Publication Date

May 1, 1996

"Are movies accurate from the standpoint of history?"

That question, again. In order to buy time I cleared my throat, forgetting the yellow cough button, which glared with reproach. From behind a thick glass window a radio technician looked down at a pad of paper, shaking his head. "Nowadays," I began, my voice dropping in poor imitation of the questioner, "historians avoid the word 'accuracy,' for it implies a certainty that is seldom warranted in human affairs."

"You mean, Dr. Carnes, that you can't say whether Pickett led a brigade at Gettysburg?"

"Well, yes." I sighed. I had been here before, knew that it was about to slip into the murky depths, but could not hold onto the words: "But, you see, movies encompass such vast imaginative realms that historians can never figure out where to post 'no trespassing' signs. At what point in Cecil B. DeMille's Ten Commandments does his evocation of ancient Egypt—its class structure, political system, family relations, music, architecture, clothing, language, rituals—fall below some standard of plausibility, especially when extant records are fragmentary and their meanings contested? Or which of Denzel Washington’s facial and hand gestures or voice tones and intonations wrongly conveyed what Malcolm X was about? From the great sweep of life to its most intricate details—feature films conjure it all. To call any film historically accurate—or inaccurate—is absurd.”

Silence. Dead air clumped about my words like footnotes to a refereed journal article. Pluckier questioners perhaps would have attempted triage, but this one, recognizing that the interview was too far gone, pronounced it DOA. He adroitly segued to the car accident on Broad and Main.

More interviews on radio, and some on TV, were scheduled for Past Imperfect: History According to the Movies, a collection of 62 essays I edited as part of a fund-raising campaign to endow the Parkman and Cooper book prizes of the Society of American Historians. So, of necessity, I came up with a different script in response to The Question, and it went somewhat as follows: “I’m glad you asked that. In one sense, historical films are surprisingly accurate. Relatively few people in this nation or elsewhere visit costume institutes or historical museums, yet millions of us have been exposed to their contents because of movies. Hollywood has in this way, at least, done much to promote historical consciousness.”

I then recounted an example from Richard Slotkin's essay in Past Imperfect about the 1939 film The Charge of the Light Brigade. Slotkin explained how Warner Brothers promoted “historical authenticity” by replicating Victorian postage stamps and requiring their use on interoffice mail, though none would appear on screen. This preoccupation with historical accuracy, however, did not extend to the plot itself, which attributed the Light Brigade’s charge to one Surat Khan, emir of Suristani, who had massacred a British garrison at Chukoti in India. A remarkably sharp-eyed Errol Flynn, peering “half a league onward” toward Balaklava Heights, spots Khan and orders the Light Brigade forward to avenge the massacre and preserve British honor. The Light Brigade, Balaklava Heights, and British India existed; the rest of the story was (im)pure invention, or so I related on air. (I was once taken to task by a historian who countered that the movie’s rendering of the “Chukoti” massacre accurately replicated the Cawnpore massacre of 1857. A good point, I conceded, though I doubted that it had inspired the Light Brigade’s daft dash in 1854. Flynn’s farsightedness, I suppose, was of the temporal kind.)

Just after I had finished telling the story on National Public Radio's "Talk of the Nation," someone from Boston (as I remember) called in to say that she was glad I had mentioned Flynn's "charge," because the movie was her favorite and had launched her upon a lifelong study of the Crimean War.

"But didn't it bother you that the movie got nothing right?" I asked.

"No, not in the least, because what has endured in my imagination all of these years are the movie's vivid pictures.

A fair proposition, and I have thought about it often since then. If historical movies represent the visual elements of the past as accurately (that word!) as can be expected, and if those visual impressions are all that stay with us once we've left the theater or returned the video, does it much matter that Hollywood deconstructs historical chronology, causation, and characters with the giddy zeal of a panel of French semioticians at a meeting of the Modern Language Association? And shouldn't we historians, at least those of us who persist in the clay-footed drudgery of teaching what we think happened, applaud Hollywood for inscribing something upon our student’s all-too-rasa tabulas? Spartacus may not have been born into slavery, or have survived the final battle to deliver a poignant farewell from the cross to some former-day Jean Simmons, but at least Stanley Kubrick’s movie imprinted upon the nation’s historical consciousness the image of white slavery, an arresting contribution to discussions of race in 1960.

But is it true that we mostly remember only the visual elements of movies? My own dim memory responds in the affirmative; furthermore, the strong connection of visual perception to memory goes way back. Cicero, citing earlier authorities, maintained that ideas could best be remembered if they were "conveyed to our minds by the mediation of the eyes," and this observation provided the foundation for the "mnemonic arts" that flourished during the Renaissance.1 American experimental psychologists rediscovered the subject in 1894, when E. A. Kirkpatrick published “An Experimental Study of Memory,” which appeared in the first volume of Psychological Review. Kirkpatrick had shown 10 objects to one group of subjects; to another group, he showed large cards bearing the names of the same objects; and to a third group, he read aloud a list of those objects. Three days later, he asked the subjects to list everything they had been shown or heard. The group that had seen the objects could recall on the average seven or eight of them, but the group that had viewed the cards could remember only two of the objects listed by name. And the third group, who listened to the list of objects, could recall only one object. Kirkpatrick concluded that we remember what we see rather than what we hear, and in regard to visual images, we’re far more likely to remember pictures than words.2

The study of visual perception and memory faded in subsequent decades, perhaps because Kirkpatrick's insight was so, well, obvious. During the first half of the 20th century, human-learning theory focused almost exclusively on how the mind processes and recalls words. (Academic psychology’s preoccupation with text was shared by many social scientists and humanists, culminating in the deconstructionist dictum that we are what we write, though perhaps academic writers simply prefer to write about what they are.) In any case, when experimental psychologists turned resolutely to the subject of visual perception and memory in the 1960s, they repeatedly (though far more elegantly) confirmed Kirkpatrick’s strong link between visual perception and memory. One psychologist, seeking to compare the qualitative aspects of different types of perceptual recall, found that people remembered pictures so much better than words that he could get roughly comparable results by testing readers 15 minutes after they had seen a text and picture viewers 72 hours after having seen pictures. In 1973 a psychologist showed subjects 10,000 slides and found that they remembered on the average 6,600 of the pictures. “Pictures are easy to remember,” another psychologist concluded with refreshing directness.3

Tests on recall and moving pictures produced similar results. In 1979 a psychologist showed a movie without dialogue called The Red Balloon to one group of subjects and read a detailed account of the story to another group. When asked to describe what they had read or seen immediately afterward, readers and movie viewers were equally successful at recalling plot elements. But when asked to do so a week later, readers had forgotten 37 percent of the plot elements, and movie viewers, only 14 percent. More striking still, 98 percent of the readers retold the story in the past tense, while 76 percent of the movie viewers used the present tense, proof that moving pictures remain fresh in our consciousness and readily accessible to recall long after the film script has been forgotten.4

Why people remember visual images better than texts is beyond the scope of this essay and the competence of its author; experimental psychologists are themselves of several minds on the subject. What matters here is the staying power of the visual elements of movies and their tendency to convey a reasonably fair sense of the physical setting and context of the past. When we embarked on Past Imperfect, I had imagined that our historians would find plenty of howling visual anachronisms: Roman senators wearing Timexes, or Revolutionary War soldiers with repeating rifles. To be sure, Michael Grant spotted busts of Hadrian (76–138 A.D.) in MGM’s 1953 version of Julius Caesar (100?–44 B.C.), and Catherine Clinton noticed cotton being harvested in the spring in Gone with the Wind; but errors of visual detail were remarkably rare. On the contrary, our authors were impressed with Hollywood’s often heroic efforts at reproducing historical costumes, interiors, and settings. My favorite example came from William Leuchtenburg’s essay on All the President's Men, for which Warner Brothers built an exact replica of a Washington Post newsroom on a sound studio in Burbank, even shipping authentic trash from the Post wastebaskets to the set.

In the letter accompanying his essay, Leuchtenburg explained that the movie had long been one of his favorites, and he had often shown it in class. But after repeatedly watching the video in order to write his essay, he became aware of a subliminal message conveyed through lighting effects: the agencies and agents of the federal government were nearly always shrouded in "menacing darkness," while Woodward, Bernstein, and the Post were “bathed in light.” The movie’s lighting was in fact its message: the Post functioned—almost literally—as a beacon of truth, piercing the dark and drear recesses of the Nixon presidency and chasing him from office. This was good cinema but bad history, Leuchtenburg concluded, for it omitted the crucial role of the federal government—the Justice Department, the Supreme Court, and Congress—in removing Nixon from office. This dualistic light-dark imagery has evidently been reprised in Oliver Stone’s Nixon (1995). The opening scenes show an ominously darkened White House (the published script describes the mansion as “dark, silent. Like a tomb”). Nixon is hunkered down listening to the Watergate tapes; Haig turns on a lamp and a shaft of light slashes Nixon’s face. He throws up his hands and cowers in fair imitation of Bela Lugosi. (The script calls for Nixon to “gesture awkwardly … he hates the light, slurs a strange growl.”) Later on, E. Howard Hunt makes the point explicit by telling John Dean that Nixon is “the darkness reaching out for—the darkness.”5

The preceding paragraph juxtaposes the customary convention of historical writing (textual citation) with that of film analysis (textual description of a visual phenomenon). The latter approach is tautological, using words to summon up visual recall and in so doing inevitably influencing what is recalled. The literally obvious alternative is to share the evidence with the reader by including plenty of film stills. But there are two serious obstacles to this approach.

Most studios charge hefty fees for use of film stills, as much as $500 each. This does not include the technical cost of pulling a particular shot from a video for reproduction in a book, which can easily cost an additional $300. Past Imperfect included hundreds of pictures and stills, and permission charges exceeded $40,000. (This did not include technical charges and the cost of agents who specialized in wresting permissions from the studios.)

An even more insuperable problem is censorship within the studios. The biggest culprit in our project was Warner Brothers, which agreed to a fee in December 1994 for 12 stills. We assumed that this implied licensing approval. On March 16, 1995, when Past Imperfect was in galleys (and due for imminent publication as the September main selection for the History Book Club), we sought confirmation. A week later Warner Brothers asked to review the text that was to address five of the movies. Accordingly, we faxed Stanley Karnow’s essay on JFK, Clayborne Carson’s on Malcolm X, William Leuchtenburg’s on All the President's Men, Richard Reeves’s on PT-109, and Nancy Cott’s on Bonnie and Clyde. On April 17 Warner Brothers informed us that permission had been denied. Why, they did not say. My repeated calls to the permission’s attorney went unanswered. Executives at Time-Warner chatted in friendly fashion but provided little help and no answers. (“How,” I asked a spokesperson for Time-Warner, “can a media empire, which often asserts a legal right to unrestricted access to information, deny such rights to scholars?”) When I refused to go “off the record,” everyone clammed up. We finally printed a second version of Past Imperfect, minus the stills of the Warner movies.

Five months later I learned exactly what happened at Warner from an unexpected source: Oliver Stone. Several weeks after Past Imperfect had appeared in print, Stone faxed sharp criticisms of Karnow’s essay on JFK (“Stanley Karnow considers himself an historian”). I, in turn, complained of Warner’s withholding of the film stills, and the studio’s subsequent conspiracy of silence. “A great media/news empire must not engage in censorship,” I pontificated, “nor throw a curtain of secrecy over its own actions.”

Stone said he knew nothing of the Warner decision and, though in the midst of the final edits of Nixon, he immediately sent a letter to Judith Singer, director of “licensing/theatrical/legal” at Warner. Stone faxed me her response. Singer, who perhaps assumed that her letter would not become public, confirmed my worst fears about studio censorship: she explained that the requisite permissions had been denied because several Warner films were “treated negatively” in Past Imperfect. This was a matter of general policy, she added, and had been submitted for review in this instance to top studio executives. She explained that the studio undertook such policies “out of concern for our filmmakers.”

Appalled, Stone rebuked the Warner executives. "I make films like JFK and Nixon to stimulate discussion of the past. I expect and even welcome criticism, and am frankly pleased when writers refer to actual stills of my films rather than rely on their subjective recollections of them. Please, in the future, do not justify censorship out of some imagined concern for my feelings.”

The selectively restrictive policies of the studios have recently been the subject of considerable discussion among scholars. Jon Wiener described in the July/August 1994 issue of Lingua Franca how researchers seeking access to Disney archives were obliged to agree to submit “any text which relates to the Walt Disney Company” for “review” and “corrections.” Last year Lingua Franca published an article on film and picture permissions whose thesis was contained in its title, “Just Do It,” a recommendation that makes considerable legal sense (the use of one or two stills out of the tens of thousands in a feature film doubtless falls within fair-use guidelines) but with which the legal departments of major publishing houses rarely concur. Without some modification of copyright law, or an expansive judicial ruling on the doctrine of fair use as it applies to researchers and film stills, scholars will be unable to engage readers in a thoughtful consideration of the meaning of perhaps our most important visual artifacts.

The studios' chief defense against the "encroachment" of researchers–the need to protect the fragile sensibilities of their artists–is refuted by the example of Oliver Stone, who has received perhaps more criticism from scholars (including me) than any other filmmaker, and yet remains committed to unfettered scholarly access to film stills and studio archives. "I fought for public access to the secret Warren Commission Archives, and to the Nixon Watergate tapes, and I'll fight for uncensored access to the film stills that comprise an important part of our cultural heritage," he told me. On this issue, at least, we historians can—and should—make common cause with Oliver Stone to rescue these moving pictures—our memories!— from corporate control.

—Mark C. Carnes is professor of history at Barnard College and general editor of Past Imperfect: History According to the Movies (1995).


1. Francis A. Yates, The Art of Memory (1966), 4, and Jonathan D. Spence, The Memory Palace of Matteo Ricci (1983).

2. E. A. Kirkpatrick, “An Experimental Study of Memory,” Psychological Review 1 (1894): 602–09.

3. L. Standing, “Learning 10,000 Pictures,” Quarterly Journal of Experimental Psychology 25 (1973): 207–22; L. Postman, “Verbal Learning and Memory,” The Annual Review of Psychology 26 (1975): 322. See also Allan Paivio, T. B. Rogers, and Padric C. Smythe, “Why Are Pictures Easier to Recall than Words?” Psychonomic Science 11, no. 4 (June 5, 1968).

4. Patricia Baggett, “Structurally Equivalent Stories in Movie and Text and the Effect of the Medium on Recall,” Journal of Verbal Learning and Verbal Behavior 18, no. 3 (June 1979): 333–56.

5. The screenplay appears in Eric Hamburg, ed., Nixon: An Oliver Stone Film (1995); quotes from pp. 88–89, 91.

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. Attribution must provide author name, article title, Perspectives on History, date of publication, and a link to this page. This license applies only to the article, not to text or images used here by permission.