Willingness to recognize the bare possibility of analysis of the media in terms of a propaganda model, as in work of the past years cited earlier, is so uncommon that the few existing cases perhaps merit a word of comment. Lemann's critique of our Manufacturing Consent is, in fact, one of the rare examples. His review contains several allusions to the book, few of which even approach accuracy; the example just cited is typical. We may dispense with further discussion of the falsehoods,¹⁸ the stream of abuse, or the occasional apparent disagreement over facts, for which his evidence reduces to "the literature" or common knowledge, which allegedly does not confirm what he claims that we assert.

Consider, rather, Lemann's criticisms of our presentation of the propaganda model. His main point is: "in no instance do they prove" the claim that the press "knowingly prints falsehoods and suppresses inconvenient truths." He is quite right. In empirical inquiry, nothing is ever literally proven; one presents evidence and tries to show that it can be explained on the basis of the hypotheses advanced. A critic could then rationally argue that the evidence is mistaken, poorly chosen, or otherwise inadequate, or that there is a better theory to explain the facts. Lemann suggests no inadequacy of the evidence (when we eliminate false allegations), but does appear to suggest an alternative theory. It is that "the big-time press does operate within a fairly narrow range of assumptions" and "concentrates intensely on a small number of subjects at a time," shifting attention "unpredictably from country to country" and reflecting "what Herman and Chomsky, meaning to be withering, call `patriotic premises'." He does not, however, proceed to say how this conception of the media explains the facts we discuss, or others, if he regards these as poorly chosen for unstated reasons. Thus, to take virtually the only reference to the book that is accurate, he notes with much derision that we give actual figures (worse yet, in "tabular lingo") concerning the relative attention given to the murdered Polish priest Father Popieluszko and 100 Latin American religious martyrs. Clearly, the case confirms our hypothesis ("which, of course, turns out to be correct," he writes with further derision). Does the case support Lemann's alternative theory? Insofar as his proposals differ from ours, they plainly have nothing whatsoever to say about these facts, or about any others that might be relevant.

In response to a letter by Edward Herman raising this point, Lemann elaborates: "As for Father Popieluszko, he was killed when the U.S. press was most focused on Poland. Archbishop Romero was killed before the press had really focused on El Salvador. Popielusko's murder wasn't more important; the discrepancy can be explained by saying the press tends to focus on only a few things at a time." This, then, is the explanation of why the media gave far more coverage to the murder of Father Popieluszko than to the murder of 100 religious martyrs in Latin America, including Archbishop Romero and the four U.S. religious women raped and murdered, and why the coverage was so radically different in character, as shown in detail. Let us ask only the simplest question: how much coverage were the media giving to El Salvador and to Poland when Archbishop Romero and Father Popieluszko were murdered? We find that the coverage was almost identical, eliminating this proposed explanation without any further consideration of its quite obvious flaws.¹⁹

Once again, the only plausible conclusion is that it is the very idea of subjecting the media to rational inquiry that is outrageous, when it yields conclusions that one would prefer not to believe.

Confirming further that this is precisely what is at stake, Lemann condemns us for "devot[ing] their greatest specific scorn to liberal journalists...in the time-honored tradition of the left," particularly Stephen Kinzer, Sydney Schanberg, and William Shawcross. He does not, however, explain how one can investigate the coverage of Central America and Cambodia by the New York Times while avoiding the work of its correspondents there; or how one can explore the remarkable success of the idea that the left imposed "silence" on media and governments during the Pol Pot years -- by publications that went to press after the overthrow of Pol Pot, no less -- without reference to its creator. Quite evidently, it is the topics addressed that Lemann finds unacceptable, for reasons that can readily be discerned. These observations apart, Lemann appears not to understand the elementary point that discussion of the most dissident and critical elements of the media is of particular significance, for obvious and familiar reasons, in exploring the bounds that are set on thinkable thought.

Throughout, Lemann is particularly incensed by attention to fact, as his derisive comments about "tabular lingo" indicate. Thus he writes that we "dismiss the standard sources on the countries they write about," as in discussing coverage of the Nicaragua election, making use instead of such absurd sources as the report of the Irish Parliamentary Delegation of largely center-right parties and the detailed study of the professional association of Latin American scholars (whom we call "independent observers," he adds derisively, apparently regarding Latin American scholars as not "independent" if their research does not conform to his prejudices). Asked by Herman to explain why he finds our use of sources inadequate in this or any other case, he writes: "By standard sources, I mean the American press, which usually weighs the government handouts against other sources." What he is saying, then, is that in investigating how the media dealt with the Nicaraguan election, we must rely on the media that are under investigation and not make use of independent material to assess their performance. Following this ingenious procedure, we will naturally conclude that the media are performing superbly: what they produce corresponds exactly to what they produce. Quite apart from this, Lemann does not seem to comprehend that our account of how the media radically shifted the agenda in the case of El Salvador and Nicaragua in no way depended on the sources he finds unacceptable and exotic.

The same is true throughout. It is difficult to believe that such performances are intended seriously. A more plausible interpretation is that the questions raised are so intolerable that even a semblance of seriousness cannot be maintained.

It is sometimes argued that the propaganda model is undermined by the fact that some escape the impact of the system. This is an "anomaly" that the model leaves unexplained, Walter LaFeber alleges. Thus, a "weakness" of the model is "its inability to explain the anti-contra movement that has -- so far -- blunted Administration policy." LaFeber argues further that proponents of the model want "to have it both ways: to claim that leading American journals `mobilize bias,' but object when I cite crucial examples that weaken" their thesis; the only example cited, the "key exception," is the case of the nonexistent MiGs. He also puts forth a third argument against the model, as it is presented in our book Manufacturing Consent: "If the news media are so unqualifiedly bad, the book should at least explain why so many publications (including my own) can cite their stories to attack President Reagan's Central American policy."

This is one of the very rare attempts to evaluate a propaganda model with actual argument instead of mere invective, and is furthermore the reasoning of an outstanding and independent-minded historian. It is therefore worth unravelling the logic of the three arguments.

Consider the first argument: the model is undermined by the fact that efforts to "mobilize bias" sometimes fail. By the same logic, an account of how Pravda works to "mobilize bias" would be undermined by the existence of dissidents. Plainly, the thesis that Pravda serves as an organ of state propaganda is not disconfirmed by the fact that there are many dissidents in the Soviet Union. Nor would the thesis be confirmed if every word printed by Pravda were accepted uncritically by the entire Soviet population. The thesis says nothing about the degree of success of propaganda. LaFeber's first argument is not relevant; it does not address the model we present.

¹⁸ I mention merely one, because Lemann gives it as the clinching evidence of our lack of "commitment to truth": "Herman and Chomsky say that `principled and courageous resistance' was a more common response of draft-age Americans to Vietnam than the seeking of deferments." The quoted phrase can be found on page 252, in the course of our discussion of how the PBS series on the Vietnam war gave "short shrift" to the peace movement. As one example, we noted that the search for deferments "hardly defined `the spirit of the times'" as the series claimed (interviewing Lemann's colleague James Fallows), "although it is a facet of this `spirit' that is far more acceptable to mainstream opinion than the principled and courageous resistance of many thousands of young people." Lemann's falsification of this accurate statement merely shows that he falls within the mainstream, as there described, putting aside the matter of "commitment to truth."

¹⁹ Letters, The New Republic, March 6, 1989. Romero was murdered on March 24, 1980; Popieluszko was abducted on October 20, 1984, then murdered. One natural comparison, then, is the columns of the New York Times index for El Salvador in 1980 and for Poland from August 1984 through July 1985 (the comparable period), obviously excluding the coverage of these incidents themselves. Coverage of El Salvador is slightly higher by this measure.