The Propaganda Model after 20 Years: Interview with Edward S. Herman and Noam Chomsky

Noam Chomsky and Edward S. Herman interviewed by Andrew Mullen

Westminster Papers in Communiction and Culture, November 2009

Andrew Mullen (AM) – Question 1: The first of the three hypotheses put forward in the Propaganda Model is that, where there is elite consensus, the media will serve elite interests uncompromisingly. Do you have any observations, or are there any general rules, concerning the existence (or not) of consensus among the elite (i.e. are particular issues prone to consensus/dissensus) or should media analysts approach the question on an issue-by-issue basis through empirical investigation?

Edward S. Herman and Noam Chomsky (EH/NC): We have several observations, but no rules. One observation is that elite consensus usually follows when the imperial state projects power abroad, although it may erode if that power projection becomes bogged down and excessively expensive. In the case of the United States (US) and the Vietnam War, that erosion of support forced Lyndon Johnson to step down in 1968, although even then and later there was enough elite support to keep the mainstream media in line and cause them to exclude any fundamental criticism of the war. This was replicated in the case of the Iraq invasion- occupation, 2003-8, when the elite was eventually splintered, but many stayed with the official line and the media continued to support the war (oppose any withdrawal date, etc.). In both cases the media failed to give serious space to news and opinion reflecting the position of the majority. In neither case would the media ever refer to the US government’s action as ‘aggression’. This suggests that the propaganda model can do very well even without a firm elite consensus.

The war-apologetic role was well illustrated in a New York Times ‘Review of the Week’ on 23 November 2008, which offered a collection of think-pieces on the state of the world for the new US president. As regards Iraq and Afghanistan, on which these experts focused, every one presupposed that the use of force by the US was legitimate, even noble. They differed only on tactics and the usefulness and necessity of military action — as with Barak Obama himself, if it fails to work it is a ‘strategic blunder’, with no other objections admissible. In short, there is a firm elite consensus on the legitimacy of state violence — in fact, it is a simple presupposition, which is much more insidious than assertion.

This is all clearly evident in the chapters on Indochina in Manufacturing Consent. [1]

A second observation is that the elite consensus is likely to be strong when fundamental class interests are at stake. This is the case with the just mentioned external projections of power of the imperial state, and the demands of the military-industrial complex with which it is closely associated, and which has grown ever stronger in the US since President Dwight Eisenhower’s January 1961 warning of its threat to US democracy. ‘Free trade’ (i.e. mainly investor rights) issues are also important to elite interests, and this is reflected in vigorous and close-to-unanimous mass media support of ‘free trade’ agreements and opposition to ‘protectionism’. Labour organization and labour power are also of elite class interest, and associated issues frequently produce elite solidarity and mainstream media uniformity in the elite interest. Even here, however, there may be exceptions based on place, time and specific issues. Several commentators on Manufacturing Consent have pointed to cases where the clash of local interests was associated with local media support of local labour’s protests and strikes. [2]

AM – Question 2: The second hypothesis is that, where the mass media is under corporate rather than state control, media coverage will be shaped by what is, in effect, a ‘guided market system’ underpinned by five filters — the operative principles of the Propaganda Model. Can you say how each of these has fared over the last 20 years — would you change any, expand upon any, etc.?


EH/NC: What you refer to as the Propaganda Model’s ‘five filters’ requires some clarification. (a) Ownership and (b) advertising belong to straightforward institutional analysis — these are the kinds of institutional arrangements that predominate among US media firms and elsewhere. (c) Sourcing and (d) flak are two well-established processes to which any elite-serving media will adapt, whether we are talking about the elite US or British media or the elite media under Stalin and Hitler. On the other hand, (e) anti-communism, as a major theme of media production during the twentieth century, was reflective of the prevailing system of belief in the Western states, and has evolved with the collapse of the Soviet bloc since the first edition of Manufacturing Consent. In a crucial sense, and extending from the most minor comic books and cartoons all the way up to the highest academic discussions of the so-called Cold War (i.e. the system of propaganda known as the ‘Cold War’), anti-communism was a staple that provided content, narratives, heroes and villains. Since 1989, this staple has morphed into an array of substitutes. But the structural role that anti-communism and its successors have played, namely, the provision of an Enemy or the Face of Evil, remains as relevant as ever.

To answer the question about the five filters more specifically:

(a) Ownership is more concentrated and more globalized in 2008 than it was in 1988, in good part because of accommodating state policy. This includes weakened anti-trust and relaxed rules on market share limits and cross-ownership rights, most notoriously as regards radio station ownership within and between markets.

(b) Advertising is a more important force in 2008 than in 1988, because of greater competition among traditional media outlets and between those outlets and the internet. Right-wing political forces have also pushed public radio and television into greater dependence on advertising. The result of all this (including ownership concentration) has been a more intense bottom-line focus, a greater integration of editorial and business operations, more product placements, cutbacks in investigative reporting and analysis, more controversy-avoidance, and greater manageability by governments and other power centres.

(c) The above-mentioned changes have made for a greater sourcing dependence on wire services, public relations offerings, and official and establishment-expert claims and press releases.

(d) Flak has very likely become more important as a constraint on the media than it was in 1988. Although aggressive government attacks on problematic media reporting and opinion existed then and earlier,[3] and media dependence on government favours and support has long been a force constraining the media, government has become even more aggressive in favouring and punishing media deviations from the official line. Furthermore, flak from within the media, including the numerous right-wing talk shows and from blogs — constituting a right-wing attack machine and echo-chamber — has become more important in recent decades. One of their high points of achievement was the success of the Swift Boat Veterans for Truth in getting the mainstream media to cooperate during the 2004 presidential election campaign in denigrating the military record of Vietnam veteran John Kerry, while downplaying the truly dubious military record of George W. Bush. [4]

(e) Anti-communism has receded as an ideological factor in the Western media, but it is not dead and is still used when needed to denigrate individuals who can be tied to Stalin or Mao or Soviet Russia more broadly (e.g. Milosevic), and the crimes of Stalin or Mao and the Black Book of Communism [5] can be featured periodically to warn against socialism and wrong-headed state intervention. The ‘war on terror’ has provided a useful substitute for the Soviet Menace. Also, the antithesis of communism, the ‘free market,’ has been elevated to more prominent ideological status, and has proven to be a strong co-replacement for anti- communism and the basis for the new world order of neoliberalism [6] now in some disarray but without an ideological rival resting on any kind of power base.

We should note that the ideology of the ‘free market’ prevails despite the fact that ‘exceptions’ crop up with great regularity, as in the case of the Agreement on Trade-related Aspects of Intellectual Property Rights (TRIPS) and periodic bailouts, and that the entire advanced economy has long relied heavily on the dynamic state sector to socialize costs and risks and privatize profit, as in the cases of funding the development of computers and the internet, and the ongoing Pentagon system. One of the most amusing cases is that of US President Ronald Reagan, anointed as High Priest of the free market — while breaking all records for protectionism in the post-war period, and much else.

AM – Question 3: If you were devising the Propaganda Model now, in 2008, what would it look like?

EH/NC: It would look very much like the 1988 version, with the ‘free market’ as a principal ideological underpinning along with ‘anti-terrorism’ and the ‘war on terror’ that have provided the needed Enemy or Face of Evil, with anti- communism pushed into a back-up and reminder/ideological role. Within this slightly revised framework, we would probably place more emphasis on globalization and dependence on government for favours and service; on aggressive government news management; on the rise or strengthening of right- wing mass media institutions (Sinclair, Clear Channel, Fox News), talk shows and blogs; and on the real but thus far weaker growth of other alternative media (including those based on the internet).

AM – Question 4: Your methodological approach is based upon the use of (a) empirical data (specifically content analysis, i.e. counting column inches), (b) the use of official sources and (c) the use of paired examples. Can you explain why you adopt such a framework in your analysis? Could such an approach be augmented by others (discourse analysis for example)?

EH/NC: We don’t confine ourselves to newspapers or official sources; e.g. in our important Table 2-1 on Worthy and Unworthy Victims [7] we have a column on CBS television news coverage, and our main focus is on the media themselves as sources, however they derive their own information (admittedly, very heavily from official sources). We do feature heavily the print media, partly because of ease of access and use, but also because they are more complete, with fuller information than television news, the latter often derivative from print media sources. Any weaknesses in objectivity or sourcing in the leading print media would commonly be amplified in television reporting. Where properly chosen, paired examples highlight bias, with quite similar evils (or virtues) treated very differently in alignment with the political interests of the state. This is a classic method of analysis [8] that fits well an analysis featuring propaganda service.

Our method IS a form of discourse analysis. We do not claim a monopoly of routes to the establishment of truth, so other forms of discourse analysis and other frames of analysis can complement, supplement and on occasion possibly yield better results than our own. [9] However, we believe that our broad framework continues to be serviceable. We note that it is often admitted by critics to work well in the important cases that we have discussed, but they rarely stop to explain why this is so and its implications.

AM – Question 5: The third hypothesis is that the Propaganda Model, media performance (specifically media that serve elite interests), and studies of such media performance will be ignored and marginalized. This is where critics of the Propaganda Model have focused their attacks. Can you deal with each of these criticisms in turn:

(a) The Propaganda Model presents a conspiratorial view of the media.
(b) It is deterministic, functionalist and simplistic.
(c) It ignores journalist’s professionalism and the fact that they can be agents for change within the system.
(d) It is overly ambitious and projects a ‘total and finalizing’ view.
(e) In the post-Cold War world, anti-communism is redundant.
(f) By deploying notions such as ‘brainwashing under freedom’, ‘propaganda’ and ‘thought control’, the Propaganda Model is indeed concerned with media effects and not just media behaviour.
(g) The Propaganda Model is not relevant in countries outside the US, which have different media and political systems. (Given the global nature of many media operations today, do you think that the Propaganda Model is applicable in most places?)

EH/NC: (a) We are very clear that the Propaganda Model does not rest on any conspiracy assumption but is rooted mainly in market-oriented processes. But many critics have not been able to see how similar results could arise without conspiracy, hence there must be an underlying conspiracy assumption. But in fact what seems to be conspiratorial behaviour is easily explained by natural market processes (e.g. use of common sources, laziness and copying others in the mainstream, common and built-in biases, fear of departure from a party line, etc.). We should note that some critics who claim that ours is a conspiratorial view do this by latching on to an occasional word or phrase we made that suggests planned action. It is true that occasionally common results arise at least in part from knowing joint action, sometimes by government request or pressure, but these are the exceptional cases. The market can do the job well, and we are very clear and explicit that this is the main mechanism through which the PM does its work.

An important factor in the charge of ‘conspiracy theory’ (and general hostility to the Propaganda Model) is that many journalists find it difficult to accept the notion that institutions like those comprising the mainstream media can work to produce outcomes that run contrary to the self-understanding of the social actors who work for these institutions, and who contribute to these outcomes. Thus, harking back to something we asserted in the first edition of our book, whereas this type of critic appears to believe that the societal purpose of the media is to enlighten the public, and to enable ‘the public to assert meaningful control over the political process by providing them with the information needed for the intelligent discharge of political responsibilities’, we believe, to the contrary, that the evidence shows that the societal purpose of the media is ‘to inculcate and defend the economic, social, and political agenda of privileged groups that dominate the domestic society and the state’. This difference in view of the media’s role is hard to bridge.

(b) Every model is deterministic; the issue is whether it is useful in understanding and forecasting. It is also simple, but our belief is that that is appropriate for a model that deals with a very complex and changing reality and tries to spell out basic parameters that will only provide broad guidelines and insight into the broad sweep of news coverage and permissible opinion. On the question of functionalism, by definition a propaganda model lays out a framework that points to a propaganda function, although most of the participating individuals may be unaware of this. We do not consider this a serious criticism, unless it is shown that the model fails to enlighten or predict on the propaganda service that the model is being used to describe.

(c) If those professional standards were sufficiently powerful and relevant the Propaganda Model would soon be shown to be false — and the media would not go along with staged elections in a terrorized El Salvador, a highly questionable claim of the Bulgarian-KGB involvement in the 1981 shooting of the Pope, and the Bush-era claim of the menace of Iraq’s weapons of mass destruction, among other blatant cases of propaganda service. The criticism neglects the questions of power, ideology and the problematic standards of journalist objectivity, which have long pushed them towards excessive if not exclusive reliance on establishment sources and to easy manageability in consequence.

(d) It doesn’t claim to do everything and cover every facet of news- making and opinion dissemination. It only shows the main thrust in many important cases, especially where elite interests are clear and strong, elite opinion is unified, and oppositional interests are unclear and disorganized.

(e) Redundant because its counterpart, free market ideology, and anti- terrorism and the ‘war on terror,’ have moved front and centre. However, anti- communism is still a force that can be mobilized to show the lesson of excessive government intervention and questioning of free market principles.

(f) We are certainly interested in the effects of propaganda, and believe that it is often a powerful force. However, our model focuses on the media’s work and performance, not the impact of the media’s propaganda efforts. We have repeatedly pointed out that propaganda is more important for the elites in a relatively democratic country like the US than in a totalitarian country, given the restraints on the use of violence in the former. It follows that propaganda service is important, but its effectiveness is variable and uncertain, and we don’t try to measure it.

(g) Globalization and cross-border integration, and the spread and increased importance of commercial media and advertising as a funding source, have made the Propaganda Model ever more widely applicable, but it has to be evaluated on a case-by-case basis given the varying degrees and forms of penetration, and different cultural conditions and levels of government intervention. In some cases, like Russia, we may have a slow merging of an older form of state propaganda with an emerging market-based system.

AM – Question 6: Do you have any opinions or thoughts on the following reasons, put forward by Mullen, [10] as to why the Propaganda Model, and your wider work on the media, has been ignored within the field of media and communication studies?

a) You are ‘outsiders’ to the discipline and, therefore, not legitimate analysts.
b) You have been smeared as apologists for totalitarian regimes and are considered by many to be ‘controversial’ figures, to be avoided.
c) Following the ‘cultural turn’ in the 1980s/1990s, and the rise of postmodernism, there has been a retreat from empiricism, grand narratives and political economy approaches to media analysis.
d) The Propaganda Model challenges the liberal-pluralist, mainstream view of how the media operate. That liberals ignore the Propaganda Model may be no surprise, but why, have many critical and Marxist analysts not been overly receptive to the Propaganda Model and your wider work?

EH/NC: Each of these has an element of truth. We are outsiders to the discipline and have brought a new and radical critique of the mainstream media into the arena. This causes resentment from some members of the media-analysis establishment at our turf encroachment, and even more resentment at the implication of the media’s systematic propaganda service and non-reformability within current institutional structures (see 5(a) above). This has been true even of left critics, many of whom still can’t abide these radical ideas that make their ‘practical’ policy proposals seem misdirected and liberal utopianism. Many leftists and Marxists don’t like radicalism, especially when not of their own authorship.

Of course we are out of step with postmodernism, but that trend has represented a retreat from serious radical criticism altogether. And our controversial character and (falsely) alleged apologetics for totalitarian regimes are largely modes of rejection and back-handed attacks on our radical analysis rather than reasons for that rejection. Radicalism automatically produces ‘controversy’, and criticism of an attack on a quickly demonized enemy automatically makes the critic an ‘apologist’ for the leader(s) of the victimized state, not a defender of the victimized people or of the rule of law — or media integrity in dealing with national enemy’s and targets.

AM – Question 7: You applied the Propaganda Model to newspaper coverage. Do you think that it is applicable to other forms of media (television for example)?

EH/NC: Yes, it should apply well to television, which is even more dependent on adverts than the print media, often shares ownership with the latter in conglomerates, and is even more sensitive to flak than the press. [11] It will also increasingly apply to the internet, where the ‘old media’ have a growing place and advertising has become steadily more important in the newer internet-based media institutions.

AM – Question 8: Does the internet age (blogging, podcasting, etc.), and the challenge this poses to the traditional media (as the internet provides people with alternative sources of news), mean that the Propaganda Model will become increasingly marginal in its applicability?

EH/NC: It is possible that this might happen, but it hasn’t yet and there are several factors protecting the hegemony of the traditional media: (1) the traditional media themselves have occupied the internet and are dominant news providers there; (2) they have the resources and pre-existing audiences to give them a huge advantage over alternative media potential rivals; (3) the alternative operators on the internet seek advertising revenue to fund their operations, compromising their alternative character, and the biggest, like Google and Yahoo, are heavily dependent on advertising revenue (and they are not inclined to put resources into original news origination); (4) much of the new media on the internet is oriented toward facilitating social connections, with politics secondary at best, and the best of the new alternative media have limited resources and outreach and specialize in critical analysis rather than news-making. [12] What would make the Propaganda Model more ‘marginal in its applicability’ is not the rise of blogging, podcasting and other potential media vehicles, but rather the diminution of class and hierarchically organized social orders, and the spread and deepening of egalitarianism. As long as highly unequal and unfair economic and social orders persist, their dominant elites will have to justify themselves and they will continue to need supportive propaganda. The media structures that will help them will keep the Propaganda Model and its filters relevant.

AM – Question 9: The Propaganda Model is concerned with media performance. Do you have any thoughts about the impact of the operation of the Propaganda Model on media audiences? Are there any models of media effects which you think could complement the Propaganda Model?

EH/NC: We believe that propaganda is often effective, but not always and in varying degrees that depend on many circumstances, such as the target audience’s interest and knowledge, the forms of appeal by the propagandists (fear, patriotism, economic self-interest, etc.), the intensity of the propaganda campaign, the degree of media cooperation, and perhaps other factors. Propaganda can be useful to those who employ it even if only a substantial minority are persuaded or confused into passivity — this persuaded or inactive minority may be enough to allow the propagandists to ignore majority opinion.

It is true that audiences can pick and choose among media offerings, but only within the limits of what the media will offer, and audiences may be virtually forced to deal with messages they don’t want and even reject. The audiences are not sovereign even if they have a (constrained) freedom of choice. It should also be kept in mind that propagandists have the power to raise the frequency and intensity and adjust the character of their messages in the face of audience resistance. Active audience analysts, beware!

AM – Question 10: What hope do you see for future resistance to the corporate domination of the mass media? Can you think of any practical examples of such resistance? How do we get from here, a corporate controlled media, to where we want to be, enjoying a truly democratic and plural media scene?

EH/NC: It is hard to be optimistic today as the trend has been toward strengthening the power of the Propaganda Model, which, along with increasing inequality, works toward a plutocratic politics that has a symbiotic and mutually supportive relationship with a concentrating commercial media. This has, however, stimulated a resistance movement that struggles to limit further Federal Communications Commission (FCC) and congressional actions increasing concentration, and which has also created pockets of local and alternative media. We have hundreds of community radio stations, hundreds of public access television channels, the Deep Dish television network, the multi-city news operation and communication system, Indymedia, Fairness and Accuracy in Reporting (FAIR) — which produces a radio programme as well as the journal EXTRA! — and numerous internet operations that transmit alternative views and information that deserves attention, even if not much original newsÑamong them, Electric Politics, ZNet, Counterpunch, Information Clearing House, Dissident Voice, Upside-Down World, Cold Type and Swan’s.

We cannot move to a truly democratic and plural media scene without a radical change in the political economy, which is to say, some kind of economic and political revolution. As noted, the existing media and political systems are mutually supportive, so that media change is not on the media’s agenda, nor on the political agenda. A strong resistance movement gathered momentum in the Bush era, attempting to at least halt the FCC’s relaxing of ownership rules. Featuring the 2002 organization of Free Press, which focused on educating and organizing against FCC-sponsored consolidation, building a membership of 400,000 by early 2008, and with an estimated 165 organizations working on media reform, this movement has had some successes in constraining the FCC. [13] These successes involved a considerable struggle, and they were essentially stopping further regression, not reversing past structural change or even halting the trend occurring in the global market itself. We must continue to struggle, but reducing the relevance of the Propaganda Model has so far proved a very difficult task.


[1] See Edward S. Herman and Noam Chomsky (1988) Manufacturing Consent: The Political Economy of the Mass Media, New York: Pantheon, pp.169-296.

[2] See Deepa Kumar (2007) Outside the Box: Corporate Media, Globalization and the UPS Strike, Champaign (Illinois): University of Illinois Press.

[3] See Mark Hertsgaard (1989) On Bended Knee: The Press and the Reagan Presidency, New York: Schocken Books.

[4] See Eric Boehlert (2006) Lapdogs: How the Press Rolled Over for Bush, New York: Free Press – chaps 6-7.

[5] StŽphane Courtois et al. (1999) The Black Book of Communism: Crimes, Terror, Repression, Harvard University Press.

[6] See David Harvey (2005) A Brief History of Neoliberalism, New York: Oxford University Press.

[7] See Manufacturing Consent, pp.40-41.

[8] See Marc Bloch (1961) The Historian’s Craft, New York: Alfred Knopf – chap.3 ‘Historical Criticism’.

[9] See W. Lance Bennett, Regina G. Lawrence and Steven Livingston (2007) When the Press Fails, Chicago: University of Chicago Press; Robert Entman (2004) Projections of Power, Chicago: University of Chicago Press.

[10] Andrew Mullen, A. (forthcoming) ‘Twenty years on: the second-order prediction of the Herman-Chomsky Propaganda Model’, Media, Culture and Society.

[11] For illustrations, and showing the great power of advertising more than 30 years ago, see Eric Barnouw (1978) The Sponsor, New York: Oxford University Press.

[12] See Theodore Hamm (2008) The New Blue Media, New York: New Press.

[13] See Robert W. McChesney (2008) The Political Economy of Media, New York: Monthly Rview Press – chaps 24-25.