about their identity and activities, rather than simply publishing particular opinions. Coordination, concealed identity, and voluntary amplification are consequently separate questions; an organized campaign can recruit genuine supporters without directing every subsequent repetition of its messages. (oac.cdlib.org)

Symbols, Narrative, and Persuasive Techniques

Framing organizes a message around selected problems, causes, moral judgments, and remedies. Robert M. Entman’s 1993 formulation emphasizes selection and salience: interpretation depends not only on what appears, but also on what receives prominence and what remains absent. Selective omission can therefore produce a misleading account without making individual statements false—for example, by presenting benefits while excluding costs or competing explanations. A narrative connects these selections into a sequence of threat, struggle, and promised resolution. These devices also occur in ordinary journalism and rhetoric; no single technique establishes propagandistic intent independently of its purpose and context. (onlinelibrary.wiley.com)

Repetition keeps a claim or association prominent, while slogans compress an argument into a portable formula that can recur across speeches, posters, and broadcasts. Emotionally loaded vocabulary supplies an evaluation in advance of supporting evidence: labels can make a policy appear patriotic, shameful, threatening, or inevitable. American Liberty Bond posters during World War I combined short exhortations with images invoking fear, pride, obligation, and guilt. Euphemism works in the opposite direction, reducing the apparent violence of an action. During the Holocaust, German officials described deportations as “resettlements” and stamped passports “evacuated,” concealing coercion and murderous destinations behind administrative language. (arxiv.org)

Authority appeals invoke an office, institution, or recognized expert; testimonials attach a named person’s endorsement to a cause. Neither necessarily establishes the endorser’s relevant competence or the claim’s validity. Symbolic association transfers prestige through flags, uniforms, religious imagery, or national personifications. James Montgomery Flagg’s 1917 Uncle Sam recruiting poster combined national symbolism with a pointing hand and direct address; more than four million copies were printed in 1917–1918. “Plain folks” presentations instead emphasize ordinary clothing, domestic life, or familiar speech to suggest shared experience. A brochure dating to about 1939 advertised Heinrich Hoffmann’s photographic books showing Hitler at home, with children, and among admiring crowds. Bandwagon claims present participation as normal or widespread: a Liberty Bond poster explicitly urged viewers to “join the crowd.” (d17j94wz7065tl.cloudfront.net)

Scapegoating assigns responsibility for complex difficulties to a designated group, while stereotyping substitutes a supposedly uniform group character for individual differences. Dehumanization goes further by representing people as less than human or associating them with contamination and disease. A Nazi poster published in occupied Poland in March 1941 combined an antisemitic caricature with imagery of lice and death. Conspiracy narratives supply a complementary explanation: apparently separate events become evidence of an alleged hidden enemy’s coordinated activity. Nazi propaganda falsely linked Jews collectively to war and communism, turning political developments into accusations against an entire population. These were mechanisms of exclusion and persecution, not factual descriptions of the targeted group. (encyclopedia.ushmm.org)

False dilemmas restrict the apparent choices to two alternatives, excluding intermediate positions or other solutions. A message may thus equate disagreement with disloyalty rather than address the objection itself. Distraction shifts attention from the disputed issue: a red herring introduces irrelevant material, while whataboutism redirects criticism toward another actor’s alleged misconduct. Such moves differ from a relevant comparison because they substitute a change of subject for an answer. The technique inventory used in the 2020 SemEval propaganda-identification task distinguished these argumentative devices from repetition, slogans, authority appeals, and loaded language. (arxiv.org)

Visual arrangement can perform the same selective work as wording. Size, position, contrast, and captions determine which figure appears authoritative, vulnerable, or threatening. Maps can privilege a territorial claim through color, projection, boundaries, and selective inclusion; Cornell University Library’s persuasive-cartography collection documents these practices. Numerical presentation can similarly emphasize selected comparisons while withholding contextual information. Photographs need not be fabricated to mislead: selection, captions, and sequencing can give genuine images an unrepresentative meaning. In the Nazi film made at Theresienstadt in 1944–1945, scenes of recreation and productive activity presented a false account of ghetto life. Editing also controlled wartime newsreels: twelve to eighteen hours of weekly footage were reduced to twenty to forty minutes, making selection an integral part of the resulting representation. (persuasivemaps.library.cornell.edu)

Psychological Mechanisms and Social Context

Attention to a message does not necessarily entail scrutiny of its arguments. Research in social psychology distinguishes effortful evaluation from reliance on cues such as a communicator’s apparent expertise. In a 1981 experiment, Richard Petty, John Cacioppo, and Rachel Goldman presented 145 undergraduates with arguments for comprehensive senior examinations. When students expected the examinations to affect them the following year, argument quality mattered more; when implementation was placed ten years ahead, source expertise mattered more. Personal relevance thus affected how recipients processed persuasion, rather than simply making them more or less susceptible. (doi.org)

Repetition can influence judgment through familiarity and memory. The illusory truth effect describes the tendency to judge repeated statements as more truthful than unfamiliar ones, partly because repetition makes them easier to process. Lisa Fazio and colleagues demonstrated in two experiments published in 2015 that this effect could occur even for statements contradicted by participants’ existing knowledge: stored information was not always retrieved or applied during judgment. Memory for content and memory for its origin can also diverge. Carl Hovland and Walter Weiss’s 1951 source-credibility experiments found that recipients initially discounted communications attributed to untrustworthy sources, but that this discounting diminished over time as content became less closely associated with its source. These findings concern particular judgment and memory processes, not an automatic progression from repeated exposure to durable political allegiance. (pubmed.ncbi.nlm.nih.gov)

Prior commitments shape both exposure and interpretation. In her 1990 account of motivated reasoning, Ziva Kunda distinguished motivation to reach accurate conclusions from motivation to reach preferred conclusions; the latter can influence which memories, arguments, and evaluative strategies people use. Political identity can also affect what information people select before evaluating it. Shanto Iyengar and Kyu Hahn’s 2009 experiment found partisan preferences for news-source labels even on comparatively nonpolitical subjects such as travel. Such selection can repeatedly expose audiences to congenial interpretations without requiring centralized control over every encounter. Emotional arousal is likewise not a single mechanism: Nicholas Valentino and colleagues’ 2011 study, combining an experiment with electoral surveys, found a distinctive mobilizing role for anger compared with anxiety and enthusiasm. (pubmed.ncbi.nlm.nih.gov)

Group settings introduce pressures distinct from private conviction. In Solomon Asch’s line-comparison experiments, reported in 1955, participants faced a unanimous group whose other members deliberately gave incorrect answers. Participants followed the erroneous majority on 36.8 percent of the critical judgments, compared with errors below 1 percent under ordinary conditions. Yet resistance remained substantial, and an ally disrupted the pressure of unanimity. These results identify conditions under which expressed judgments follow a group; they do not establish that everyone privately accepts a majority’s claims. For propaganda analysis, social norms, anticipated isolation, and visible agreement must therefore be distinguished from internalized belief. (web.mit.edu)

Messages also pass through interpersonal relationships. Elihu Katz’s 1957 discussion of the two-step flow of communication examined how opinion leaders mediate between mass media and other recipients. Friends, colleagues, and community contacts can select information, explain its relevance, or challenge its interpretation. Katz treated the model as a hypothesis refined by successive studies, not a universal rule requiring exactly two stages. The mechanism makes audience participation central: communication is filtered through relationships and discussion rather than transmitted unchanged from an institution into isolated minds. (academic.oup.com)

Different responses consequently require separate descriptions. Persuasion changes an attitude; reinforcement strengthens an existing commitment; demobilization reduces willingness to act; confusion introduces uncertainty rather than agreement; apparent compliance concerns outward behavior without necessarily indicating private acceptance. Haifeng Huang’s 2018 survey experiment using Chinese campaign messages illustrates this separation: overt official propaganda worsened evaluations of the regime while also signaling state power and reducing reported willingness to protest. A recipient’s interpretation of official communication can therefore coexist with skepticism toward its substantive claims, and behavioral withdrawal need not indicate ideological conversion. (journals.uchicago.edu)

Media, Performance, and Material Culture

Printed formats organize encounters with political messages differently. Posters combine prominent images and short texts for repeated viewing in streets, workplaces, and recruiting offices; pamphlets provide portable arguments that readers can retain, circulate, and discuss. Newspapers place political messages within a recurring mixture of reporting, commentary, advertisements, and illustrations. These formats therefore offer different balances of visibility, detail, and repetition rather than interchangeable channels. James Montgomery Flagg’s 1917 recruitment poster, depicting Uncle Sam pointing outward above “I Want You,” exemplifies a design addressed to a passing individual: the gesture attracts attention, while the text directs the viewer toward enlistment. More than four million copies were printed during the First World War, making reproduction and physical placement integral to its circulation. (loc.gov)

Radio carries speech and music across distances without requiring audiences to read, while public-address systems make listening possible within a particular street, factory, or assembly. Unlike a retained pamphlet, an unrecorded broadcast unfolds at the broadcaster’s pace; scheduled repetition can reproduce the same message across dispersed locations. In Nazi Germany, speeches entered homes and workplaces through radio and reached streets through loudspeakers. Government subsidies supported the inexpensive Volksempfänger, or “People’s Receiver”; approximately 1.5 million had been sold by 1935. Broadcasting and amplification thus connected domestic listening with organized communal reception, although access to a receiver did not establish agreement with its contents. (encyclopedia.ushmm.org)

Cinema combines moving images with an ordered sequence that exhibitors can reproduce for successive audiences, but circulation requires prints, projection equipment, venues, and booking arrangements. The British official film The Battle of the Somme (1916), photographed by Geoffrey Malins and John McDowell, combined frontline footage with some recreated scenes. Its distribution notice advertised a screening at London’s Scala Theatre on August 10 and explicitly rejected exclusive exhibition rights. Commercial cinemas thereby supplied an existing infrastructure through which official material could circulate beyond government premises. Television subsequently brought audiovisual political messaging into domestic programming. Lyndon B. Johnson’s “Daisy” advertisement, broadcast as a paid spot only once, on September 7, 1964, juxtaposed a child counting flower petals with a nuclear countdown and explosion; its subsequent news coverage illustrates how circulation could extend beyond purchased airtime. (assets.publishing.service.gov.uk)

Literature, theater, songs, animation, and entertainment can carry political meanings without taking the form of an explicit address. Commissioned productions must nevertheless be distinguished from independently created works that officials subsequently promote or appropriate. The American military’s Private Snafu cartoons, introduced in 1943, provide a documented example of commissioned entertainment: Theodor Seuss Geisel contributed to scripts, Warner Bros. supplied animation, and the films appeared in the Army–Navy Screen Magazine. Their incompetent protagonist demonstrated undesirable conduct through humorous consequences, including failures of secrecy and security. Conversely, the war-related writing of figures such as Siegfried Sassoon and J. R. R. Tolkien drew on personal experience; political subject matter alone does not demonstrate an official commission or campaign function. (unwritten-record.blogs.archives.gov)

Rallies and exhibitions turn spectators’ movement and participation into parts of the presentation. At Nuremberg, where Nazi Party rallies took place from 1933 to 1938, monumental architecture provided settings for choreographed assemblies; uniforms, formations, and ceremonial gestures made organizational membership visually legible. Exhibitions could similarly direct interpretation through spatial arrangement rather than argument alone. In Munich’s 1937 “Degenerate Art” exhibition, organizers crowded more than 600 confiscated artworks into an improvised gallery, displayed some without frames or with incorrect labels, and surrounded them with hostile slogans. The installation attempted to discredit the works through their conditions of encounter, not merely through their selection. (museums.nuernberg.de)

Monuments and commemorations extend communication through durable material culture and recurring ritual. Their inscriptions, placement, and ceremonial use require separate interpretation: a memorial is not automatically propaganda. At Bushey, England, a monument unveiled on March 17, 1922, combined a figure holding a laurel wreath with a cenotaph and named local war dead. Its inscription presented their deaths through the language of virtue, sacrifice, and attachment to the country. Such objects preserve a particular public vocabulary of remembrance, while later ceremonies and additions can give the same physical structure further historical meanings. (memorials.iwm.org.uk)

Digital Networks and Computational Propaganda

Computational propaganda combines algorithms, automation, and human management to organize political communication through digital networks. Samuel Woolley and Philip Howard’s 2018 formulation emphasizes the purposeful distribution of misleading information through social media: operators select narratives and audiences, software schedules or multiplies activity, and platforms rank and distribute the resulting messages. The term therefore describes a sociotechnical operation, not simply machine-written content. Oxford’s 2020 Global Inventory of Organized Social Media Manipulation, published in 2021, documented organized manipulation in 81 countries and automated accounts in 57; these figures record documented activity, not the proportion of users exposed or persuaded. (navigator.oii.ox.ac.uk)

Political bots can automatically publish posts, repeat slogans, or redistribute links, while human operators maintain fabricated biographies, answer comments, and manage groups. Networks may combine automated, manually controlled, impersonated, and authentic accounts. Automation and deception are separate properties: a disclosed automated account need not conceal its origin, whereas a manually operated persona can falsely claim to be an independent citizen. Facebook’s 2019 account of “coordinated inauthentic behavior” accordingly focused on misleading audiences about an operation’s identity and organization, rather than requiring every message to be false. Oxford’s inventory also identified private contractors, paid citizens, and influencers within influence operations; payment and concealed sponsorship cannot be inferred merely from sympathetic posting. (demtech.oii.ox.ac.uk)

Paid promotion adds selective distribution to these networks. Operators can use advertising infrastructure to direct messages toward particular audience segments rather than publishing the same appeal to everyone; Oxford’s 2020 inventory identified data-driven targeting strategies in 30 countries. Circulation also extends beyond public feeds. Its 2018 inventory documented campaigns using WhatsApp, Telegram, and WeChat in approximately one-fifth of the 48 countries examined. A concrete cross-platform mechanism appeared in Facebook’s September 2020 takedown report: networks combined fake and authentic accounts to manage pages and direct users toward external websites. Such arrangements connect platform-based amplification with separately controlled publishing infrastructure. (demtech.oii.ox.ac.uk)

Recommendation systems mediate distribution without necessarily sharing an operator’s political objective. Meta’s 2023 technical explanation described models that use content characteristics, interaction histories, and engagement signals to predict users’ interest and rank candidate posts. Likes, comments, views, and shares consequently function both as visible feedback and as inputs into distribution systems, although their significance varies across systems. Visibility does not establish persuasion: in Andrew Guess and colleagues’ 2023 study of the 2020 United States election, replacing Facebook and Instagram’s algorithmically ranked feeds with chronological feeds substantially changed exposure and platform use but produced no significant change in measured polarization or other key political attitudes during the three-month study. (ai.meta.com)

Generative artificial intelligence adds synthetic portraits, cloned voices, and generated text to this repertoire. In its March 2021 report, Facebook said it had disrupted seven operations using synthetic profile photographs since 2019, describing the photographs as likely machine-generated. On January 21, 2024, robocalls to New Hampshire voters used an artificial imitation of Joe Biden’s voice to discourage participation in the presidential primary. OpenAI’s May 2024 investigation separately documented operations generating multilingual comments, articles, account biographies, and translations; the company reported no meaningful increase in their reach or engagement attributable to its services. These findings distinguish demonstrated production capabilities from demonstrated influence. They also require separate attribution of the synthetic artifact, the distributing accounts, and the sponsoring organization: identifying generated media alone does not establish who commissioned it or what effect it had. (about.fb.com)

Propaganda Scholarship and Major Interpretive Traditions

Walter Lippmann’s Public Opinion (1922), written in the aftermath of the First World War, examined the gap between complex events and the simplified representations through which people understand them. His “pseudo-environment” comprised the mental pictures on which individuals act, although their actions have consequences in the external world. Stereotypes, restricted access to events, and the selection routines of journalism helped explain this gap. Propaganda could exploit these conditions, but Lippmann’s analysis extended beyond deliberately deceptive campaigns: it questioned whether citizens could acquire the knowledge required by conventional accounts of democracy. His proposed organized intelligence services would supply decision-makers with independently investigated information, rather than merely improve persuasive messaging. (gutenberg.org)

Harold D. Lasswell approached wartime persuasion through political science and social psychology. Propaganda Technique in the World War (1927) compared American, British, French, and German experience. His accompanying article, “The Theory of Political Propaganda,” defined the subject through the management of collective attitudes using culturally meaningful symbols. A collective attitude, in this account, was a pattern of individual responses, not a separate “group mind.” Symbols could reinforce existing commitments as well as redirect them; enlistment and war-bond purchases were examples of the participation propagandists sought. This analytical orientation should be distinguished from endorsement of any particular campaign. Lasswell also advanced content analysis and the communicator–message–channel–audience–effect framework, connecting propaganda research to broader communication studies. (mitpress.mit.edu)

Edward Bernays wrote as a public-relations practitioner seeking to establish the legitimacy of his profession. Crystallizing Public Opinion (1923) and Propaganda (1928) combined observations about group affiliations and opinion leadership with advocacy of organized persuasion. In the latter, he argued that institutions could influence conduct indirectly by securing respected intermediaries and creating events that became news. His discussion of promoting velvet, for example, described cultivating fashion authorities and designers rather than simply advertising directly to consumers. Unlike a purely descriptive account, Bernays’s writing presented professional opinion management as a necessary instrument of modern social organization. His June 1928 speech “Propaganda and Impropaganda” explicitly distinguished uses he considered proper from improper ones and advocated technically efficient political publicity. (tile.loc.gov)

Jacques Ellul shifted attention from identifiable campaigns to the organization of modern society. Propagandes (1962), translated as Propaganda: The Formation of Men’s Attitudes (1965), argued that propaganda depended on technological media, mass organization, and audiences’ needs for explanation and participation. His “sociological propaganda” included the diffuse transmission of a society’s way of life, not necessarily centrally planned persuasion. “Integration propaganda” sustained adjustment to an established order, whereas agitation mobilized people against an opponent. Compared with campaign-centered approaches, Ellul’s account therefore broadened both the timescale and the location of influence: institutions and everyday cultural environments could prepare people to accept subsequent political appeals. These were propositions within his sociological interpretation, not proof that every instance of cultural transmission constituted propaganda. (ellul.org)

Edward S. Herman and Noam Chomsky’s Manufacturing Consent (1988) proposed a propaganda model of institutional filtering in American mass media. Its five original filters were concentrated, profit-oriented ownership; advertising dependence; reliance on official and business sources; organized punitive criticism, termed “flak”; and anticommunism. They argued that these pressures systematically shaped selection, emphasis, and the boundaries of acceptable debate. The explanation did not require a central authority issuing instructions or journalists consciously falsifying reports: professional adaptation, resource dependence, and internalized assumptions could produce patterned coverage. Their model concerned institutional news production rather than a demonstration that every audience accepted the resulting messages. (chomsky.info)

These traditions place different objects at the center of inquiry. Rhetorical analysis examines symbolic appeals; psychological approaches investigate attitudes and responses; Ellul emphasizes social integration; and Herman and Chomsky examine institutional constraints. Computational research extends this interdisciplinary field rather than replacing those questions. The Oxford Internet Institute’s Computational Propaganda project, for example, combines social and computer science to study interactions among automation, algorithms, and human political activity. Its use of computational investigation alongside qualitative inquiry connects network-scale patterns with the organizations and practices producing them. (mediastudies.press)

Research Methods and Standards of Evidence

Archival research reconstructs campaigns by connecting surviving messages to the institutions that commissioned, financed, edited, and distributed them. Provenance analysis establishes an artifact’s creator, date, intended audience, and custodial history rather than treating its imagery or wording as sufficient evidence of sponsorship. The United States National Archives’ Record Group 63, for example, preserves Committee on Public Information correspondence, shipment reports, press releases, posters, and foreign-office records from 1917–1919, allowing comparisons between administrative instructions and published material. Interviews with producers can clarify working routines, while audience testimony documents encounters and interpretations absent from production files. Neither is transparent evidence: recollections may reflect faulty memory or retrospective self-presentation and require comparison with contemporary records. (archives.gov)

Content analysis converts selected features of communication into systematically interpretable evidence. Qualitative readings examine framing, symbolism, omissions, and context; quantitative coding measures their distribution across a defined corpus. Corpus construction specifies dates, outlets, languages, inclusion rules, and units such as articles, images, or individual claims. Multimodal interpretation also considers how captions, visual composition, speech, and sound work together rather than reducing an audiovisual message to its transcript. A coding manual makes category definitions explicit, and independent coding of a shared sample tests reproducibility. Klaus Krippendorff’s alpha measures agreement while accounting for chance disagreement; high agreement nevertheless does not establish that a category validly measures propaganda. Automated text analysis likewise requires validation against substantively informed human judgments, a central principle of Justin Grimmer and Brandon Stewart’s 2013 methodological review. (doi.org)

Multilingual comparison raises an additional measurement problem: identical translated labels need not identify equivalent political meanings. Researchers must distinguish comparisons of word frequencies from interpretations of irony, euphemism, or culturally specific associations. A 2018 study, No Longer Lost in Translation, found substantial correspondence between machine-translated and human-translated texts for comparative bag-of-words applications. That result supports particular computational tasks, not an assumption that translation preserves every rhetorical distinction. Language-specific validation samples and documented translation procedures help establish which comparisons the evidence can sustain. (cambridge.org)

Surveys measure reported exposure, interpretation, and attitudes; experiments use randomized exposure to isolate responses to specified messages. Panel studies repeatedly observe the same people, distinguishing baseline differences from subsequent change, although attrition and inaccurate exposure measurement remain concerns. Gregory Eady and colleagues’ 2023 study linked longitudinal survey responses to Twitter feeds during the 2016 United States election, illustrating how behavioral traces can supplement self-reports. Quasi-experimental designs instead exploit variation not assigned by researchers. David Yanagizawa-Drott’s 2014 study of Rwanda used terrain-induced differences in radio reception to investigate participation in violence. Such causal inference depends on assumptions about whether reception differences are independent of other determinants of the outcome, not merely on a correlation between broadcast availability and behavior. (nature.com)

Network analysis maps relationships among accounts, shared URLs, reposts, and posting times to identify diffusion pathways or candidate coordinated groups. These patterns do not independently establish a common operator or sponsor. Oxford’s Computational Propaganda project combined computational detection with qualitative fieldwork precisely because bot identification and impact estimation present distinct difficulties. Automated classifications also introduce measurement error: a 2020 study of Botometer documented false positives, false negatives, and score variation over time. Consequently, prevalence estimates depend on sampling, classifier validation, and decision thresholds. Platform-access restrictions, deleted accounts, and unavailable private communications further delimit the observable network; estimates of bot activity describe that observation window, while claims about influence require separate evidence connecting activity to human exposure and outcomes. (demtech.oii.ox.ac.uk)

Effectiveness, Reception, and Limits

Campaign expenditure and publication volume measure resources and output; impressions measure opportunities for exposure; engagement records actions such as sharing or commenting; and recall measures memory. None, by itself, establishes agreement or behavioral change. A conspicuous campaign may repeatedly reach existing supporters rather than persuade opponents. Gregory Eady and colleagues’ 2023 study linked longitudinal surveys to respondents’ Twitter timelines during the 2016 United States presidential election. Within their sample, 1% of users accounted for 70% of exposures to posts from Russian Internet Research Agency accounts, and exposure was concentrated among strong Republicans. The researchers found no meaningful relationship between exposure and changes in attitudes, polarization, or voting behavior. This finding concerns measured individual-level outcomes on Twitter, not every possible consequence of the wider interference campaign. (csmapnyu.org)

Reception evidence can reveal interpretations opposite to those intended by sponsors. In September 1939, Britain’s Ministry of Information distributed a poster promising that the public’s courage, cheerfulness, and resolution would bring victory. Mass-Observation’s investigation recorded objections to its abstract vocabulary and to the contrast between “your” sacrifices and “us” receiving victory, which some readers understood in class terms. Remembering the slogan therefore did not necessarily indicate acceptance of its appeal to national unity. Humor likewise complicates any opposition between successful indoctrination and outright resistance. Patrick Merziger’s 2015 research on entertainment in Nazi Germany examined the failure to popularize aggressive anti-Jewish satire and audiences’ preference for other comic forms. Rejection of a particular propagandistic format cannot automatically be equated with rejection of the regime or its broader ideological assumptions. (ww2poster.co.uk)

Reinforcement and selective acceptance also make average effects potentially misleading. Maja Adena and colleagues’ 2015 study of German radio found that the effects of antisemitic broadcasting varied with local predispositions. Using historical pogroms and earlier nationalist voting as indicators, they estimated stronger effects on denunciations and other expressions of antisemitism where existing hostility was greater. For some outcomes, estimated effects were negative in localities with historically low antisemitism. Such geographical comparisons do not directly reveal every listener’s reasoning, but they show why a uniform message need not produce a uniform response: reinforcement, indifference, and resistance can coexist within the same broadcasting system. (micro.econ.kit.edu)

Claims that propaganda determined an election, war outcome, or social transformation face a counterfactual problem: what would have happened without the campaign? Confounding arises when political organization, economic conditions, or prior commitments affect both exposure and outcomes. Self-selection occurs when committed supporters seek congenial messages; reverse causation occurs when existing attitudes shape media consumption rather than simply result from it. The concentration of Russian campaign exposure among already strongly partisan Twitter users illustrates why exposure–attitude correlations cannot establish persuasion. Conversely, failure to detect an effect on specified survey measures does not establish the absence of indirect effects through journalists, interpersonal discussion, or unmeasured outcomes. (csmapnyu.org)

Even stronger causal designs can yield different conclusions about the same campaign. David Yanagizawa-Drott’s 2014 study used terrain-related variation in radio reception and village-level prosecution data to estimate the contribution of RTLM broadcasts to participation in Rwanda’s 1994 genocide. It attributed approximately 10% of participation to broadcasting, including effects transmitted through neighboring communities. A 2024 study by Hollie Nyseth Nzitatira, Trey Billing, and Jared Edgerton instead found that RTLM coverage was associated with an earlier onset of violence, but not higher participation. These findings distinguish the timing, scale, and interpersonal diffusion of violence rather than supporting a single measure of propaganda’s overall effectiveness. (yanagizawadrott.com)

Duration is a separate evidentiary question. Alan Gerber and colleagues’ 2011 study experimentally varied the timing and volume of television advertising during a 2006 gubernatorial campaign involving approximately $2 million in television and radio advertising. Daily polling found substantial but short-lived shifts in candidate preferences. Although political advertising is not automatically propaganda, this experiment demonstrates why immediate responses cannot establish durable conversion. Its results concern a particular electoral audience and competitive media environment; transferring them to sustained state broadcasting, another language, or coercive wartime conditions requires additional evidence rather than assuming equivalent effects. (cambridge.org)

Law, Rights, and Public Accountability

The International Covenant on Civil and Political Rights (ICCPR), adopted in 1966 and in force since 23 March 1976, distinguishes freedom of opinion from freedom of expression. Article 19(1) protects holding opinions without interference; Article 19(2) protects seeking, receiving, and imparting information and ideas across borders and through different media. Restrictions on expression must satisfy Article 19(3): they must be provided by law and necessary to protect others’ rights or reputations, national security, public order, public health, or morals. Article 20 imposes two separate obligations on states parties: to prohibit propaganda for war, and to prohibit advocacy of national, racial, or religious hatred that constitutes incitement to discrimination, hostility, or violence. Neither provision establishes a general international prohibition of political propaganda, partisan advocacy, or government publicity. (2covenants.ohchr.org)

The United Nations Human Rights Committee’s General Comment No. 34 (2011), paragraphs 50–52, explains that Articles 19 and 20 are complementary: restrictions justified under Article 20 must also comply with Article 19(3). Article 20 differs by requiring specified forms of expression to be prohibited by law, not by removing safeguards against excessive restrictions. The Committee requires restrictions to meet tests of necessity and proportionality, and treats freedom of opinion itself as permitting no exception or restriction. Thus, identifying communication as “propaganda” does not dispense with identifying its legally relevant content, the applicable statutory rule, and the justification for restricting freedom of expression. (docstore.ohchr.org)

In the United States, federal election law uses sponsorship disclosure and financing restrictions rather than a general statutory category of prohibited propaganda. Under 52 U.S.C. §30120 and 11 C.F.R. §110.11, covered campaign communications must identify their payer and, where applicable, whether a candidate authorized them. The internet-disclaimer rules adopted in December 2022 specifically addressed paid online advertisements, including requirements for readable written notices and video notices visible for at least four seconds. Separately, 52 U.S.C. §30121, as reproduced in the 2023 United States Code, prohibits foreign nationals from making election-related contributions and specified expenditures, and prohibits soliciting, accepting, or receiving their contributions. Its definition excludes lawful permanent residents from the individual foreign-national category; the restriction is therefore not a blanket prohibition on participation by every noncitizen. (uscode.house.gov)

The European Union’s Regulation (EU) 2024/900, adopted on 13 March 2024, establishes a different framework for transparency and targeting of political advertising, with most provisions applying from 10 October 2025. Articles 11–12 require political advertisements to carry identifying labels and provide access to transparency notices covering sponsorship, relevant elections or referendums, financing, and targeting information. Article 5(2) limits eligible sponsors during the three months preceding an election or referendum: qualifying categories include EU citizens, certain permanently resident non-EU nationals with voting rights, and EU-established legal persons not ultimately owned or controlled by specified third-country persons or entities. These rules concern the provision and traceability of advertising services, rather than declaring political persuasion itself unlawful. (eur-lex.europa.eu)

Public expenditure is subject to distinct controls. In Great Britain, section 2 of the Local Government Act 1986 prohibits local authorities from publishing, or arranging publication of, material appearing designed to affect support for a political party. The assessment considers content, style, timing, surrounding circumstances, likely audience effects, and the apparent purpose of a wider campaign. Section 2(3) also prevents an authority from financing or otherwise assisting another person to publish material that the authority could not lawfully publish itself. Unlike an election-advertising disclaimer, this is a restriction on institutional publication and assistance: identifying the public sponsor does not make otherwise prohibited party-political publicity permissible. (legislation.gov.uk)

Counterpropaganda, Verification, and Media Literacy

Counterpropaganda is communication intended to counter the influence of an opposing propaganda campaign. Its methods address different objects: rebuttal contests particular claims; exposure identifies concealed sponsors, fabricated identities, or coordinated distribution; and alternative narratives supply a competing account rather than answering every allegation separately. Restrictions on distribution—such as removing accounts or blocking broadcasts—operate on access and circulation rather than directly establishing whether claims are true. These activities need not occur together. NATO’s approach published in February 2025, for example, distinguishes proactive communication, monitoring, debunking, and exposure, while acknowledging that directly engaging a provocative message can amplify it. Such institutional statements describe response strategies and objectives, not independent evidence that those strategies succeed. (en.wikipedia.org)

Professional verification establishes what evidence supports a claim before deciding how to characterize it. In journalism, this includes locating the original photograph, recording, or document; comparing independent reporting; and checking whether an image actually depicts the asserted event. The Associated Press’s 2023 standards on generated content specify original-source identification, reverse-image searching, and comparison with trusted reporting. Its verification operations also use geolocation and frame-by-frame video examination. Archival authentication adds examination of handwriting, paper, original negatives, and records of an item’s origin and custody. The United States National Archives identifies disputed authorship, date, and authenticity as reasons to preserve originals for physical examination. These procedures distinguish an authentic artifact from an accurate account: a genuine campaign document establishes what its producer communicated, not necessarily the truth of its assertions. (ap.org)

UNESCO treats media and information literacy as broader than detecting false statements. Its curriculum combines information, media, and digital competencies, encompassing critical evaluation, understanding how media institutions and technologies operate, and participation in producing and communicating information. UNESCO issued its first model curriculum in 2011 and a second edition, Media and Information Literate Citizens: Think Critically, Click Wisely!, in 2021. The revision addresses changes including artificial intelligence, digital citizenship, disinformation, and online hate speech, with provision for formal, non-formal, and informal learning. Its educational objectives concern critical engagement and participation; publication of a curriculum or adoption by an institution does not itself measure improvements in audience judgment. (unesco.org)

Research on fact-checking evaluates narrower outcomes. In a 2021 study, Ethan Porter and Thomas J. Wood reported simultaneous experiments conducted in September–October 2020 in Argentina, Nigeria, South Africa, and the United Kingdom, testing 22 distinct fact-checks through 28 experiments. Corrections reduced belief in false claims by an average of 0.59 points on a five-point scale. Most effects remained detectable more than two weeks later in the three countries with follow-up surveys. No tested correction produced an identified backfire effect, meaning increased belief in the corrected falsehood. However, participants were assigned exposure rather than choosing what to read, and the study did not establish corresponding changes in political attitudes or behavior. The authors explicitly distinguish improved factual accuracy from a comprehensive solution to misinformation. (doi.org)

Prebunking intervenes before subsequent exposure, often by explaining manipulation techniques and presenting weakened examples that recipients can learn to recognize. Drawing on inoculation theory, Jon Roozenbeek and colleagues tested short videos about techniques including emotionally manipulative language and false dichotomies. Their 2022 study comprised six preregistered randomized experiments with 6,464 participants and a YouTube field study with 22,632 respondents. Reported outcomes included better technique recognition, greater discernment between trustworthy and untrustworthy content, and improved sharing judgments. These measures were not equivalent to observed long-term sharing behavior. The field experiment assessed recognition within a 24-hour window, with a median delay of 18.4 hours; it therefore did not establish durable resistance over months or across repeated campaigns. (pubmed.ncbi.nlm.nih.gov)

Ethical Questions and Unresolved Analytical Problems

Truthfulness and respect for audience autonomy are distinct questions in the ethics of propaganda. A message may contain accurate statements while using selective presentation, emotional pressure, or concealed motives to impair informed judgment. Conversely, an unsuccessful attempt at manipulation may remain ethically objectionable even when it produces no measurable harm. Philosophical accounts disagree about whether manipulation principally involves deception, pressure, bypassing rational deliberation, or treating people as instruments. These differences matter for evaluating persuasion undertaken for beneficial purposes: an improvement in the recipient’s circumstances does not, by itself, establish that the means respected the recipient’s agency. (plato.stanford.edu)

Concealed sponsorship raises a separate problem because audiences use information about a speaker’s interests when assessing credibility. A commercial example illustrates the mechanism without making advertising and propaganda interchangeable. In its 2016 Lord & Taylor proceeding, the United States Federal Trade Commission alleged that the retailer had given dresses to 50 Instagram influencers and paid each $1,000–$4,000 for posts during March 27–28, 2015, without disclosure of compensation. The company also approved the posts before publication. The ethical issue extended beyond whether the photographs accurately depicted the product: paid, supervised communication appeared to be independent endorsement. (ftc.gov)

Unequal access to communication concerns the conditions under which public judgment forms, rather than simply the accuracy of individual messages. In How Propaganda Works (2015), Jason Stanley connects propaganda with inequalities that obstruct democratic deliberation and permit democratic vocabulary to disguise undemocratic arrangements; this is a particular philosophical account, not a universally accepted definition. (assets.press.princeton.edu) Targeting vulnerable audiences introduces another dimension. UNICEF’s work on children and digital marketing distinguishes harmful content from harmful context and design: undisclosed influencer promotion can obscure commercial intent, while reward-based advertising and targeting mechanisms can exploit children’s developing cognitive capacities. Such distinctions locate ethical responsibility in distribution and interface design as well as wording. (unicef.org)

Public-interest campaigns, patriotic commemoration, and civic education create classification difficulties because socially approved objectives do not determine how communication operates. An analytical comparison can distinguish explanation that permits questioning from instruction designed to secure a predetermined response, while examining disclosure, omitted alternatives, and institutional control. In discussing educational examples, Garth Jowett and Victoria O’Donnell recount Elliot Aronson’s 1980 observation that arithmetic problems involving wages, purchases, and interest may also normalize an economic order. This does not establish that every culturally situated textbook is propaganda; it exposes disagreement about where instruction ends and ideological reproduction begins. Patriotic rituals similarly require attention to sponsorship and intended integration, rather than classification by national symbolism alone. (csmeyns.github.io)

The word “propaganda” can itself delegitimize an opponent by treating disputed conclusions as evidence of manipulation. Consistent analysis therefore separates disagreement with a viewpoint from evidence about its production and presentation. Philosophical discussion of manipulation explicitly cautions against inferring impaired autonomy merely because people make choices an observer considers mistaken. (plato.stanford.edu) The same criteria—source disclosure, selective presentation, control over circulation, and opportunities for contestation—can be applied across political systems without assuming that their institutions are equivalent. Cross-cultural comparison also raises the question of whether definitions identify comparable practices or import the evaluator’s assumptions about legitimate persuasion; Jowett and O’Donnell’s distinction between neutral and pejorative usage makes this evaluative problem explicit. (csmeyns.github.io)

Intent becomes particularly difficult to assign when production is decentralized. Kate Starbird, Ahmer Arif, and Tom Wilson’s 2019 study of three online information operations describes interactions between organized operators and crowds whose participants may neither share the organizers’ objectives nor recognize their involvement. Voluntary participation and organized manipulation can therefore coexist within one circulation network; sharing a message does not establish knowledge of its origins. Their analysis also shifts attention from isolated statements to campaigns and information environments. Unresolved questions concern whose intentions define a distributed campaign, how responsibility follows successive adaptations, and how cumulative changes in perceived consensus can be distinguished from independently emerging agreement. (faculty.washington.edu)

References

1. Propaganda & Persuasion, Sixth Editioncsmeyns.github.io
2. Congregationarchiviostoricopropaganda.va
3. Press Conference to present the International Study Conference “Euntes in mundum universum”press.vatican.va
4. The Globalization of Chinese Propagandaperlego.com
5. Long Live Chairman Mao: Propaganda About Mao Zedong in Chinese Primary School Textbooks (1984–1999)ijoc.org
6. PRSA Milestonesprsa.org
7. About Us – Under Secretary for Public Diplomacy and Public Affairs2021-2025.state.gov
8. Video News Releases: Unattributed Prepackaged News Stories Violate Publicity or Propaganda Prohibitiongao.gov
9. GAO-05-643T: Video News Releasesgao.gov
10. Historical Documents—Foreign Relations of the United States, 1951, Volume I, Document 333history.state.gov
11. Propaganda: The Formation of Men’s Attitudestheanarchistlibrary.org
12. Propaganda: The Formation of Men's Attitudes—Google Booksbooks.google.co.uk
13. Coin Judaea Capta Coin of Vespasiancollections.thejewishmuseum.org
14. Add MS 33241 — Encomium Emmae reginaesearcharchives.bl.uk
15. Defender of the Faithbl.uk
16. The New Testament of Jesus Christ. 1582.catalog.folger.edu
17. Shakespeare, Sidney and Spenser in an Early Continental Libraryapi.repository.cam.ac.uk
18. Wellington House and British Propaganda during the First World Warssoar.info
19. British wartime propaganda and publishingfiles.eric.ed.gov
20. Records of the Committee on Public Informationarchives.gov
21. The Four Minute Men of Chicagolibsysdigi.library.illinois.edu
22. Remember Belgium. Enlist To-day, 1914collection.nam.ac.uk
23. Archivio Centrale dello Stato — Ministero della cultura popolare (1926–1945)search.acs.beniculturali.it
24. Il fondo archivistico Ufficio Censura Teatrale (1931–1944)movio.beniculturali.it
25. Joseph Goebbelsencyclopedia.ushmm.org
26. Ministry of Propaganda and Public Enlightenmentencyclopedia.ushmm.org
27. Nazi Propagandaencyclopedia.ushmm.org
28. Internal Workings of the Soviet Union — Revelations from the Russian Archivesloc.gov
29. About Shashin Shuhojacar.go.jp
30. A Window into the Early Showa Period — Shashin Shuhojacar.go.jp
31. A Reel Story of World War IIarchives.gov
32. Propaganda — The National Archivesnationalarchives.gov.uk
33. Attention! Moscow Calling: BBC Monitoring and the Cuban Missile Crisisiwm.org.uk
34. Radio Free Europe and Radio Libertywilsoncenter.org
35. RADIO FREE EUROPE AND RADIO LIBERTYcia.gov
36. Cultural Diplomacy — Hope for America: Performers, Politics and Pop Culturewwws.loc.gov
37. Гагарин в плакатном фонде ГА РФstatearchive.ru
38. The Sleeping Giant: Posters & The Chinese Economy — Exhibition Archiveposterhouse.org
39. Solidarity and design: An introduction to OSPAAALvam.ac.uk
40. Sejarah Penyiaran Radio di Indonesiampn.komdigi.go.id
41. Radio Republik Indonesia Surakarta, 1945–1960s: Its Role in Efforts to Maintain Indonesian Independence and the Formation of National Cultureejournal2.undip.ac.id
42. Radio Tanzania Archivestanzaniaheritageproject.org
43. Psychological Operations Branchapi.army.mil
44. The Press in the Third Reichencyclopedia.ushmm.org
45. Polyglot Typographyarchiviostoricopropaganda.va
46. Industry Documents Libraryindustrydocuments.ucsf.edu
47. Four-Minute Men Records, 1914–1918oac.cdlib.org
48. Historical Documents: Meeting to Discuss the Crusade for Freedomhistory.state.gov
49. Historical Documents: Memorandum for the 303 Committee, January 27, 1969history.state.gov
50. A Propaganda Model, by Edward Herman and Noam Chomskychomsky.info
51. How We Respond to Inauthentic Behavior on Our Platforms: Policy Updateabout.fb.com
52. Common Persuasive Techniques Used in Propagandad17j94wz7065tl.cloudfront.net
53. Propaganda and Mediaushmm.org
54. SemEval-2020 Task 11: Detection of Propaganda Techniques in News Articlesarxiv.org
55. The Many Emotions of Purchasing a Liberty Bond: Understanding Persuasive Postersblogs.loc.gov
56. Deceiving the Publicencyclopedia.ushmm.org
57. Brochure advertising photo books about Hitlerencyclopedia.ushmm.org
58. Defining the Enemyencyclopedia.ushmm.org
59. Persuasive Mapspersuasivemaps.library.cornell.edu
60. Personal involvement as a determinant of argument-based persuasion.doi.org
61. Knowledge does not protect against illusory truthpubmed.ncbi.nlm.nih.gov
62. The case for motivated reasoningpubmed.ncbi.nlm.nih.gov
63. Opinions and Social Pressureweb.mit.edu
64. The Great War: U.S. Army Arthistory.army.mil
65. Writing the Newsencyclopedia.ushmm.org
66. The Centenary of the Sommeassets.publishing.service.gov.uk
67. The Living Room Candidate — Peace Little Girl (Daisy)livingroomcandidate.org
68. Uncle Sam-I-Am: Dr. Seuss’s Private Snafuunwritten-record.blogs.archives.gov
69. Documentation Center Nazi Party Rally Groundsmuseums.nuernberg.de
70. The 1937 Nuremberg Rallyresearch.calvin.edu
71. “Degenerate” Artencyclopedia.ushmm.org
72. About the War Memorials Registermemorials.iwm.org.uk
73. Busheymemorials.iwm.org.uk
74. What is Computational Propaganda?navigator.oii.ox.ac.uk
75. 2020 Global Inventory of Organized Social Media Manipulationdemtech.oii.ox.ac.uk
76. Challenging Truth and Trust: A Global Inventory of Organized Social Media Manipulationdemtech.oii.ox.ac.uk
77. Removing Coordinated Inauthentic Behaviorabout.fb.com
78. Introducing 22 system cards that explain how AI powers experiences on Facebook and Instagramai.meta.com
79. How do social media feed algorithms affect attitudes and behavior in an election campaign?networkscienceinstitute.org
80. March 2021 Coordinated Inauthentic Behavior Reportabout.fb.com
81. Federal Communications Commission DA 24-102docs.fcc.gov
82. Public opiniongutenberg.org
83. Propaganda Technique In World War Imitpress.mit.edu
84. Propagandagutenberg.org
85. Featured Books by Jacques Ellulellul.org
86. Conclusionschomsky.info
87. Computational Propagandaoii.ox.ac.uk
88. U.S. Army Guide to Oral Historygovinfo.gov
89. Text as Data: The Promise and Pitfalls of Automatic Content Analysis Methods for Political Textsdoi.org
90. No Longer Lost in Translation: Evidence that Google Translate Works for Comparative Bag-of-Words Text Applicationscambridge.org
91. Exposure to the Russian Internet Research Agency foreign influence campaign on Twitter in the 2016 US election and its relationship to attitudes and voting behaviornature.com
92. DemTech: Computational Propagandademtech.oii.ox.ac.uk
93. The False positive problem of automatic bot detection in social science researchpubmed.ncbi.nlm.nih.gov
94. Exposure to the Russian Internet Research Agency Foreign Influence Campaign on Twitter in the 2016 US Election and Its Relationship to Attitudes and Voting Behaviorcsmapnyu.org
95. University of Birmingham: Spencer_Bennett3.pdfpure-oai.bham.ac.uk
96. ‘Totalitarian Humour’? National Socialist Propaganda and Active Audiences in Entertainmentopenurl.ebsco.com
97. Radio and the Rise of the Nazis in Prewar Germanymicro.econ.kit.edu
98. Propaganda and Conflict: Evidence from the Rwandan Genocideyanagizawadrott.com
99. How Large and Long-lasting Are the Persuasive Effects of Televised Campaign Ads? Results from a Randomized Field Experimentcambridge.org
100. International Covenant on Civil and Political Rightsohchr.org
101. UN Human Rights: International Covenant on Civil and Political Rights2covenants.ohchr.org
102. Human Rights Committee, General Comment No. 34docstore.ohchr.org
103. Internet Communication Disclaimers and Definition of “Public Communication”fec.gov
104. 52 U.S.C. §30121: Contributions and donations by foreign nationalsgovinfo.gov
105. Regulation (EU) 2024/900 on the transparency and targeting of political advertisingeur-lex.europa.eu
106. Local Government Act 1986, section 2legislation.gov.uk
107. Counterpropagandaen.wikipedia.org
108. NATO’s approach to counter information threatsnato.int
109. Standards around generative AIap.org
110. Verification at APap.org
111. AP introduces AP Verify to strengthen, streamline online content verificationap.org
112. Intrinsic Value in Archival Materialarchives.gov
113. Media and Information Literate Citizens: Think Critically, Click Wisely!unesco.org
114. Media and information literate citizensunesco.org
115. Psychological inoculation improves resilience against misinformation on social mediaaspredicted.org
116. The Ethics of Manipulationplato.stanford.edu
117. Complaint: Lord & Taylor, LLCftc.gov
118. Princeton University Press Philosophy Catalog: How Propaganda Worksassets.press.princeton.edu
119. How Propaganda Works — Introductionassets.press.princeton.edu
120. Impact of Digital Marketing on Children: Understanding Harmful Content, Context and Designunicef.org
121. Disinformation as Collaborative Work: Surfacing the Participatory Nature of Strategic Information Operationsfaculty.washington.edu