When Erick D. Langer of Carnegie Mellon University reviewed my book Bolivia: Press and Revolution, 1932–1964 (1986), for the Hispanic American Historical Review, he seemed surprised that newspapers could be regarded as something more than mere chroniclers of the passing scene. My goal was to examine how Bolivian newspapers either instigated or hindered social change in the Bolivian National Revolution (1952–1964), the second social and economic revolution in Latin American history. But Langer wrote, for example, "For an analysis of the tin baron Simón Patiño's role in Bolivian society, the author depends exclusively on the judgment of MNR [National Revolutionary Movement] newspapers, hardly a complete or objective source."

Unfortunately, too many historians seem to share this view of newspapers only as sources of factual information. Anything not "complete" or "objective" must be discarded. (It should be added, however, that in a response to my criticism of his review, Langer later stated, "I gladly concede that La Calle and other newspapers might have played a central role in the triumph of the 1952 revolution.")

Too often in the banquet hall of history, newspapers have been relegated to the corner like country cousins or scullery maids. In fact, newspapers were shunned as historical sources by generations of historians. As the Swedish bibliographer Folke Dahl described their attitude, "The publications were said to contain mostly lies, false reports, and the like, while their editors and their publishers were termed newsmongers, news-scribblers, or gossip writers." Not until John Bach McMaster's History of the People of the United States began publication in 1883 did any prominent historian in this country make copious use of newspaper quotations. To be sure, many historians have consulted newspapers since then—traditionally for the political record—but few have recognized the wider significance of the role of the press in both reflecting and shaping society.

In my view, history is concerned—or should be concerned—not only with what actually happened in any given time or place, but also with what people thought was happening, as revealed to them through the means of mass communication, which may have conditioned their subsequent actions. Thus, the perception of events as filtered through the press may have changed the historical outcome. According to this concept, it does not matter if the news is false or distorted as long as readers believed it and acted on their belief. To the historian trying to understand public opinion, then newspapers become primary rather than secondary sources.

Viewed in this light, newspapers and the later electronic media take on a reality of their own, as catalysts of social change or roadblocks of repression. I am not referring to editorials—recent studies have shown that they may reinforce existing beliefs but rarely change attitudes—but rather to the news itself: the selection, gathering, writing, editing, and display of what journalists of any one period or culture consider significant.

Traditionally, newspapers used as factual sources have severe limitations. Under the pressures of time, limited access to information, and available space, a truncated view of society may be presented by the periodical press. And the historian's selection of "facts" from the newspapers adds another subjective dimension to the process. As Frank Luther Mott, whose five-volume History of American Magazines won the Pulitzer Prize, commented:

It is true that quotations without context, arguments wrenched from their proper moorings, or almost any statement once sanctified by print may, if skillfully and unscrupulously used, be made to enforce and buttress almost any thesis. A dishonest historian, by artful selection of current comment and cunning arrangement of such materials, may give a wholly distorted view of conditions and opinion.

Likewise, in the press itself there are sins of commission and omission. As stated earlier, for these very reasons most early historians shied away from newspapers as historical sources altogether. As James Ford Rhodes wrote as late as 1909, "A modern newspaper, though probably true, if quoted in a book as testimony, would be laughed at; but the letter of a court gossip, if written some centuries ago, is thought good historical evidence."

Even today newspapers seem to be peripheral sources for many historians, who may give them a cursory glance for a few juicy quotes to enliven the pages of already completed research. The emphasis is changing, however, as some scholars realize that newspapers and other forms of communication strike responsive chords in the public; otherwise, they could not exist economically. To a certain extent, then, newspapers are also gauges of public opinion. Since their inception in 1609 they have become the lingua franca of society, the most valuable index we have of measuring popular attitudes. It is a two-way street. As Herbert L. Matthews, the distinguished New York Times reporter who covered the Spanish Civil War and the Cuban Revolution, once phrased it, "The picture they [the American mass media] draw is a response to a predisposed public opinion which is both satisfied and moulded by it."

A few historians have wrestled with the problem of newspapers as historical sources. Best known perhaps is Lucy Maynard Salmon whose monumental 566-page study of The Newspaper and the Historian and 505-page sequel The Newspaper and Authority both appeared in 1923. Salmon was the first to recognize the value of studying advertisements for social history, but she concerned herself mainly with the reliability of newspapers as historical sources. Speaking more to my thesis is Elizabeth L. Eisenstein, who wrote The Printing Press as an Agent of Change (1979), but she confines herself in an otherwise admirable study to books and pamphlets.

In 1908 a session of the annual meeting of the American Historical Association dealt with the use of newspapers by historians, but to my knowledge such concern has not surfaced again in the AHA. (Information to the contrary would be greatly welcomed.) More recent comment and analysis are also rare. A computer search turned up only three articles on newspapers as historical sources, including one which deals with the use of newspapers for local history projects in high school classes.

Nevertheless, there have been some brilliant studies by American historians using newspapers as primary source material, and these works are increasing in number. The historiographical breakthrough came with Marcus M. Wilkerson, Public Opinion and the Spanish–American War (1932) and Joseph E. Wisan, The Cuban Crisis as Reflected in the New York Press (1934). After Joseph Goebbels focused attention on mass manipulation in the events leading up to World War II, a flurry of works appeared which considered newspapers as actors in the historical drama, not mere spectators seated comfortably in the box seats.

Illustrative of this trend are Philip Davidson, Propaganda and the American Revolution (1941); Arthur M. Schlesinger, Sr., Prelude to Independence: The Newspaper War on Britain, 1764–1776 (1958); and Bernard Bailyn and John B. Hench, eds., The Press and the American Revolution (1981). And one must mention such works as Donald E. Reynolds, Editors Make War: Southern Newspapers in the Secession Crisis (1970), to indicate that analysis of the press has been done in other periods of United States history.

Still, most of these works deal with the more specific—and more narrow—concept of propaganda, relying (except for those dealing with the American Revolution) mainly on editorials after comment began to be separated—ostensibly at least—from the news in the first decade of the nineteenth century. But propaganda suggests conspiracy, and the newspaper must be taken as a whole, more frequently than not reflecting unconscious biases. As Salmon pointed out, "For the historian is concerned not simply with the accounts of material events, he is equally interested in the interpretation of the spirit of a time or locality. This spirit is revealed both by the true and by the false accounts given by the press."

But how does one measure the effect of the news on readers of a bygone era? There is no easy answer because any one solution appropriate for one period or topic may be wholly inappropriate for another. Quantitative research is a good case in point. I once asked my Ph.D. advisor Dumas Malone, winner of the Pulitzer Prize for his six-volume biography of Thomas Jefferson, what Malone thought about quantitative history and he replied, "If you can count them, count them, but if you can't count them, don't try." In other words, quantitative research may be quite helpful in studying voting patterns or medieval land tenure, but foolish for assigning numerical values to loaded adjectives in the yellow press and deciphering them, a process the trained historian could have done directly by a careful reading in the first place.

Content analysis is a fairly recent technique advanced for studying the news media. In its most primitive form it involved measuring column inches of a news story and size of the headline to determine the play or importance given an event, but again this technique has its limitations. It neglects the question of what is contained in those column inches: a two-paragraph filler may be more significant than a two-column story. Establishing categories of content have refined this technique, but backlash to both quantitative research and content analysis has led to the formation of a qualitative research division within the Association for Education in Journalism and Mass Communication, which holds boldly that there is not necessarily magic in numbers.

And Calder M. Pickett, professor emeritus of journalism at the University of Kansas, points out that the most sophisticated sampling scheme might inadvertently skip over significant events. For my dissertation, "The Jefferson Years: Response by the Press, 1801–1809," I went through fifty-six years of issues of selected Jeffersonian and Federalist newspapers, page by page, and about sixty-five years of newspapers for my book on Bolivia. For projects of lesser scope, my technique is what I call the "cluster method"—examining newspapers and magazines carefully for ten days before an event and ten days afterward, or for longer periods of time if one is dealing with a trend or issue rather than a specific event.

Still, how does one evaluate the effect on the public by newspapers of the past? Today scientifically designed questionnaires and interviews provide some index as to how persons perceive the news, those things selected from the "glut of daily occurrences," as one British colonial printer phrased it. Except when working in the recent past when oral history can be used—I conducted about twenty-five interviews in my Bolivian research—one is forced to rely on traditional historical sources for comments on the contemporary impact of the press—letters, diaries, memoirs, and other documents.

Yet this raises another question: does the person of the past immersed in any one period have a clear perspective on it? Trends may be unfolding or undercurrents of opinion flowing of which he or she is completely oblivious but which may be evident to trained historians who presumably have more sources at their command and the dispassionate distance of time. At the very least, the historian can examine and present the newspaper record of what was offered to the public and let readers form their own conclusions.

In this regard, it must be remembered that in earlier times or less advanced societies readers were less sophisticated, so the press had greater credibility. What was presented was likely to be believed. As Carlos Montenegro, intellectual bellwether of the Bolivian National Revolution put it: "The [Bolivian] public did not have, throughout half a century, any other source of cultural nutrition than journalism, and it learned to attend to and judge things in consultation with the printed [newspaper] page."

My concern about relying exclusively on one historical technique is that it may wrench the newspaper out of the context of its time. Here ownership of the means of mass communication—what media critic Ben H. Bagdikian today calls "the lords of the global village"—should be divulged. Herbert S. Klein, for example, relies heavily on La Paz newspapers as factual sources for his Parties and Political Change in Bolivia, 1880–1952 (1969) but fails to inform his readers that the three leading newspapers were owned or controlled by the Big Three tin mining magnates, each with political axes to grind.

Newspapers are big business and tend to be conservative, reflecting the economic and political interests of their owners. Even the renowned New York Times missed the point entirely of the Mexican Revolution of 1910, calling editorially for the assassination of its agrarian reform leader Emiliano Zapata and solemnly reporting in its news columns that Zapata maintained a harem of fifty women. Magazines, with the shining exception of the muckraking era (1902–1912), can be even more freewheeling. During the Mexican upheaval, for example, Harper's Weekly denounced in an article "the slouchiness, the laziness, the stupidity, and the cowardliness of the average Mexican. ..." Is this the stuff of history? Yes, if it turned Americans against the Mexican struggle for national redemption.

Newspapers can also play a more direct role by intervening directly in the historical process, whether in broad sweeps or isolated incidents, sometimes leaving historians who neglect them stranded. In Mexico in 1911, for instance, when the fighting was underway, Francisco I. Madero, political leader of the revolution, responded to some written questions put to him by William Randolph Hearst by stating: "I knew that General [Porfirio] Díaz could only be defeated by means of arms, but the democratic [electoral] campaign [of 1910] was indispensable in order to make a revolution because this would prepare public opinion and justify the armed uprising." Without knowledge of this document, which I discovered in the Biblioteca Nacional in Mexico City, Madero's U.S. bibliographers Stanley R. Ross stated that Madero opposed revolution; Charles Curtis Cumberland cited Madero's "desire to have recourse to arms only as a last resort"; and Robert E. Quirk wrote there was "no indication that he [Madero] planned or even favored a revolution."

A better known incident was the famous interview by James Creelman with aging dictator Porfirio Díaz (1876–1911) published in Pearson's Magazine in March 1908, which contained the Mexican leader's assertion that he would not run for an eighth term in 1910. Intended only for foreign consumption, the interview was translated and circulated by the underground press in Mexico, which set the political pot bubbling and led to revolution. The Mexican government considered this article so decisive that it recently printed a facsimile reproduction, along with commentary and Spanish translation.

On the other hand, few U.S. historians seem to realize the role played by the oligarchical press of Mexico City in undermining the fragile democratic government of Madero after he gained power in 1911, making him appear vulnerable and encouraging the successful counterrevolt and his murder in 1913. The newspapers are there, but they have not been thoroughly examined.

Only the more spectacular incidents recounted in newspapers seem to force themselves on the attention of historians, such as the historic trek into the Sierra Maestra of Cuba by Herbert L. Matthews in 1957 to prove that Fidel Castro was still alive after the disastrous Granma landing a year earlier. Critics charged that Matthews in his three-part story in the New York Times glorifies Castro and was instrumental in his coming to power, but the journalist replied that this was like blaming the weatherman for the weather.

Later, conservative newspapers in Chile undermined the elected Marxist government of Salvador Allende (1970–1973), aided with $4.3 million in subventions from the Central Intelligence Agency. This massive propaganda campaign magnified every violent incident, faked news such as the "March of the Pots and Pans," escalated political passions, polarized Chileans, and sabotaged the economy. The Sandinistas in Nicaragua in the 1980s cited the Chilean experience to justify censorship and periodic closures of the opposition newspaper La Prensa of Managua. Again, the role of the press in either of these historical periods has not been sufficiently studied.

For those interested in taking a fresh look at the colonial or United States press, a number of bibliographical guides are indispensable. First came the lifework of Clarence S. Brigham, History and Bibliography of American Newspapers, 1690–1820, published in 1947. His endeavor was carried on by Winifred Gregory and her staff at the Library of Congress who compiled American Newspapers, 1821–1936: A Union List of Files Available in the United States and Canada (1937). By microfilm and microfiche many of these newspapers are now widely available. An introduction to the more recent period is the two-volume work by Richard A. Schwarzlose, Newspapers: A Reference Guide (1987).

The sheer volume of the accumulation of these and other newspapers makes indexing a noble endeavor. Anita Cheek Milner crossed this threshold with her three-volume Newspaper Indexes: A Location and Subject Guide for Researchers (1977–1982). Now, with database compilations possible, the number of indexed newspapers is growing by leaps and bounds.

Even with these tools, however, it is necessary for anyone working with newspapers to have at least a nodding acquaintance with the historical development of the press itself and its social, economic, and cultural context. Still standard is Frank Luther Mott, American Journalism: A History, 1690–1960 (3rd ed., 1962). For the more recent period, researchers may want to consult Michael Emory and Edwin Emory, The Press and America, an Interpretive History of the Mass Media (7th ed., 1991) and Wm. David Sloan, et al., The Media in America: A History (2nd ed., 1993).

Today, to a large extent, study of the role of the press in societal change has gone by default to communication specialists, although they usually deal with internal press histories and their theoretical work is sometimes incomprehensible to the uninitiated (editors now ask contributors for "accessible language"). Nevertheless, a number of valuable journals to complement Journalism Quarterly and Journalism Monographs have appeared in recent years. These include Mass Comm Review (1973), Journalism History (1974), and American Journalism (1983).

It has not been my intent here to emphasize unduly the role of newspapers as formulators or indicators of public opinion. Obviously, the press is but one piece in a complex historical mosaic, but I have stated my case in terms which might redress a long-standing imbalance and stimulate discussion on this important issue, not excluding the proper use of newspapers as factual sources. In this discussion, there should be greater exchange between historians and journalism scholars, for here is a rich vein for all to mine. As Lucy Maynard Salmon wrote, "the periodical press still remains the most important single source the historian has at his command for the reconstruction of the life of the past three centuries."

Jerry W. Knudson, professor of communications at Temple University, is a former newsman who earned his Ph.D. in history at the University of Virginia and has published widely in the field of journalism history.