Turning to the different question of actual media impact on opinion, comprehensive and systematic studies are lacking, but there is little doubt that the impact is substantial, surely among the educated classes.[20] Analysis of a kind not as yet undertaken would be required to determine more closely just how much impact to attribute to media distortion and filtering, and how much to narrowly conceived self-interest and other causes, in establishing the remarkable illusions that prevail on critical issues. It is also true that, with great effort, some are able to find ways to think for themselves, even to act effectively in the political arena, thus bringing about a "crisis of democracy." But that neither confirms nor refutes an account of how the media function.

Let us put aside for a moment the matter of "the anti-contra movement," and turn to the second argument, based on the "key exception." This we have already discussed. It is no exception, but conforms to the propaganda model (see note 8). This fact eliminates the second argument. But suppose that real cases had been presented of media failure to conform to the government line. Proponents of the model would not "object," as LaFeber believes; this is exactly what the model predicts, as we see when a persistent misinterpretation is overcome.

The propaganda model does not assert that the media parrot the line of the current state managers in the manner of a totalitarian regime; rather, that the media reflect the consensus of powerful elites of the state-corporate nexus generally, including those who object to some aspect of government policy, typically on tactical grounds. The model argues, from its foundations, that the media will protect the interests of the powerful, not that it will protect state managers from their criticisms; the persistent failure to see this point may reflect more general illusions about our democratic systems. In the present case, a propaganda model is not refuted if the media provide a platform for powerful domestic elites that came to oppose the contra option for destroying Nicaragua; rather it is supported by this fact. As noted earlier, by 1986 80 percent of "leaders" (executives, etc.) objected to the contra policy -- as flawed, too costly, and unnecessary to achieve shared goals, to judge by public discussion. A propaganda model therefore predicts that these views should be reflected in the media, thus conflicting with the government line. In fact, the model arguably does fail in the case of the contras, though in a manner opposite to what LaFeber believes: as we have seen, the media not only adopted without thought or question the basic doctrines of the narrow (and quite remarkable) elite consensus on Central America policy, but even kept largely to the extremist position of the incumbent state managers, thus showing a degree of subordination to state authorities beyond what the model expects.

Having clarified this point, let us return to the "anti-contra movement that has...blunted Administration policy." Here some care is necessary. There are two very different anti-contra movements, just as there were two very different movements against the Vietnam war. One opposed administration policy on tactical grounds, the other on grounds of principle. After the Tet offensive, much of the corporate elite came to oppose the war as unwise or unnecessary. The same has been true of the contras, as just noted. The popular and principled opposition to the U.S. attacks against Vietnam and Nicaragua did "blunt administration policies," but not through the media. These movements raised the costs to the perpetrators, and in this way were in large part responsible for the ultimate emergence of the narrowly based and self-interested elite critique. But however important these matters, we need not explore them more closely here. The point is that there were two very different kinds of "anti-contra movement"; the media reflected the narrow tactical objections in conformity with their societal function, but never offered more than the most marginal opening to the principled critique, as illustrated by the samples reviewed earlier. Again, the predictions of a propaganda model are confirmed.

What is more, a propaganda model is not weakened by the discovery that with a careful and critical reading, material could be unearthed in the media that could be used by those who objected to "President Reagan's Central American policy" on grounds of principle, opposing not its failures but its successes: the near destruction of Nicaragua and the blunting of the popular forces that threatened to bring democracy and social reform to El Salvador, among other achievements. Analogously, the assertion that the Soviet press transmits government propaganda and tries to "mobilize bias" is in no way refuted when we find in it -- as of course we do -- material undermining the claim that the heroic Soviet military is marching from success to success in defending Afghanistan from bandits dispatched by the CIA. The point is obvious in the latter case; equally so in the former. The third argument thus collapses as well.

Note finally LaFeber's belief that administration policy was unsuccessful. True, in the terms of official propaganda, the policies failed: the U.S. did not "restore democracy" to Nicaragua or establish "democracy" fully in El Salvador and Guatemala. As the propaganda model predicts, the media with virtual unanimity describe the policy as a failure, adopting official pretenses without skepticism or inquiry. If we permit ourselves a measure of critical detachment, thus granting the right to analyze the U.S. ideological system in the manner of other societies, then the conclusions are rather different. Administration policies met with substantial success in achieving the basic goals, though maximal objectives were not attained and the partial failures were costly to the interests represented by the planners -- not exactly an unknown event in history, the Indochina wars being another case.

Perhaps it is worth stressing a point that should be obvious. If the media function as predicted by a propaganda model, then they must present a picture of the world that is tolerably close to reality. Investors have to make judgments based on the facts of the real world, and the same is true of state managers. Privileged and politically active elites, who rely on the media, must have some awareness of basic realities if they are to serve their own interests effectively and play their social roles. Often, these realities demonstrate the ineptness, incompetence, corruption, and other failings of the state managers and their policies. These realities are detectable, even emphasized, in the media, and would be even if their sole function were to provide services to the powerful. To appeal to these facts to show that the media do not attempt to "mobilize bias" is to betray a serious misunderstanding of social realities, not to speak of the logic of explanation.

It is rare to discover in the mainstream any recognition of the existence or possibility of analysis of the ideological system in terms of a propaganda model, let alone to try to confront it on rational grounds. The failure of argument in the few examples that can be found again suggests that the model is indeed robust.

One of the most appropriate ways to test the propaganda model, or any other conception of how the media function, is by close comparison of paired examples. Of course, history does not provide perfect experiments, but there are many cases that are close enough to permit an instructive test. A number of examples are discussed in the text and appendices, many more elsewhere. To my knowledge, they confirm the propaganda model with a degree of consistency that is surprising in a complex social world, and in a manner that is often dramatic.

[20] The few studies that do exist confirm the conclusion. See Manufacturing Consent on studies of the impact of the media, primarily television, in mobilizing support for the Vietnam war, including the self-refuting study published by Freedom House on coverage of the Tet offensive.