12

Psychiatry of Character Disorders—Part I

► Dialogue

Well-written dialogue evokes a sense of reality, even though it is, in essence, unreal. Where people in everyday life may speak on and on, in movies their dialogue usually confines itself to one or two sentences at a time. In a sense, dialogue is more about the listener than the speaker, since most of us, most of the time, don’t listen well to anything that doesn’t immediately impact us. Tell someone about how your day went, a dream you had, or what you did on vacation, and his attention drifts in and out. Tell that same person that the tests came back positive and he only has three weeks to live, and you have got his full attention.

With dialogue we become distressed when too much information is crammed into one character’s speech. As an audience we prefer to discover the betrayal, the attraction, the warning through the back and forth pattern of conversation. In editing we also want to reinforce that enjoyment of making this discovery.

In the area of dialogue cutting, digital nonlinear editing has produced several surprising developments. Even though other aspects of editing, such as action cutting, have also experienced a paradigm shift brought on by the digital dawn, their evolution has proceeded pretty much as could be expected. In the section on action editing we saw how instant access to diverse takes, a highly effective trimming system, and the ability to mock up effects, such as motion ramping and greenscreen compositing, have kicked down the doors of the traditional action sequence. It has fostered even faster chases, more daring escapes, and heightened devastation. But in the realm of dialogue editing, the progress has been subtle, though perhaps of greater significance.

In the days of editing on film, a common deterrent to a well-cut dialogue scene was the editor’s unwillingness, uncertainty, or incorrectness in determining the overlaps. In a dialogue scene, an overlap occurs when it becomes necessary to shift the audience’s attention from the speaker to the listener. At the moment when Character A proclaims something of importance to Character B, the audience wants to see Character B’s reaction, even while Character A is still speaking. What occurs in that character’s reaction often tells us more about the scene and about the character than what the character has stated. If the character says, “Luke, I am your father,” we want to see Luke’s reaction.

The timing of the overlap with the reaction is crucial to the audience’s understanding of the scene and the characters’ involvement in it. While action scenes supply the excitement and thrills in many films, dialogue scenes reveal the deeper meaning. To understand how film dialogue communicates meaning, we first need to examine the value of subtext.

► Subtext

Good writing, like good editing, is about relaying complex and even conflicting information in a subtle and interesting way. This is generally the realm of subtext. In the case of anthropology, the concept is known as metacommunication, the visual and auditory signals that we telegraph to each other—wittingly or unwittingly—that underlie the stated message. The Invention of Lying (2009), a comedy about a world where everyone says exactly what is on his or her mind, intrigues us because it runs counter to typical social interaction. In most cases, even a simple “Hello” can project all sorts of complex or hidden messages, as reflected in Dorothy Boyd’s classic line from Jerry Maguire (1996), “You had me at hello.” Suffice it to say, how an actor delivers a line, how the other character reacts to the line, and how the editor places it in the scene are the essence of effective dialogue.

Take another simple greeting that projected a much larger meaning in Hotel Chevalier (2007). Jack Whitman reclines indolently on the king-size bed in his Parisian hotel room when the phone rings. In the wide shot he puts down his newspaper and picks up the phone. “Hello?” he says. A female voice on the other end says, “Hi.” As the single syllable crosses the cut from the wide shot to Jack’s tight profile, the drastic change of angle signals us that the caller holds far greater than passing significance for him.

The end of the Diane Keaton/Mandy Moore romantic comedy Because I Said So (2007) exemplifies the use of subtextual dialogue to subtly communicate meaning. In it Diane Keaton plays an “overprotective, overbearing, over-the-top mother” and caterer with three grown daughters. Determined to find a good husband for her youngest daughter Millie, played by Mandy Moore, she resorts to all sorts of meddling, including interviewing perspective husbands whom she recruits through an internet dating service. The story’s middle revolves around two particularly eligible suitors, a successful and charming architect and a kind, devoted musician with a young son and little money. The audience senses that Millie really loves Johnny, the musician, even though her mother would prefer to see her with the wealthy architect. Eventually she falls for Johnny, while maintaining the relationship with the architect to satisfy her mother. Everything goes fine until both relationships fall apart.

In the end, a heartbroken Millie is again alone, conducting a cooking class for the love-lost geriatric set. As she explains, “When you’re cooking for one, it’s important to look forward to the end result.” At that point an off-camera voice, which we recognize as Johnny’s, interrupts to say, “What if you want to make it for two?” The scene cuts to Johnny standing behind Millie’s students. Then it cuts back to Millie as she replies, “It’s a little more complicated, but it can be done.” After that, Johnny approaches her, and they embrace and kiss. Though more overt than the subtext in a dramatic scene, this dialogue displays an amusing and basic use of subtext. Rather than Johnny appearing and professing his love to Millie—perhaps while admitting he was wrong to leave her even after she’d admitted her love for him and explained why she’d been with the other guy whom she didn’t care about—it’s more satisfying to use the cooking analogies. This allows the audience to participate in discovering the meaning beneath the lovers’ statements. “Cooking for one” obviously refers to Millie’s lonely life. Here she is even teaching a cooking class to elderly people who have grown old perhaps without finding Mr. or Mrs. Right. Johnny’s question about cooking for two offers the possibility that he and Millie might be together. Millie’s final response, “It’s a little more complicated, but it can be done,” acknowledges the difficulties of being in a relationship but her awareness of its value and her willingness to commit to Johnny.

One of the best examples of good writing and dialogue editing resides in the Mike Nichols LGBT comedy The Birdcage (1996), based on the play La Cage aux Folles and edited by Arthur Schmidt. Here, the editor’s use of reaction shots injects the subtextual dialogue with added meaning. In the long but compelling climactic scene (Figure 12.1), every character has something to hide. This scene epitomizes a wonderful exaggeration of life’s subterfuges and stands out for its use of subtext and reaction.

Robin Williams and Nathan Lane play a gay couple, Armand and Albert, who own a thriving drag club in Miami’s South Beach. The film opens with a musical number performed to the pointed lyrics of “We Are Family,” before introducing Albert in the guise of the drag queen star, Starina. Initially we are introduced to the flamboyant, fussy, and colorful life of this successful gay couple. They’re not prepared for the announcement by Armand’s son Val—a child he fathered 20 years ago during a brief heterosexual relationship—that he’s getting married—to a woman. And not just any young woman; he’s marrying the daughter of the ultra right-wing Senator Keeley, played by Gene Hackman, cofounder of the Coalition for Moral Order. We learn that he has no sympathy for homosexuals, Jews, or liberals. Armand and Albert fit all three categories. This forms a perfect dramatic, as well as comedic, setup, bringing together characters from opposite ends of the social spectrum.

Figure 12.1 Hosting dinner guests in The Birdcage (1996)

Photo courtesy of United Artists

As the story progresses, Val pleads with his father to help make the meeting of the two families a successful one. In his mind that goal can be accomplished by portraying their household as a typical God-fearing, heterosexual American home, supporting the values that the senator holds dear. But at this point in the story a dark spot has already appeared in the senator’s half of the yin yang. His colleague, Senator Jackson, has been found dead in a prostitute’s hotel room. The Republican contingent, of whom Senator Keeley appears to be the most outspoken, is busy doing damage control. The senator welcomes the opportunity to get out of town with his wife and daughter and drive south for a dinner with his prospective in-laws in Miami.

As the senator makes his escape, avoiding reporters waiting in the snow outside his home, the action cuts back to Armand, sweating in the heat of South Beach, as he agrees to compromise for the good of his son. He’ll portray himself as a typical male dad married to a typical female mom. He even goes as far as to enlist the help of Val’s biological mother, whom he hasn’t seen in 20 years, ever since the ill-conceived tryst that led to the boy’s conception. Armand’s live-in lover, Albert, who raised the boy, grudgingly agrees to make himself scarce and to allow Val’s birth mother to take his place at the dinner. Further, Armand removes all suggestive objects, wall hangings, and books from their home, creating a sparse, spartan-like environment centered on a large cross hanging from the wall, despite the fact that he is Jewish. He has even agreed to alter the pronunciation of his last name, Goldman, to the gentile Coleman. All in all, the preceding action is a brilliant setup for the mayhem that will follow.

In the climactic dinner scene, the conflicting elements—liberal/conservative, gay/straight, loose/uptight, cold/hot, muted/colorful—all come together. As the evening proceeds, with the forces of deception hard at work, the tension burrows under each character’s calm facade, and the audience delights in knowing what others do not. This knowledge, fueled by the reactions of the characters, unleashes an outrageously comical, barbed, and enjoyable experience. Early in the dinner scene, a long and rambling speech by the senator about his car ride from the north to the south highlights the awkwardness and distance between the two families. It is told mainly in cutaways to the reactions of his listeners. The actual dialogue is neutral and prosaic, but the characters’ reactions tell an oblique and compelling story.

As the dinner proceeds and the senator pontificates about his concern for upholding family values, various incidents threaten to upset the well-planned evening. The pretend wife gets caught in traffic and will arrive late, the drag queen-turned-butler stumbles around and screws up the cooking, and reporters from the National Enquirer figure out where the senator has escaped to and follow him. At each tear in the fabric of deceit, the audience shares in the delightful suffering of the characters. At one point in the dinner, the quasi-butler has neglected to observe the dinnerware and serves his inedible soup in bowls decorated with Greek images of young men copulating. We never see what hides in the bowl, and the senator, having misplaced his glasses, thankfully can’t make out the image. The audience, however, realizes what is depicted beneath the puddle of murky soup, based on the reactions of Armand, his son, and the son’s fiancée. When Albert, who’s been sequestered in another part of the house, can’t stand it any longer and finally prances onto the scene dressed in full drag, portraying Val’s mother, another round of humorous reactions ensue. As Albert exuberantly introduces himself, even correcting the senator who calls him Mrs. Coleman rather than Goldman, Armand, the son, and the future daughter-in-law struggle to maintain the deception. Throughout the sequence the strategic use of overlapping dialogue and reaction shots creates the subtext that fuels the humor.

The Overlap

But how does the editor know when to cut away from one character to another and how to overlap one character’s dialogue over the reaction of another character? For many, this process has evolved as a somewhat arbitrary decision. Convinced of the necessity to supply overlaps in order to maintain a scene’s momentum, inexperienced or less expert editors will determine these moments based on whim, chance, or guess. The dictum to overlap is so deeply ingrained that many will not proceed to the next line of dialogue without first considering an overlap. Yet this is one of the primary ills that film doctors encounter. In previous forms of linear editing, the condition was exacerbated. In the case of videotape, as mentioned previously, it was necessary to build the cut by laying down one shot at a time. This was accomplished by playing back the desired shot so it could be recorded onto a fresh roll of tape encoded with SMPTE timecode. Since the audio and video tracks were laid down separately at predetermined timecode spots, it gave the editor the option of inserting the video at a different position in relation to the audio, creating what is known as a split edit or L-cut. The tendency for editors, producers, and directors working in this manner to make paper cuts on a notepad and then build the edit based on timecode notes meant that overlaps were built in as the cut developed.

Moviola editors working in film had a slightly different workflow, more akin to what would become known as nonlinear editing. The Moviola editor viewed individual takes by running individual rolls of workprint and soundtrack through the machine, marking the selected areas, and then cutting the film and attaching it to the previous shot on the reel. As mentioned earlier, experienced editors actually paperclipped the selected picture and track sections together to be assembled later by an assistant. Only after the film was physically spliced together did the editor see the outcome of his decisions. The discipline required a mindset similar to that of a chess player calculating many moves ahead so as not to fall into checkmate.

To save time and to avoid making too many physical cuts in the film, many Moviola editors worked with a synchronizer by their side, building in the dialogue overlaps as they went. Likewise with the linear-based KEM editor who was inclined to take a similar approach. With the introduction of digital NLE systems, it became even easier to build in overlaps along the way. But some of the best editors knew that the most effective way to cut overlaps on the first pass of a dialogue scene was to not do so at all.

Why is this? Because overlaps supply emotion, feeling, and response. How can an editor accurately gauge the feeling of a scene and, by extension, the character’s response without viewing it in the context of the overall scene? In this regard, it is better to cut the whole scene first before supplying any overlaps. This is particularly easy on a NLE system where subsequent trimming is easier than ever. Because of this, cutting dialogue is the easiest and hardest procedure an editor can perform. It is easy because all one really needs to do in a two-person dialogue scene is checkerboard the cuts onto the nonlinear system’s timeline—in other words, cut straight across without thought of overlapping back and forth from Character A to Character B and so on. Cut just before the character speaks, without extra air, and then cut out as soon as the character stops speaking. Proceed in this manner, back and forth, for each character until reaching the scene’s conclusion. If using only one take per character, you can even expedite the process by cutting all of Character A’s lines and assembling them like railroad cars. Then go back and cut all of Character B’s lines, interspersing them with those of A. Using this approach it is possible to cut an entire dialogue scene before lunch.

Of course, it is not quite as simple as it sounds. Each selection requires an evaluation of performance—a decision as to which take best serves the purpose of the scene. In this regard, many editors comb the variety of takes in search of the best reading for each line in the scene. Others, touting the integrity of performance, will settle on one take for each character and use that throughout the scene. The problem with this approach is that if the actor’s intensity is particularly strong in one take, there may be moments, particularly at the beginning or end of the scene, where you need to ramp up or ramp down that intensity. If the actor has started at full throttle, in comparison to the scene’s overall energy, the performance may seem one-note after a while. Regardless of the approach to selection, most of the editing time should be devoted to determining which performances best fulfill the needs of the story. Once this is clear, the pieces fall into place.

After the checkerboard is complete, it is time to sit back and watch the scene. Let the scene speak to you. This is where the next phase of cutting occurs. As you watch, you’ll be struck by impulses that tell you this is where to cut, this is where we need to overlap. Watch carefully and trust your feelings. Maybe the actor says, “What do you think? Should I take the offer?” and your inclination is that you’d like to see the “you” he’s referring to and see what he’s thinking before he responds. So after the word “think,” you cut away to the listener while the rest of the dialogue, “Should I take the offer?” plays over the listener’s expression. As the overlapped dialogue finishes, the actor responds, “Sure. Go for it.” But by the time he says this, the audience will already have a sense of his true feelings and the motivations for his response. This is the magic of overlapping.

The editor also encounters other issues of clarity in reviewing the scene. For instance, what if the actor in close-up says, “Someone sent me this photo of you. Can you explain who you’re with?” It’s an opportunity to cut wide to see the hand reaching into the purse and taking out the photo. Or maybe you don’t need to establish where the photo resided and an even tighter shot, an insert of the hand holding up the photo, will work best. Or perhaps cutting the dialogue too tight misses a needed beat that would allow the viewer to question the speaker’s integrity or sympathize with his plight. By laying out the basic foundation for the scene, it becomes much easier to evaluate its needs and prescribe the proper audio and visual cues.

With the advent of digital nonlinear editing this highly linear approach to cutting dialogue has been infused with new life—and almost by default. Because extending or shortening a clip in the timeline requires a separate procedure—an activation of the trim mode functions, whether in Avid, Premiere Pro, or others—the editor’s tendency now is to proceed in the checkerboard fashion just described. Because the access to and selection of the material happen so swiftly, it is possible to throw together an initial cut in a matter of minutes. Often, without appreciating the underlying wisdom of this approach, the editor is compelled by the nonlinear system to cut the first pass in this manner, using straight cuts without overlaps. But generally it helps to understand why one is doing something, in the editing room as well as in life.

By discovering the underlying premise, one can carry the procedure through all times and places. In this way the insightful editor maintains the ability to address a variety of situations on various films rather than finding himself foiled by dependence on the elusive and unreliable pattern of habit. Yet, as in previous years, many novice editors still feel compelled by the urgency to overlap dialogue and begin the procedure before laying down all the cuts. Again, this leads to confusion and, potentially, disaster. In viewing scenes constructed in this premature manner, an audience senses that something isn’t quite right, even if they can’t articulate it. In many cases, because they have the advantage of viewing a completed scene with no preconceptions, the audience unconsciously anticipates certain moves that, when they aren’t forthcoming, frustrate their involvement in the movie.

Granted, at times it is advantageous to outwit the audience’s expectations, as David Lean did with the clever cut in Ryan’s Daughter (1970) where the audience fully expects the character of the British officer-turned-illicit-lover, Randolph Doryan (Christopher Jones), to commit suicide by igniting the dynamite and blowing himself up. But the scene ends before the anticipated explosion. Only in the next scene, from a distance, do we hear from across the landscape the powerful reverberation of a detonation. At other times it is best to jump ahead of the audience’s expectations and, in a sense, beat them to the punch or surprise them. But in emotional scenes, where the realm of human emotion makes everyone an expert, generally it is best not to mislead the audience.

Exposition Infection

To the other extreme from good subtextual dialogue lies excessive exposition. In dialogue, the film doctor must beware of this particularly noxious infection. Once it begins to spread, it can take over and destroy its host in a matter of scenes.

What do we mean by exposition or expository dialogue? Basically, it is dialogue that explains what has happened, is happening, or will happen. The old tenet in television of “tell them what they’re going to see, tell them what they’re seeing, and then tell them what they’ve seen” reflects this approach. Such a prescription posits that the audience is so stupid, presumably because only someone without much intelligence would buy the products advertised, that they had to have it all laid out for them like a TV dinner. In recent years, the proliferation of cable channels (including those that carry feature films), more film-savvy audiences, and greater competition have forced broadcasters to elevate their approach to TV writing. As the writing becomes more sophisticated, producers find that audiences gravitate toward shows like Mad Men (2007–2015) or House of Cards (2013–2017). On the other hand, purposely confounding an audience or keeping them guessing about essential bits of information can be deadly as well.

Show, Don’t Tell

The often pronounced prescription of “show, don’t tell” is particularly important with regard to a film story. What a viewer sees with her own eyes holds much more validity than what she’s told. Experiencing a boxing match is much more satisfying, informative, and memorable than having it described to you. Even spectacular, highly visual films like Avatar (2009) have to wrestle with excessive exposition. The concept of the avatar, a host body possessed by a person’s psyche, becomes clear when Jake Sully (Sam Worthington) enters the scientist’s pod and finds himself transported among other avatars and, ultimately, among the Na’vi as one of their own. Yet a lot of dialogue is expended explaining what the audience can easily glean through the stunning visuals.

If we look back at the earliest films, we are reminded that those filmmakers didn’t have the luxury of sound to tell their story. Generally, these silent movies were accompanied by a musical score played live before the audience. The essential information that had to be communicated in words was displayed on title cards, some intricately drawn, some simple white text on black background. Even films like Charlie Chaplin’s Modern Times, appearing in 1936, a decade after sound had entered the arena, relied more on visuals than on sync dialogue, even to the point of presenting title cards in place of some on-camera dialogue. Either way, these filmmakers knew that if they piled on too many cards, it would interrupt the story’s flow and disrupt the audience’s involvement. The early Chaplins, Pickfords, Keatons, and others used these sparingly. The images on the screen had to tell the story.

The arrival of the talkies offered a great privilege to filmmakers, and some used it wisely, often enlisting the talents of great playwrights or novelists to craft the dialogue. Others did not. Some silent film stars, who had grown accustomed to miming their way through a movie, had a hard time adjusting to sound. It revealed their unattractive voices and unconvincing speech. Silent film directors also struggled with the unforgiving demands of sound. Occasionally, it would go awry, as in the intriguing though overly long monologue at the end of The Great Dictator (1940) where Chaplin explains his philosophy to the audience. We must remember that as valuable as a message may be, it will mean more to an audience if they experience it for themselves. Filmmakers today would be well-served to consider the effectiveness of early silent films when deciding how much exposition they need to tell their story. (For more about expository dialogue, see Chapter 13).

► New Territory

Aside from bringing into common practice a process of dialogue cutting that had been the purview of a select group of editors, digital editing has also made way for a sophisticated new dialogue technique. For practical purposes, this method couldn’t have existed before now. It involves the selective use of multiple dialogue takes for multiple characters within a single shot. Where the option of multiple takes has always given a dialogue editor the chance to select and craft performance throughout a series of close-ups or over-the-shoulder shots, computerized editing allows the determined editor to shape performances within a stationary 2-shot as well. One of the early proponents of this approach is the comedy editor Bruce Green, who has spoken about it on a number of occasions to my students. He originally used this technique in the Mark Waters film Just Like Heaven (2005).

Generally, the shot consists of two characters engaged in conversation. Each actor depends on the other’s performance as well as her own to make the scene succeed. Until now, if one character’s performance is spot on but the other’s is not, the take is either rejected or used with its compromised performance. But today the offending performance can be removed and a better performance substituted for one character while not affecting the other character in the shot. This is accomplished by splitting the screen in half and employing a different take for Character B than was used for Character A, while retaining both within the same frame.

Obviously, this technique can only work if the camera is locked down so all the takes are identical in terms of subject matter, framing, and camera steadiness. With the use of motion control, where a computerized camera mount repeats the exact movement, such as a pan or tilt, for each frame, conceivably the editor could combine different takes with identical camera moves within the same scene.

As this technique makes clear, the computer’s ability to alter the frame through such effects as picture-in-picture, cropping, scaling, repositioning, warping, flipping, flopping, and so on, affords the editor seemingly limitless control over the once immutable frame. Where previously any effects had to be produced by a film laboratory according to the editor’s specifications using a lengthy photochemical process, now the editor can experiment freely, altering the frame at will. What he produces falls well beyond the means of earlier laboratories and postproduction facilities. In the coming years, with increased resolutions and decreased processing and rendering times, the picture editor will be able to output the final product directly from his editing machine, ready for final release. The main requirement, as in all film editing, still involves the aesthetic sense and skill to determine what constitutes a good performance.