The Propaganda Model
Who shapes the news before it reaches us?
Follow the relationships among media, money, sources and power. Learn to explain the five filters, examine their interaction and test a claim without assuming its conclusion.
In this chapter · Open the contents
Read the headline. Investigate the system behind it.
- Explain the model’s structural argument and five filters.
- Explain how ideology and enemy framing can narrow debate.
- Distinguish source dependence, flak, factual accuracy and audience effects.
- Compare coverage and consider evidence that challenges your explanation.
- Assess the model’s strengths, criticisms and limits across settings.
Look behind the headline. Follow the pressures that shape it.
Two newsrooms receive the same account of a factory accident. One has a reporter on the ground, another relies on an official briefing. One can fund a week of investigation; another must publish within an hour. The resulting stories may differ before anyone issues an order about what to write.
The Propaganda Model examines how the organisation and financing of media can shape patterns of news selection, emphasis and exclusion. Edward S. Herman and Noam Chomsky developed it in Manufacturing Consent: The Political Economy of the Mass Media (1988). Their central setting was the dominant, commercially organised news media of the United States.
The model argues that unequal resources and institutional relationships can favour powerful interests while narrowing the space available to sustained challenges. It asks how this can happen even where news organisations compete, journalists value professional independence and direct state censorship is absent.
The shift: from individual bias to institutional conditions
An individual-bias explanation asks whether a reporter dislikes a politician or supports a cause. A structural explanation asks who owns the outlet, how it earns money, which sources it can readily use, what pressures follow publication and which assumptions organise debate.
Think of a newsroom as a workshop. Its workers make choices, but the tools, budget, deadlines, contracts and rewards shape what they can repeatedly produce. The analogy helps locate constraints without erasing professional judgment.
Which institutional pressures favour which kinds of coverage, through what mechanisms, and with what evidence? A suspicion about influence starts an inquiry; it does not establish the explanation.
Three claims the model does not establish by itself
- That every journalist knowingly follows a coordinated plan.
- That every published statement is false, or every challenging report is excluded.
- That audiences believe everything they encounter.
Herman’s later clarification distinguishes a model of media performance from a theory of guaranteed audience effects. News patterns, journalists’ intentions and audience responses require different evidence.
Foundations: Herman and Chomsky (1988); Herman (2000). Links and further readings appear at the end of the chapter. Contemporary cases below are fictional teaching scenarios.
Propaganda involves more than an obvious lie.
Propaganda scholarship studies organised efforts to shape perceptions, judgments and conduct in the service of a sponsor’s aims. Such communication can use accurate facts, selective emphasis, emotional symbols, misleading claims or fabricated material. Truthfulness matters, but checking individual facts does not exhaust the analysis.
For example, a campaign may present verified achievements while suppressing relevant costs. Another may conceal its sponsor. A third may spread a false allegation. These are different practices and should not be collapsed into one label.
White, grey and black propaganda
This classification concerns source identification as well as the accuracy and presentation of information. It is separate from Herman and Chomsky’s five institutional filters.
| Type | Main distinction | Teaching example |
|---|---|---|
| White propaganda | The source is correctly identified; information tends to be accurate but is selected and presented to advance the sponsor’s purpose. | An openly sponsored campaign highlights verified successes while offering a one-sided account. |
| Grey propaganda | Source attribution and/or the accuracy of the information remain uncertain. | An unattributed message makes claims whose origin and reliability cannot yet be established. |
| Black propaganda | The communication deceives about its origin or claims a false authority, often using fabrication and other forms of deceit. | A campaign circulates a forged statement that appears to come from its opponent. |
An anonymous message is not automatically black propaganda. An openly identified source is not automatically balanced. Examine origin, purpose, accuracy and presentation together.
From wartime campaigns to a political economy of news
World War I intensified organised propaganda and later scholarly inquiry. Harold D. Lasswell’s Propaganda Technique in the World War appeared in 1927. Herman and Chomsky’s 1988 model asks a different, more specific question: how can the routine operation of privately organised news media favour powerful interests?
Keep the levels distinct. Studying a deliberately deceptive campaign is not the same task as explaining a recurring pattern of news production. The two may overlap, but one does not prove the other.
For propaganda concepts and forms, see Snow, Jowett and O’Donnell (2025), Propaganda & Persuasion, 8th edition. Their framework is distinct from the Herman–Chomsky model.
Five questions about the conditions of news.
The filters describe pressures and relationships, not five people checking a story in a fixed order. They can operate before a reporter begins, during routine production and after publication.
Five interacting filters · Questions behind news production
Who owns, governs and finances the outlet?
Capital, corporate interests and management priorities.
Which audiences and environments attract revenue?
Dependence on advertisers and commercially desirable attention.
Who can supply regular, usable and credible-looking material?
Unequal access to information and the cost of verification.
Who can impose costs after unwelcome coverage?
Complaints, threats and pressure that may discipline reporting.
Which assumptions define acceptable debate?
Shared premises and enemy narratives can narrow the space for dissent.
The fifth filter here focuses on dominant ideology and enemy framing. This broader formulation draws on later discussion of ideological change. The original 1988 formulation reflected the Cold War setting; applications elsewhere must identify the assumptions and pressures operating in that setting.
Use the map to generate questions, then look for mechanisms and evidence. Knowing that an outlet carries advertising, for example, does not establish that a particular advertiser altered a particular story.
Trace the mechanism before naming the cause.
Filter 1Ownership, scale and profit orientation
The first filter concerns the resources needed to enter and sustain a large media operation, its ownership and governance, and the demands placed on it as a business. An owner can influence the organisation through senior appointments, investment priorities, risk tolerance and broad editorial direction without rewriting every sentence.
Corporate connections may create conflicts between public-interest reporting and other business interests. But ownership is not a complete explanation of every item. Different owners, business models, professional cultures and institutional safeguards may produce different outcomes.
Look for: ownership and governance records, documented commercial interests, appointment patterns, editorial policies and accounts of relevant decisions. Compare the outlet’s treatment of similar matters involving unrelated firms.
Filter 2Advertising and the market for audience attention
In an advertising-supported operation, revenue depends partly on delivering audiences that advertisers want to reach. This can affect which audiences receive investment and which environments are considered commercially attractive. Pressure may operate through revenue expectations and anticipated objections as well as direct intervention.
Advertising does not mean that all journalism is purchased by sponsors. The question is how dependence, concentration of revenue and the ability to withdraw funding may constrain decisions. Subscription, membership, public-service and philanthropic models alter these relationships; each also requires its own scrutiny.
Look for: the revenue model, exposure to particular advertisers where records are available, sponsorship terms, editorial safeguards and documented pressure. Distinguish labelled advertising from news content.
Filter 3Sourcing and the economics of gathering news
Newsrooms need a continuing flow of material. Governments, corporations and well-resourced organisations can supply briefings, data, spokespersons and ready-made accounts on schedule. Their regular availability can make them easier to use than people who lack press offices, transport, time or recognised credentials.
This is a question of unequal access and routine dependence. It does not mean that an official statement is always false or that a non-official source is always reliable. The concern is whether some accounts receive an unearned advantage in defining the issue and setting the terms of verification.
A source network can become self-reinforcing: regular access helps meet deadlines, repeated use builds familiarity, and familiarity can be mistaken for sufficient evidence.
Look for: who speaks first, who is quoted directly, whose claims are checked, who is absent, and whether affected people can challenge an official account. A source count is a start; the authority each source receives also matters.
Filter 4Flak and the power to make reporting costly
Flak refers to negative responses to media output: complaints, organised campaigns, threats, legal pressure or appeals to advertisers and managers. Such responses can discipline future coverage when they impose substantial costs or make journalists anticipate punishment.
Distinguish pressure from accountability. A well-supported correction request can improve a report. A forceful complaint is not proof that journalism should be shielded from criticism. Analyse the claims, resources, tactics, proportionality and consequences of the response.
Look for: the complaint’s evidential basis, coordination, resource backing, threats or sanctions, and subsequent editorial decisions. Do not assume that every large online disagreement is an organised operation.
Filter 5 · Contemporary adaptationDominant ideology and enemy framing
This filter directs attention to assumptions that make some positions appear reasonable and others disloyal, dangerous or beyond discussion. An enemy frame can compress a complex dispute into a division between “us” and “them”, making criticism easier to dismiss without addressing its evidence.
Study the boundary of debate. A panel may vigorously argue about the most efficient way to pursue a policy while leaving its purpose unquestioned. Conversely, shared assumptions may be challenged when political groups, journalists or public movements open new questions.
Do not treat every warning about a threat as propaganda. A threat may be real. The inquiry asks whether evidence is assessed consistently, which voices receive a hearing, and whether a label displaces examination of the claim.
Look for: recurring labels, taken-for-granted premises, acceptable expert roles and the treatment of dissent. If applying this adapted filter in a new setting, define the relevant ideology and explain how it could shape selection or interpretation.
The institutional filters are developed in the supplied lecture and Herman and Chomsky’s work. The broader fifth-filter treatment follows later discussion, including their 2009 interview with Andrew Mullen. Examples and suggested indicators here are teaching applications.
A pattern can emerge without a single command.
Consider a fictional newsroom with limited reporting funds. It draws heavily on a regular official briefing. A major advertiser dislikes prolonged controversy, while earlier complaint campaigns have made editors wary of the topic. These pressures can converge even if each actor acts separately.
Ownership, revenue dependence, source access, potential sanctions and shared assumptions.
Time, staffing, source selection, verification, prominence and follow-up.
Recurring differences in attention, authority, context and criticism.
Where does professional judgment fit?
Journalists may verify claims, resist pressure, widen sources and publish investigations. The model’s structural focus asks what makes those practices easier or harder to sustain. A sincere commitment to objectivity can coexist with dependence on a narrow set of routine sources.
The presence of critical journalism therefore matters. It should be examined, not dismissed as a convenient exception. Ask whether criticism is sustained, what it challenges, how prominently it appears and which institutional conditions permit it.
Do not count favourable coverage as proof of control and every challenging report as proof that control is well disguised. State in advance what results would weaken the proposed explanation.
Whose suffering receives a name, a face and a follow-up?
Herman and Chomsky use the deliberately ironic terms worthy and unworthy victims to describe unequal media treatment. The labels concern the political usefulness assigned to victims’ stories; they do not describe the worth of human beings.
The argument proposes that suffering connected with an adversary may attract stronger condemnation and fuller humanisation than comparable suffering connected with an ally or favoured interest. Testing that argument requires carefully chosen comparisons.
Case A · Adversary-linked event
Measure prominence, named victims, personal accounts, images, explanations and demands for responsibility.
Case B · Ally-linked event
Use the same coding rules, outlet sample and observation period. Record differences and similarities.
Count more than the number of stories
Two events can receive similar amounts of coverage but different treatment. One report may name victims and explain their lives; another may present them as an anonymous total. One may trace responsibility; another may describe harm as an unexplained tragedy.
Code prominence, source authority, agency, humanisation, historical context and follow-up as separate dimensions. Keep the coding rules consistent, including when the result challenges your expectation.
A comparison trap to avoid
Suppose Case A produces extensive verified footage while Case B remains inaccessible. A difference in coverage could partly reflect evidence availability. That does not settle the wider political question, but it must be considered. A stronger design compares several events, documents access constraints and tests whether the proposed pattern survives those differences.
Adapt the questions. Do not merely rename the filters.
Digital communication adds relationships among publishers, platforms, creators, advertisers, hosting services and audiences. A website can publish material that receives little visibility elsewhere. A creator can distribute independently while depending on platform access or income.
Separate production, distribution and visibility. The original model is not a technical description of recommendation algorithms. Applying it to platform systems requires evidence about the additional mechanisms involved.
| Relationship | Investigate | Do not assume |
|---|---|---|
| Ownership and infrastructure | Who controls publishing, hosting, distribution or access to an audience? | That low publishing costs remove every barrier to reach. |
| Revenue and monetisation | Which dependencies arise from advertising, sponsorship, subscriptions or platform income? | That every creator or publisher uses the same business model. |
| Source circulation | How do briefings, public relations material, posts and verified reporting move across outlets? | That repetition provides independent confirmation. |
| Pressure and sanctions | Can organised complaints, access restrictions or commercial threats change decisions? | That every moderation decision or complaint is evidence of coordinated flak. |
| Ideology and framing | Which assumptions and enemy labels shape acceptable positions in the setting studied? | That the adapted fifth filter has one universal form. |
Applying the model beyond its original setting
Specify how the media system differs from the original US commercial setting. Public funding, party ownership, regional-language markets, regulatory relationships, subscriptions and different source networks may alter the pressures. These features should be investigated rather than forced into an unchanged template.
For a study in India, for example, choose an outlet type, language market, issue and period. Map its actual ownership, revenue, source access and documented pressures. The framework supplies questions; it does not supply findings about an entire country in advance.
For debates on digital and international applications, see Pedro-Carañana, Broudy and Klaehn’s edited collection (2018). Its chapters offer extensions and disagreements rather than a single settled revision.
Different theories ask different questions about the same story.
| Approach | Primary question | Relation to the Propaganda Model |
|---|---|---|
| Gatekeeping | How are potential items selected, shaped and circulated? | The Propaganda Model offers a particular account of structural pressures on selection and treatment. |
| Framing | How does coverage organise an issue, explain causes and make responses meaningful? | Frames can be studied as content patterns; a political-economic explanation of them needs additional evidence. |
| Agenda-setting | Which issues become prominent, and how does prominence relate to public priorities? | Selection and prominence may overlap with the model’s concerns, but audience-priority effects require separate research. |
| Functional analysis | What consequences do media practices have for specified units? | The Propaganda Model emphasises power and institutional alignment rather than a general list of social functions. |
| Critical political economy | How do ownership, resources and power organise cultural production and circulation? | This is the wider family of inquiry within which the model is commonly placed. |
A dramatic headline is not sufficient evidence of any one filter. Neither is agreement among outlets: they may share a wire report, verified facts, deadlines, sources or institutional pressures. Distinguish these possibilities.
Continue with News Framing and the Frankfurt School to compare approaches to meaning, culture and power.
Identify the pressure. Keep the alternative explanation in view.
Select a case, write a short diagnosis and then reveal the analysis. All organisations and events are invented. Each answer identifies an inquiry, not an established finding.
The native radio selector supports keyboard navigation: Tab to the group, then use the arrow keys. Selections and answers are not submitted or saved.
Make the argument answerable to evidence.
A useful study links institutional conditions, production mechanisms and content patterns. No single measure establishes the entire chain. Ownership records do not reveal every editorial decision; a content imbalance does not reveal its cause.
- Define the setting and prediction.
Choose outlets, an issue, a period and a reasoned comparison. State what coverage pattern the proposed filter mechanism predicts. Define the institutional interests independently of the coverage you plan to explain.
- Operationalise the concepts.
Decide how to code prominence, source roles, verification, responsibility, humanisation and follow-up. Explain what counts as an instance of pressure or an ideological boundary. Pilot the coding rules and compare coders’ judgments.
- Collect different kinds of evidence.
Combine content analysis with available ownership and revenue records, editorial documents, interviews and publication timelines. Protect confidential participants and document access limits. Do not infer private motives from a headline alone.
- Test competing explanations.
Examine event scale, timeliness, access, verification, audience relevance, professional routines and resource constraints. These may interact with institutional pressures; they are not automatically irrelevant alternatives.
- State what would weaken the claim.
Specify the expected relationship before reading all results. A repeated absence of the predicted pattern, or decisions better explained by documented alternatives, should lower confidence in that application. Record counterexamples and revise the explanation.
Separate three kinds of conclusion
- Descriptive
- “These reports quote official sources more frequently than affected residents.”
- Explanatory
- “Evidence connects that imbalance with routine dependence on institutional briefings.”
- Audience effects
- “Exposure changes what audiences believe or do.” This requires reception or effects evidence beyond the content pattern.
What the model offers—and what remains disputed
| Issue | Contribution | Question to press |
|---|---|---|
| Institutional power | Directs attention beyond individual prejudice to resources and organisational dependence. | How strongly can the proposed mechanism be documented in the selected setting? |
| Pattern and comparison | Encourages systematic comparison of attention and treatment. | Were cases and coding rules chosen fairly, including inconvenient cases? |
| Elite differences | Raises questions about shared interests and limits of debate. | How do conflicts among owners, political actors and other powerful groups change coverage? |
| Journalistic autonomy | Shows why professional commitments operate within material conditions. | What explains resistance, investigative success and meaningful variation among outlets? |
| Scope and adaptation | Supplies questions that can travel across media systems. | Which relationships need revising when financing, governance or distribution differs? |
| Testability | Can generate specific predictions about news performance. | Which observations would count against the application, rather than being explained away? |
Herman (2000) defends the model and clarifies its scope. Sparks (2007) argues for refinement that takes greater account of differences within powerful groups, journalism and audiences. Corner (2003) provides a critical response in the wider debate. Read these positions alongside the original argument.
Alternatives and safeguards are research questions too
Independent reporting funds, revenue diversity, source diversity, transparent corrections and editorial safeguards may reduce particular dependencies. Alternative media can widen debate, but the label does not establish accuracy or independence. Study their funding, methods and accountability as carefully as those of larger outlets.
Peace journalism, mentioned at the close of the lecture, offers a distinct approach to reporting conflict. It is not a sixth filter or an automatic remedy for every structural pressure.
“The findings support a sourcing-based explanation in this setting, while leaving ownership influence unresolved” is stronger than claiming that every filter has been proved. Match the conclusion to the evidence.
Use the lecture to build and test your own explanation.
Watch Dr. USP’s lecture on the Propaganda Model, then revisit the cases and research questions. The chapter develops the discussion for self-study and uses the broader fifth-filter formulation explained above.
| Start | Focus | Pause and ask |
|---|---|---|
| 00:00 | The wider concept of propaganda. | Why does fact-checking not settle every question about propaganda? |
| 01:09 | Forms of propaganda and wartime context. | How do source attribution and accuracy differ? |
| 01:55 | Transition towards Herman and Chomsky’s model. | What changes when the unit of analysis becomes the media institution? |
| 02:41 | News filters and their interaction. | Why should I avoid treating the filters as isolated boxes? |
| 04:18 | Professional practice, ownership and advertising. | How might pressure operate without direct rewriting of a report? |
| 05:59 | Source dependence and subsequent discussion of flak. | What makes a source easy to use, and what makes criticism costly? |
| 08:24 | Unequal coverage and alternatives. | What would make a comparison fair? |
Entry points are approximate guides derived from the supplied transcript, not official chapter markers. They may open in YouTube. The terminology of the fifth filter has been adapted for this chapter.
Map it. Apply it. Challenge it.
First pass: name the five pressures. Second pass: explain one mechanism through an example. Third pass: write what evidence could show that your explanation is incomplete or wrong.
Explain the pressure. Defend the inference.
Choose and justify your answer before opening the explanation. No responses are collected or automatically scored.
1. What is the Propaganda Model principally designed to explain?
- Recurring patterns of media performance in relation to institutional power
- Why every person believes every headline
- The technical operation of all recommendation algorithms
- A universal rule that every journalist receives secret instructions
Reveal answer 1
Answer: A. The model concerns news production and performance. Intentions and audience effects require additional evidence.
2. A news outlet depends on one advertiser. What does that fact establish?
- That every report is false
- A possible structural dependence to investigate
- That the advertiser wrote its headlines
- That the advertising filter cannot operate
Reveal answer 2
Answer: B. Dependence supplies a plausible condition for influence, but the operation and consequences of pressure require evidence.
3. Which observation most directly starts a sourcing-filter inquiry?
- A story appears on a website
- The report contains a photograph
- Regular briefings dominate the account while other relevant voices are absent
- The report uses short sentences
Reveal answer 3
Answer: C. The source pattern prompts questions about access, resources and authority. It does not alone establish deliberate exclusion or falsehood.
4. An organisation identifies a genuine factual error and seeks a correction. Which response is most defensible?
- Call every correction request censorship
- Refuse to correct the error because the critic is powerful
- Assume the entire report was fabricated
- Correct the error and examine any separate coercive pressure on its own evidence
Reveal answer 4
Answer: D. Accountability and disciplining pressure must be distinguished. They may occur in the same dispute.
5. An interview dismisses critics as enemies without examining their evidence. Which inquiry is most directly relevant?
- Whether a platform uses the same technical algorithm everywhere
- How ideology and enemy framing may narrow the space for debate
- Whether every threat must therefore be imaginary
- Whether all journalists share identical private beliefs
Reveal answer 5
Answer: B. Examine whether the label displaces scrutiny of the claim. This does not establish that every threat is imaginary or that every criticism is justified.
6. What do “worthy” and “unworthy victims” mean in the model’s terminology?
- Some people deserve less concern
- A legal classification of victims
- Ironic labels for unequal media treatment associated with political usefulness
- Two groups that must receive exactly the same story count regardless of circumstances
Reveal answer 6
Answer: C. The terms criticise unequal representation; they do not rank human worth. Comparisons still need to account for relevant differences.
7. A report uses accurate facts but omits major relevant costs. What follows?
- Accuracy of individual facts does not rule out a selective or propagandistic presentation
- Every omission proves the full five-filter model
- The report cannot be examined for propaganda
- The audience must have accepted it
Reveal answer 7
Answer: A. Propaganda can use truthful material selectively. Explaining that selection through a particular institutional mechanism is a further task.
8. Several outlets publish a similar account. What is the best first step?
- Conclude that their owners coordinated a campaign
- Conclude that all outlets are independent
- Ignore the similarities
- Check shared sources, verification, timing and possible institutional pressures
Reveal answer 8
Answer: D. Similar content can arise through different mechanisms. A sound inquiry distinguishes them.
9. Which finding would most directly challenge a proposed local application?
- A result that differs from the researcher’s politics
- Repeated absence of a pre-specified predicted pattern, with decisions better explained by documented alternatives
- Any single critical headline, regardless of the claim
- The existence of a media company
Reveal answer 9
Answer: B. A testable application must allow evidence to lower confidence in its explanation. Specify the prediction and comparison in advance.
10. What additional evidence is required to claim that a coverage pattern changed audience beliefs?
- Only an ownership chart
- Only a count of stories
- Relevant audience research that can examine exposure, interpretation and competing explanations
- Only a producer’s stated purpose
Reveal answer 10
Answer: C. Content analysis does not by itself establish audience effects. Reception and causal claims need methods suited to them.
Applied assignment · 400–500 words
Select a news issue and a small, clearly bounded sample of reports. Identify a pattern you could code. Propose one or two relevant filter mechanisms, specify the evidence needed to examine them, and consider an alternative explanation. State one result that would weaken your interpretation. Conclude with the limits of the proposed study.
| Criterion | Points | Check |
|---|---|---|
| Defined case and comparison | 2 | Are the outlets, issue, period and selection rules clear? |
| Accurate concepts | 2 | Are institutional filters distinguished from factual errors, personal bias and audience effects? |
| Mechanism and evidence | 3 | Does the evidence address how the proposed pressure could shape decisions? |
| Alternatives and counterevidence | 2 | Would the design recognise a result that challenges the explanation? |
| Proportionate conclusion | 1 | Does the conclusion remain within the scope of the evidence? |
Readiness check
These checkboxes are a temporary study aid. They do not save course progress.
Keep the concepts precise. Keep the conclusions open to revision.
Key terms
- Propaganda
- Organised communication directed towards responses that serve a sponsor’s aims; it can include truthful as well as deceptive material.
- Political economy
- Inquiry into how resources, ownership and power organise production, circulation and access.
- Filter
- An institutional relationship or pressure proposed to shape recurring news choices and treatment.
- Source dependence
- Routine reliance on actors able to supply usable information and recognised authority.
- Flak
- Negative responses that can impose costs on media and discipline coverage.
- Dominant ideology
- Assumptions that help define what appears natural, legitimate or open to debate in a setting.
- Enemy framing
- Organisation of a dispute around a threatening opponent, potentially narrowing how dissent is heard.
- Humanisation
- Representation that gives people identities, experiences and context rather than treating them only as an anonymous total.
- Media performance
- Patterns in what media produce, select, emphasise and sustain.
- Audience effect
- A change in audience beliefs, judgments or conduct; it requires evidence beyond the media content itself.
Read the argument—and the debate.
Start with the lecture and original model, then compare Herman’s defence with Sparks and Corner. The later readings support adaptation, not automatic transfer of the original scheme to every setting.
- Pandey, U. S. (Dr. USP). The Propaganda Model of Media Control [Video]. YouTube.
Open source / publication record ↗
Why read it? The video supplied for this chapter. Use the viewing guide and applied questions above. - Herman, E. S., & Chomsky, N. (1988). Manufacturing consent: The political economy of the mass media. Pantheon Books.
Open source / publication record ↗
Why read it? Read the original model. Link: an excerpt hosted on Chomsky’s website. The publisher also lists the 2002 edition with a new introduction. - Herman, E. S. (2000). The propaganda model: A retrospective. Journalism Studies, 1(1), 101–112.
Open source / publication record ↗
Why read it? A defence and clarification of the model, including its distinction between media performance and audience effects. - Herman, E. S., & Chomsky, N. (2009, November). The Propaganda Model after 20 years [Interview by Andrew Mullen]. Westminster Papers in Communication and Culture.
Open source / publication record ↗
Why read it? A later discussion of filters, ideological change and the model’s scope. Link: the interview text on Chomsky’s website. - Sparks, C. (2007). Extending and refining the propaganda model. Westminster Papers in Communication and Culture, 4(2), 68–84.
Open source / publication record ↗
Why read it? A sympathetic but critical refinement that addresses variation, conflicting interests and journalism. Open access. - Corner, J. (2003). The model in question: A response to Klaehn on Herman and Chomsky. European Journal of Communication, 18(3), 367–375.
Open source / publication record ↗
Why read it? Read a critical position alongside the authors’ defence. Publisher access conditions may apply. - Snow, N., Jowett, G. S., & O’Donnell, V. (2025). Propaganda & persuasion (8th ed.). Sage.
Open source / publication record ↗
Why read it? For wider propaganda concepts and forms. Read the publisher’s chapter extract. The edition carries a 2025 copyright; the publisher lists its release in August 2024. - Pedro-Carañana, J., Broudy, D., & Klaehn, J. (Eds.). (2018). The propaganda model today: Filtering perception and awareness. University of Westminster Press.
Open source / publication record ↗
Why read it? Open-access discussions of theory, digital media and applications across settings. - Lasswell, H. D. (1927). Propaganda technique in the World War. Alfred A. Knopf.
Open source / publication record ↗
Why read it? Historical background to propaganda research. Link: the University of Pennsylvania’s Online Books Page record.