One of the reasons I have selected Persona and Three Women for extended discussion is that both films represent considerable creative leaps for their directors-—important, and even courageous, breaks with their pasts. I would like to discuss each film more fully in terms of the artistic development of the filmmakers.

Both Bergman and Altman began their careers very firmly rooted in the traditions of narrative film. Bergman’s little-known films of the 1940s, for example, owe a great debt to the Italian Neo-Realists. Works like Ships to India (1947), with its stereotyped story of misunderstood youth, or Port of Call (1948), in which the reasons for an attempted suicide are laboriously examined, have almost nothing in common with the great works their author later created. Bergman has always been interested, one might also say obsessed, with themes involving identity, and especially the identities of female characters, but an early proto-Persona like Summer Interlude (1950) resolutely stayed on the surface of its narrative about a ballerina tormented by approaching middle age and the inability to perform.

In looking at the more than thirty films Bergman has made in the last thirty-three years, we can witness an artist in the process of discovering himself and his powers. He was not a natural filmmaker, as Fellini might be said to be. Bergman began with studied exercises in social commentary; he borrowed awkwardly from the Neo-Realists; and then he spent the 1950s developing a distinctive style and voice that was sometimes masked by his too-willing use of striking symbolism. The chess game with Death in The Seventh Seal (1956), the old professor’s nightmare in Wild Strawberries (1957), and the attic apparitions of The Magician (1958) will always remain among his most famous scenes, but they are the easy way out when contrasted with the depth, complexity, and emotional power of his consideration of death in Cries and Whispers (1972), the dreamlike night in Persona, or the nightmarish imagined visions of the schizophrenic girl in Through a Glass Darkly (1961).

Bergman had to leave his facility for striking symbolism behind if he wanted to penetrate more deeply into the themes that obsessed him, and he did, starting with his “death of God” trilogy of 1961–63 (Through a Glass Darkly, Winter Light, and The Silence). The middle film was one of the most visually barren and striking he ever made, with the bleak coldness of its little Swedish Lutheran church in winter and the starkness of the office in which a spiritually impotent minister tried to counsel a suicidal parishioner. The trilogy must have been a spiritual cleansing for Bergman, because after it he left behind, as a subject, his concern with God’s apparent silence. He also left behind (with the singular exceptions of the unsuccessful comedy All These Women in 1964 and the black-magical scenes in Hour of the Wolf in 1968) any desire to deal with his human subjects except head-on. Henceforth he was to be seen gravely regarding them as individuals, not as elements on a canvas that could also include stylistic symbolic props.

Persona was the breakthrough film. It was almost as if he had found the courage to leave behind the trappings—the medieval rituals, the devils, the flamboyant props and fantasies he had heretofore employed—and was determined to deal with his characters unadorned.

I asked him about that when I met him in 1975, during the filming of Face to Face. He had a little room, long, dim, and cool, like a monk’s cell, across the hallway from the sound stage he was working on in Stockholm’s Film House. It was furnished simply with two chairs, a cot, and a table (upon which rested two apples, a banana, a box of chocolates, and the script). When he was not sure about how a scene was going, he said, his practice was to come into the room, lock the door, and lie down on the cot until the scene was clear in his head.

“What has happened in my recent work,” he said, “is that I’ve grown up. I’ve matured. I’m more interested now in the people themselves. If I believe in anything, I believe in the sudden relationship, the sudden contact between human beings. When we grow up, we suddenly feel we are completely alone. We find substitutes for loneliness—but this feeling of a certain contact, a certain instant understanding between two people, that’s the best thing in life. It has nothing to do with sex, by the way.”

That “sudden relationship” perhaps explains what is seen to happen in the final shot of the dreamlike bedroom scene in Persona. It can also be seen in Cries and Whispers, when two sisters (Liv Ullmann and Ingrid Thulin), long at odds with one another, suddenly and compulsively begin to stroke each other’s faces while exchanging words of endearment. And also at the end of the six-part television serial Scenes From a Marriage (1974), when the man and the woman (Erland Josephson and Miss Ullmann), separated and divorced for so many years, decide on a whim to spend a night together in a friend’s cottage, “a cottage in the middle of the night somewhere in the world.”

But what can be seen at the same time in Bergman’s visual style, and especially since Persona, is his characteristic use of a particular kind of “two shot” (the cinematographer’s eminently sensible term for a shot containing two people). Again and again, so frequently that it becomes his motif, Bergman places two people within the frame but does not permit them to look at one another. Both look away, or one looks at the other, who looks away. The moments of actual visual contact in Bergman’s films, even those of his earlier years, are astonishingly rare. An underlying tension becomes apparent when all of his films are looked at in sequence, or when the films from Persona on are examined with particular care: The characters cannot, or will not, or fear to, communicate openly with one another. The barriers are always there, but then they break down in a rush of vulnerable, even confessional, emotion.

This sort of tension does not lend itself to conventional narrative structure, and especially not to the sort of storytelling to which the commercial cinema usually limits itself. If characters cannot communicate, that is one thing; but if they’re forced to inhabit stories about their lack of communication, there’s the danger of self-defeating boredom, of movies trapped forever inside a Jules Feiffer cartoon. Having abandoned the symbolic stage-trappings, Bergman had to find, to invent for himself, a valid way of dealing with “the agony of the couple,” as he called it. He did so, as I have suggested, by abandoning paraphrasable story lines, by throwing conventional narrative overboard, and by moving instinctively into the worlds of pure emotions, sensations, and dreams.

“No other art-medium—neither painting nor poetry—can communicate the specific quality of the dream as well as the film can,” he told his interviewers in Bergman on Bergman. “When the lights go down in the cinema and this white shining point opens up for us, our gaze stops flitting hither and thither, settles and becomes quite steady. We just sit there, letting the images flow out over us. Our will ceases to function. We lose our ability to sort things out and fix them in their proper places. We are drawn into the course of events—we are participants in a dream.”

Altman, of course, did film his dream. (“Sometimes while I’m dreaming,” Bergman mused a bit later in the conversation quoted above, “I think: ‘I’ll remember this, I’ll make a film of it’—it’s a sort of occupational disease.”) Like Bergman, though, it took Altman many years to arrive at such a point, to seriously consider a film totally set free from narrative structure.

If Bergman is an artist who had to learn, whose early work does not easily hint of the greatness to come, Altman certainly is as well. All of his beginnings were firmly within conventional narrative cinema—and with a vengeance, because he spent the better part of his early career directing industrial and educational films in Kansas City. He made an attempt to break into Hollywood in 1955, when he was thirty years old, with two very easily forgotten films, The Delinquents (1955) and The James Dean Story (1957). Defeated for the moment, he went back to Kansas City and then allowed himself one more chance to break into fiction film direction. He went back to Hollywood in the early 1960s and directed dozens of segments of such television series as “Combat.” Those segments that can be glimpsed on late-night reruns betray few signs of an individual directorial style; but then, television by its nature tends to force directors into the overuse of closeups, medium shots, and narrative rhythms that can tell a story quickly and simply.

When Altman finally did get another chance at a theatrical feature, he was already forty-three years old and, presumably, well accustomed to working within conventional narrative forms. That is certainly the conclusion to be drawn from his rarely seen 1968 feature, Countdown, which was all but withdrawn from commercial release after a disastrous New York opening. (Howard Thompson’s review in the New York Times described it as, among other things, “simply stultifying, slack, cliché-ridden, listless, and dreary.”) Warner Brothers cut it down from 101 to 73 minutes for its brief British release, and the shorter version occasionally turns up late at night on television. I have, however, seen the 101-minute version (and even it was cut by the producer, William Conrad, after Altman’s own final cut). Countdown might be said to be to Altman’s career what the unhappy It Rains on Our Love (1946) was to Bergman’s: evidence that it occasionally pays to persevere on the course one has charted, no matter how discouraging things seem at first.

It is not necessary to discuss the plot of Countdown in any great detail (it has to do with the first landings on the moon). What is to our purpose is to notice that Altman directed it in a perfectly straightforward Hollywood narrative style. There is only one scene, at a party, that employs what would later be called the “Altman sound,” the characteristic use of multiple recording tracks to obtain realistically overlapping dialogue. Everything else in the movie could have been directed by any other graduate of the television series of the 1960s. Altman’s next film, That Cold Day in the Park (1968), is more interesting (it tells of a repressed woman who establishes a pathological relationship with a younger boy), but, again, it is essentially a conventional film.

Altman’s career was properly launched with M*A*S*H, in 1970, and it was here that Altman first found his voice and style. Through a coincidence (I was writing a screenplay at the same studio at the same time), I was able to read Ring Lardner, Jr.’s original screenplay for M*A*S*H—a screenplay for which he later received an Academy Award, though Altman’s actual film uses only its bare bones. What Altman found at last in M*A*S*H was his special feeling for milieu and community. The camera dives into the middle of its muddy mobile field hospital in Korea, and we get to know the characters much as if we had just arrived ourselves. Familiar faces keep turning up. People refer to each other by name until we begin to remember their names. Action in the background helps to establish later action that will take the foreground.

Altman’s notions about communities have been his focal points ever since (with the singular exception of Images, which was anti-community in the sense that its schizophrenic heroine could never be quite sure who was really there in the room with her). Characters whirl about one another in Altman films. Relationships are shown, and are taken for granted, without having been really established. People meet each other during the course of a film and get on better, or worse, as the story advances; in either case they may not be major characters but simply “atmosphere,” as Hollywood calls extras who have dialogue (an example would be the interplay when the country music stars visit the local musicians’ hangout in Nashville and wait somewhat uncertainly to be introduced from the audience). Since M*A*S*H, Altman has apparently not conceived of films in such simple terms as a foreground story and the background needed to sustain it. Instead, he has taken as his subject intriguing groups of people and then has plucked out as his protagonists a few representative examples. This approach to scenario is opposite to the vast majority of feature films made in Hollywood or anywhere else, and it is Altman’s personal solution to the inhibitions of conventional narrative.

We can see this approach developing in the films after M*A*S*H. Altman handles it awkwardly in Brewster McCloud (1970), with its strange tale of a young boy who lives in an unmarked room in the Houston Astrodome and is tutored by a fallen angel as he tries to build wings that will let him fly. Altman, who stands behind all of his later films, insists that Brewster McCloud rewards additional viewings, but it has not rewarded mine. Yet it does, like his other films, occupy the center of a community of people, of purposes that are common or crossed. It is not just the story of the boy but of the immediate society which has to decide what it thinks of him. McCabe and Mrs. Miller, discussed above, is about community almost to the exclusion of narrative. McCabe walks into a smoke-filled frontier barroom with its half-heard conversations, its laconic asides, and becomes one of its regulars. As he attempts to become an entrepreneur through the construction of a bordello and the importation of Mrs. Miller and her troupe of prostitutes, the community carries on its daily affairs all around him. Church services are held. Community baths are taken. When a young boy is shot dead because he was unfortunate enough to find himself crossing a footbridge at exactly the same time as a hotheaded young killer, it is the whole community that absorbs the event, and mourns it. California Split (1974) is at pains to place its two compulsive gamblers within a clearly seen gambling community. Altman’s stylistic approach is especially evident here in a scene where the protagonists are shown into a room where a high-stakes private poker game is in progress. Murmurs on the sound track introduce the players, whose long-standing rivalries and friendships are taken for granted: We arrive, as we so often do with Altman, in medias res.

Altman’s most free, ambitious films about communities come in Nashville (1975) and A Wedding (1978). In both films he almost seems to be testing his limits, daring himself in an attempt to get as many recognizable characters as possible on the screen, and to keep them all established. Nashville was not popular in Nashville, where the residents felt the film did an injustice to the city and its primary industry, country and western music. But Altman was not doing a documentary on Nashville; he was using the title, I believe, as a specific for community. And during a mobile era when communities are reformed year by year, he was concerned not so much with the roots of the city and its industry as with the many kinds of people who found themselves drawn there for socioeconomic reasons of the moment. A presidential campaign is in town, and so are several top country singers, back from national tours for recording dates and the Grand Ole Opry. But the city’s dynamic has also attracted such characters as a would-be assassin; a man grieving for his dying wife; a star-struck young woman from the state’s rural areas; her jealous husband; two groupies (one a young soldier, one an eccentric girl from California); an Englishwoman pretending to be from the BBC; a waitress and her man friend; visiting celebrities (Elliott Gould and Julie Christie play themselves); a campaign manager and his advance man; and even an inexplicable magician on a motorcycle (who seems to be an American cousin of the anonymous cyclist who roars at random through many of the scenes in Fellini’s Amarcord).

Few of these people know each other when the movie begins, and some will not have met when it ends. But many of their paths will cross, their lives will affect one another (if only at two or three removes), and Altman seems fascinated by how this interaction takes place. There is no central character in Nashville, no person whose fate is more important than another’s, and in this respect Altman was wise to leave the presidential candidate offscreen entirely. If he had brought him on and then had him assassinated, the movie would inevitably have been categorized as “about” political assassination, and the real subject, the haphazard and human interactions of a community, would have been missed.

These larger communities are distilled into the limited and very particular community of Altman’s Three Women. If in his other films he dared to overpopulate the screen, in Three Women he limits himself so severely that many of the actors seem to play more than one character; it might be said that many of the characters seem to contain more than one person. I have mentioned the many kinds of female roles that the three women play at one time or another, adding up almost to a catalog of the most obvious personalities women can offer up for the examination of society. I now want to emphasize Altman’s creative courage in making the film. Nashville and A Wedding (and the less successful Buffalo Bill and the Indians) free themselves from conventional narrative but not from the chronological passage of time. Three Women is altogether more mysterious, turning inward, turning back on itself, allowing its characters to exchange or alter their basic natures. It is free of all narrative conventions except those such as that sentences must begin before they end. The film is an immersion in several personalities, a turning here and there, from a point of view inside the narrative. In fanciful imagery, Three Women is the story of a human soul or consciousness peering through the many windows it finds opening onto an inconsistent but apparently real universe outside.

It is most unlikely, of course, that Saturday-night moviegoers emerge from the theater with such an interpretation in mind, but that is not the point. The viewers most likely to be frustrated or unhappy at the end of the film are the ones who insist that it “make sense”—who require a paraphrasable narrative. Those who allow the film to happen to them in the ways discussed at the beginning of this essay are much more likely to enjoy it, although they would probably be just as hard-pressed to describe what happens in it as the more literal and didactic members of the audience would.

To return to the notion of Three Women as an objective representation of a soul/consciousness peering through windows into the world: That is what all of us found ourselves doing at birth, we do it every day, and some of us are really good at it. It is our habit, though, to do the peering ourselves, and to expect that the things we see outside the windows will be objective and consistent enough to permit us to weigh, touch, smell, or hear them, and arrive at decisions about them. Traditional narrative films are objective and consistent, and therefore often eminently satisfying to our expectations. But the film medium is so fluid, flexible, and complete that it makes a altogether different kind of experience possible if filmmaker and filmgoer conspire to let it come about. The capacity of film to be escapist, voyeuristic, sensual, nonverbal, and encompassing is also what allows a film to imply another soul/consciousness looking out through its windows at ourselves. And it is just here that two minds can touch.

Three Women and Persona are not on the screen to be looked at objectively, and they are not surrogate reality. Through the genius of their makers and the boldness of their designs, they are themselves self-conscious. Permit me a fancy: They are looking back at us.

It is not my purpose here to sound like a mystic who has been to the mountaintop and been shown the truth. I hope I continue for many years to enjoy films that are cheerfully organized around an old-fashioned narrative. I have tried to argue, however, that the two films discussed at length represent moments in the careers of their creators when freedom, real freedom, was found from the restrictions that bind almost all filmmakers and almost all their films.

The freedom to go beyond narrative and engage other human personalities at a level different from the literal and the linear is one that music, dance, sculpture, and painting have always shared. The same freedom also exists in literature, particularly in poetry, but it does not come so easily there, because the word and the sentence have an innate tendency to bring matters back to the particular. It is unfortunate that the cinema, since its earliest days, has drawn more readily from stories, novels, and plays than from other artistic forms (for film, as Kael reminds us, is, along with grand opera, one of the two great bastard arts). Perhaps that is what they are trying to tell us, those die-hard purists who believe silent movies were the art form in its most pure state, and that film sold its soul for sound: That without the words, without that damnable dialogue, you had to tell your story with the visual tools that made movies different from books.

Painters made their fundamental discoveries centuries ago; filmmakers are still making theirs today. The freeing of painting from subject matter—nonrepresentational, abstract art—has already, of course, been reflected in countless avant-garde films. But in terms of how we still look at mass-market narrative movies, a more useful parallel might be drawn between the state of feature films today and the development of attitudes toward art in the Renaissance. Paintings then were supposed to be about something, to illustrate, to lecture, to instruct the common people about the glories of God, the doge, or the Medici. The painters themselves knew better. There is the famous story of Veronese, unveiling his “Last Supper” with its sacrilegious supporting cast of dogs, monkeys, and Germans, and ordered by the Venetian Council of Ten to alter the painting or be put in prison. Unveiling the same painting the second time, Veronese explained that it had indeed been altered: Its title was now “Feast in the House of Levi.” The movie audience today is a mass Venetian tribunal, demanding, whether consciously or not, that movies mean something, and say what they mean, and look as if they mean it.

The irony here is that film is not as effective when it traffics in ideas as when it deals in, and with, emotions. The straightforward consideration of ideas is not as interesting in film as a more oblique approach in which the ideas, whatever they may be, are given emotional content. Persona and Three Women are obviously filled with ideas that Bergman and Altman have about the nature of character—but at no point does either film refer directly to those ideas. Instead, we are invited to feel what the directors feel, to share their observations, and then to arrive at our own conclusions through the emotional experiences that have been provided for us.

I discussed earlier the rather obvious differences between On the Beach, which warns us of nuclear peril, and Dr. Strangelove, which, by making us laugh, employs our laughter as its warning. The most numbing moments in cinema come when a character is made to turn to the audience and give us the dialogue containing the director’s message. Frank Capra found such moments obligatory, and yet they are the very moments when his populist entertainments go flat.

Jean-Luc Godard finally became paralyzed with his messages. When two proletarian garbage collectors face his camera and give us their interminable Godardian/Marxian analysis of capitalist society, we squirm. The same criticisms are made more effectively by purely visual means in Weekend (1968), in the famous tracking shot in which bourgeois society is seen at a dead standstill in a traffic jam that goes on for miles.

Godard’s long speeches were defended by critics who said the very length of these speeches, which were filmed in unbroken closeup, were intended as the director’s ironic commentary on ideas in movies. But, alas, that turns out not to have been the case. One of the saddest of recent films is Godard’s Numéro Deux (1975), in which the director stands in an editing room, operating a movieola or videotape cassette system on which we see footage involving work, families, and sex in contemporary France. The images are rarely allowed to take up as much as half of the available space on the screen, and the rest is either darkness (making us conscious that we’re watching a movie) or other, smaller images. The film’s very long prologue shows Godard, in shadow, standing in his studio next to a television set, on the screen of which we see him addressing us. He is placing himself at two removes from Warshow’s “immediate experience,” and this scene is, for me, a record of Godard going off the deep end. He speaks of his ideas for perhaps ten minutes, but with the exception of a few scattered observations he never says anything that seems even vaguely sensible. It would be charitable to say he is free-associating; in fact, he’s raving.

My notion is that Godard arrived at this point by determining to convey his political positions in a medium not suited to abstract ideas. If we want to reason, debate, discuss—if we want to marshal ideas and apply them to the discipline of logical argument-—the medium in which to do so has existed for a very long time, and it is the written word. But if we want to convey the mood of a time and a place, the look of a setting sun, the subtle flow of feelings across a human face, the way it was to be there and experience it (whatever it was), the medium is film.

That is not to deny that many films, many good ones, have existed primarily as vehicles for ideas. From The Great Dictator to Medium Cool, and in If..., Z, The Battle of Algiers, Burn!, Patton, and many more films that come to mind, the director’s primary purpose has been to make us think (and, usually, to think as he does). But these films are good not because of their ideas but because of the artistry at the service of the ideas. To praise a film because it is “correct,” or attack it for being “wrong,” is to miss the point: Even in idea-oriented films, the ideas serve primarily as the occasion for a film’s artistry, and the film’s success or failure will take place apart from them. If a film can make us feel well and deeply, it need not make us think. Or, more correctly, if a film can make us feel, then it can inspire us to analyze our feelings, and perhaps then come to a deeper understanding of their sources, powers, and (yes, now, in this context) their meanings.

In suggesting that the Saturday night audience can hardly be expected to arrive at conclusions about Three Women similar to those we have discussed here, perhaps I was implying something more. I was not, I hope, hinting that the film’s audience was somehow dense and unable to experience such a film. I meant that few works in the experience of contemporary moviegoers can have prepared them to cope with films like Three Women and Persona on the films’ own terms. It is one of the peculiar burdens that serious filmmakers and critics have to bear that mass audiences will cheerfully give themselves over to the sensual delights of Saturday Night Fever, Star Wars, or Jaws, but are wary of granting the same abandon to “art” films. Because films like Star Wars give such sharp, immediate, and rewarding emotional experiences, their audiences rarely think to ask what the films might mean, if anything.

It is true, of course, that Persona and Three Women also provide sharp, immediate experiences (or at least they did for me). But I think their experiences are of a different level. Star Wars has as its ambition the creation of escapist experience for its own sake: If the movie makes us feel, if it entertains and distracts us, it has succeeded on its own terms. But the experiences we discover in Bergman or Altman are not so immediately rewarding; they raise questions, sometimes quite disturbing ones, and they will not simply allow themselves to be consumed. For those who find such questions challenging, Three Women and Persona can be delightful—as puzzles, personality mazes, touchstones of the directors’ own personalities. For those who want only to escape, this more difficult form of experience will be harder to assimilate.

Perhaps I am talking about two kinds of audiences—or moviegoers who will want to join different audiences and different collective states of mind for different movies. The mass audience still seeks experience, pure and simple. I remember a practice that was popular during the reserved-seat engagements of 2001 in the late 1960s. Lovers of the film would mingle with the intermission crowds on the sidewalk, then slip inside to rest flat on their backs on the floor in front of the screen, the more thoroughly to enjoy Kubrick’s space-and-time “light show.” This is consuming a movie in the same way one might consume a roller coaster; while the experience is undeniably entertaining, it is not the way I would suggest to deal with a serious film.

There are, however, great and serious films that require something of the same willing giving over of moment-to-moment consciousness from the viewer to the filmmaker. The better these films are, and the more central to shared human experience, the more readily some filmgoers seem to shy away from them. Ingmar Bergman’s Cries and Whispers (1972) is a case in point. I thought it was the best work of the year and said so in an annual review of the year’s films. A reader called to ask: “But should we go to see it?” “Well, yes, of course,” I said. “My friend and I were thinking,” the reader said, “that if it’s really that good, then maybe we’d have more fun going to see something else….”

How does a critic build bridges between what is new, best, and most daring at the movies, and the built-in desire of the mass audience to see the kinds of movies it has known best and longest—and can depend on? Where do the two audiences meet? The daily newspaper reviewer is faced with this dilemma more frequently, and more bafflingly, than writers for audiences who have already made part or all of the journey to those lands where the best new movies reside. There are a great many people for whom going to the movies still means a decision in favor of the new Clint Eastwood film instead of, say, the new Charles Bronson film. There are those who would rather see Saturday Night Fever ten times than see Saturday Night Fever once and then see nine other films.

I do not mean to reject filmgoing on that level, but I do want to insist that the most original new work will not be found there. It is fine with me if there are two, ten, or a hundred cinemas, but I think we have to understand that the most important new movies will not be coming from the directors who make better and better films in conventional narrative modes, no matter how much we may admire and enjoy what they accomplish. The key films of the coming years, whether or not they are immediately (or ever) successful, will be the ones that explore and try to understand the powerful three-way connection between cinema, emotion, and the mind.