Showing Versus Telling
Visual symbolism and relationship mapping
How much of the dramatic action should a movie try to show visually, and how much of the dramatic action should be portrayed through the use of words? The writers' imperative, "Show don't tell," is often a confusing dictum that drives people to do such things as come up with rules about how much visual action there should be in a movie, and how much dialogue.
Obviously, when a character describes the action, rather than acting it out, he has fallen headlong into the pit of telling. Or has he? Perhaps there is not such a distinct line as we suppose between dramatic action and other ways of presenting the story. And if this is true, what differentiates good story telling from bad in this respect?
A movie, like a stage play, is a very complicated set of communications channels. Even poetry, which doesn't necessarily even tell a story, is a complicated communications channel.
According to Sonesson, Gotthold Ephraim Lessing, "In the Laokoon, attempted to fundamentally define the separate functions of painting and of poetry. He pointed out that whereas painting is bound to observe spatial proximity--and must, therefore, select and render the seminal and most expressive moment in a chain of events--poetry has the task of depicting an event organically and in its temporal sequence. The essence of poetry thus lies not in description but in the representation of the transitory, of movement." *1
Movement, both in time and in space, is what happens in a story. A story is the unfolding of events that depict the dramatic action. That is, the character's conflicts are expressed, or acted out.
I won't maintain that poetry is best described by movement. Some poetry touches the heart by simply describing static conditions, such as other's emotional states. Nor will I maintain that a painting has only a fixed temporal quality to it. The moment in time that is captured in a painting is preceded by many moments in time that resulted in that point. Although the other frames are not in full view, we know that they are there from their result. If the right moment is captured, the symbols in the painting tell the entire story in a very brief form. This is what pictorial storyboards do. I guess I'm rarely content with absolute definitions that are so restrictive.
Semioticians have looked at how meaning is delivered through images in movies, and they are divided into many different schools of thought. According to Göran Sonesson, "In the case of pictorial semiotics, there has been, for some time now, three leading models of semiosis: those of the Greimas school, the Groupe µ, and the Quebec school; according the Fernande Saint-Martin (1994:2), however, one may also distinguish a fourth school, represented by the present author, which she calls the 'Swedish school'.
"...according to the Greimas model, all pictorial meaning derives from binary oppositions, the terms of which are distributed into two series which serve to separate at least two fields dividing the picture; the Quebéc school claims visual meaning is embodied in topological and Gestalt terms [Whole, not reducible into parts. Ed.]; the Groupe µ holds that pictorial meaning emerges from the ruptures suffered by the norms which are posited to hold for all pictures ('general norms'), or for the very pictures which they transgress ('local norms'). *1
My view is that most images have no dependable universal meanings, but can be dependably used to access experienced based meanings within each of us. The meanings that the images tap are different for each of us. I think this is more consistent with research into human intelligence and mental modeling. For an example of these differences in perception, see VISUAL SEMIOTICS AND THE PRODUCTION OF MEANING IN ADVERTISING at http://spot.colorado.edu/~moriarts/vissemiotics.html
What in the world would "mental modeling" have to do with anything in cinema? The concept of mental modeling can help us understand how the mind works, and therefore why certain things present the story well and certain things don't. For example, why is it that one group of people can look at a picture and see only a woman quietly smoking a cigarette, musing dispassionately about who knows what, while another group of people might see a woman confronting her existential awareness of the lack of meaning in her life, and the utter futility of all effort? It has very little to do with what is in the picture, and everything to do with culture indoctrination and what is inside the people's minds.
The mind is endlessly complex, but we do have a few clues to how it works. Mental modeling is one such inroad to understanding. An example, is the way in which we interact with a highway map. A highway map is a representation of actual routes, and uses symbols to indicate cities, services such as rest areas and camping, and highway types, such as two-lane, four-lanes, and Interstate. We don't continue to look at maps as we travel; we typically make in our head some representation (a mental model) of the area that we are interested in.
For someone interested in camping, the mental model derived from the map might simply be a spatial (dimensions in space) model with interconnections between camping sites. For someone interested only in major cities, the mental model derived would be a spatial model with interconnections between cities.
We can easily see how this would work, but what about for people who can't see, and other things related to dimensions in space? Studies of those who are blind, find that blind people make similar spatial maps. They "know" where the furniture and other things are through some representation in their mind. Mapping seems to be a basic function of the brain that is independent of vision.
Spatial mapping is definitely done by more than one sensory system. For example, the mind can recognize objects learned either by feel (hands) or by seeing, and can recognize objects by sight that have been learned by feel, and vice-versa. In the same way that the mind classifies objects perceived visually, the mind also classifies objects perceived haptically (by hand).2, 3
Another type of mapping phenomenon came to light with the advent of the computer program. A program is nothing but code. But the code is represented in usable form through the visual user interface. The user interface (UI), is the part of the program that we see and operate, and it enables interaction between the user and the program. *4
Human Computer Interaction (HCI) studies show that based on usage of the program, people create mental models of functions of the program. Most functions in today's programs are accessed through menus and dialog boxes. People know where the function is "located" based on this routing, or some other usage related schema known only to the individual. A more or less spatial representation of the entire program is derived from operating the program.
"Mental models are psychological representations of real, hypothetical, or imaginary situations. They were first postulated by the Scottish psychologist Kenneth Craik (1943), who wrote that the mind constructs '"small-scale models" of reality that it uses to anticipate events, to reason, and to underlie explanation. *5
The psychologist, George Kelly (1905 to 1967), is noted as one of the first to understand that we all have constructions in our minds of the reality we perceive, and use these constructions as templates to organize information so that we can comprehend it. This is called a constructivist view, to which I subscribe.*7
"Models can also represent abstract notions, such as negation and ownership, which are impossible to visualize," according to Johnson-Laird et al. This is a very important thing to realize. Not everything can be shown visually.
"...many authors think mental models are organised structures consisting of objects and their relationships. On the other hand, Redish (1994) suggests that they consist of propositions, images, rules of procedure, and statements as to when and how they are used." *4
Personally I favor the objects and relationships approach, as I think it corresponds to how people categorize information. Objects, as I perceive them, are mental representations of something, even if it is only a suspicion, regardless of whether the object is physical or abstract. However, this is just a construct for accessing mental modeling, and like all other theories, may not accurately represent how the mind actually works. I don't think that all mental objects are necessarily permanent, but many are constructed "on-demand" as purpose and context require that information within us be put in some order, and as things change, such as attitude, that influence relationships between information.
We have moved in this article beyond mental models of spatial relationships (things located in space that we can see), which would apply to two and three dimensional pictures and movement, and gotten into the world of situations and abstract things. This puts us in the land of different symbols, such as words.
Stevan Harnad, who does research on how the mind categorizes information, says that "What language allows us to do is to "steal" categories quickly and effortlessly through hearsay instead of having to earn them the hard way..." *6 In a sense, he is talking about an action like metaphor, which allows us to understand one thing through using a similar thing as a pattern. The mind is in a constant state of creativity, which is asked of it by new information. Just as the hand feels a pattern and maps it, and the mind can then recognize an object by sight, we can also recognize and understand something by the pattern of a similar idea.
For an idea of what a mental model is, do what is called a "mind dump." That is, take a blank piece of paper, place a topic (preferably a short one), or a question, problem, or idea in the middle of the paper, and then place all of the potentially related subjects and questions in relation to the central thought. If you are having trouble thinking of an idea, use this: "How I am going to describe the benefits of a new idea I have to my boss or other person." Indicate how these things are all related. When complete, you have a visual representation of your mental model of that central thought. You might find in representing it that your visual representation is structured in outline form, in list form, or centralized with the topic in the middle and related items around the central thought.
This structure may not represent your mental model so much as your ability to spatially represent things on paper. But it does represent the organization present in your mind by the relationships you indicate on paper. Some of these may be newly formed relationships that occur through the process of thinking the subject through to put it on paper.
The words that we use every day are understood by us partially through their interconnections with other words. These form concepts. The simplest form of cognitive modeling of concepts in words is Wordnet. Wordnet is a lexical database of English nouns, verbs, adjectives and adverbs that are organized into synonym sets, each representing one underlying lexical concept. I used Wordnet as a vocabulary resource for this site in the late 1990s. For more about Wordnet, visit http://www.cogsci.princeton.edu/~wn/.
The mind makes relationship models to make information understandable, by having information in reference to other things. We commonly call this, "Seeing things in perspective." We term these relationships, "ontologies," or "classifications," or "categories," because relationships are represented. These relationships might be linear or nonlinear.
One example of a hierarchical relationship could result from a mind dump of our thoughts about a woman of 34 years of age. Thinking about age and relationships, one relationship structure might extend upward to a mother and grandmother, with a line extending downward to a daughter. A horizontal line might extend to a husband. Other lateral lines, perhaps longer to indicate more distant relationships, might extend to cousins and peers. This is a linear model, arranged by relationship and age, and indicated by lines and placement.
A non-linear spatial image of family might be a circle, or an undefined area, that contains all of the same people in no particular order. I am calling it non-linear because there is no ordering, and no significance to placement and proximity.
The relationship structure that we form is an "on-demand" structuring of information, and is dependent on our purpose for structuring the information. For example, if our purpose was sexual desirability, we might have ordered the information by sexual desire, so a very different mapping would occur, which might include all of the other women that we know.
If our purpose was medical problems, we might have ordered by risk categories which might be assigned to the woman by probabilities. If our purpose was counting friends, we might have ordered the information by our friendships, and all of our friends, both male and female, young and aged, related and unrelated might be present. Similarly, bias, attitude, and a variety of other contexts may shape how we structure information by relationships.
This one woman could occur in many different mental models that our mind makes. Context and purpose determine how we order information into structures.
As an example of context in a visual setting, the image of a reverent man might suggest reverence by including religious symbols in the picture. The religious context tells us how to interpret the image. But the same image of the man, surrounded by children, or by nature (woods, stream), or by tall buildings, or by cell bars, would suggest very different interpretations of the image. Context is very important in determining how we interpret an image.
Similarly, the meaning of words is interpreted by the context of the sentence. If you say, "The woman was hot and dripping with sweat," the word hot means overheated. If you say, "She is one 'hot' woman," the word hot means "sexy." The other words in the sentence provide the context for interpreting the word, "hot." This is fully explored in What's In A Word.
Words are symbols and have the ability to create mental images, or to engage mental models. So, if you say, "Hot woman," you get some kind of mental image and related feelings. The relationship between the words and the mental image exists regardless of the source of the mental trigger, such as hearing or seeing. What is important is context, which stimulates a purposeful structuring information. Both words and images can access the same information.
Typically seeing the image of a "hot" person is more powerful than just hearing the words. But when words are used in novels to describe a sex scene, they may get more of a physical reaction than the same scene in a movie. Words that are less specific than images have the power to employ the imagination. For example, a particular man or action in a movie might not be sexually arousing to a particular woman. But a description in words lets the mind fill in the blanks with the reader's preferences. Showing too much, instead of telling, can get the opposite reaction.
As I have mentioned (numerous) times before, black box theater, in which the characters perform without props or a background (in limbo), is just as effective, if not more so, than if the stage was filled with props and backgrounds. The audience's mind is more thoroughly engaged by imagining the settings. Engaging the audience is essential to writing. The imagination is an extremely good tool for engaging the audience, and words are often more effective than images for engaging the imagination.
Both images and words tap into the experiences, attitudes, desires, fantasies, imagination, metaphors, knowledge, and categories (and stealing categories, as Harnad said) of the individual. Both images and words can order (restructure) information within the individual's mind on demand. What are the implications for writers? When is it better to show something very specifically, and when is it better to simply suggest something.
One great clue is the makeup that actors wear. We rarely see real skin with all of the pock marks, scars, wrinkles, age spots, and natural colors. We don't see reality at all. The appearance of the actors suggests a much tidier and pleasing world - we are not to be distracted by blemishes. And we are enchanted and seduced by enhancements made by props, costumes, makeup, and airbrush. Real details are simply not relevant - the entire production is simply a suggestion of what might be real.
Not everything can be shown visually. I remember a great radio promotion from several years ago. It featured something that couldn't be shown on television. It was the world's largest cup of hot chocolate. I can't remember the thing well now, but it seems like it had a battleship floating in it. An announcer gave a blow by blow account of the action as giant cranes dropped in whipped cream and a cherry. It stimulated the imagination to create a great mental image. Sometimes it is better to create a mental image like this simply by suggesting it, than to use all of the pictures in the world; and suggesting is a lot less expensive.
Many other things can't be conveniently shown visually. Complex relationships within the individual may take an entire movie to show, but can sometimes be expressed effectively in one or a few words. Similarly, external relationships to other people and objects may be just as difficult to show, while words get there much more quickly.
Similarly with the sex theme that I mentioned before, showing too much isn't helpful. In fact, as in porno films, some people may be turned on by an image, while many others are turned off by it. The power of suggestion, and the fantasy, have much more imaginative impact than a whole film full of pictures. (Watch for a new series on Human Sexuality, in January of 2003, in The Human Condition, in which the topic of fantasy will be central.)
It isn't nuts and bolts reality that is on the screen. Suggestions are made by images and words of what might be - a portrayal of action. Suggestions tap into each individual's mental reservoir to present a story.
Telling and showing, as reflected by words and images, can both have a lot of impact in emotional distancing. For example, tension can be created in a script in the following sequence:
In the preceding list, danger becomes more and more explicit by the growing proximity of two devices. Dangerous men get closer and closer. The threat of the gun becomes more and more personal. At first only words are used, and then both words and images are used to make the threat more real and immediate. But note that the command (in words) is nearly as threatening and immediate as the gun pointed at the person. (See the article on Emotional Distancing for more information.)
So, in dramatic action in a movie, which is typically a dramatization of character motivations in conflict, a combination of words, images, and physical action (movement) are necessary and appropriate for portraying the story.
What is necessary, when using either words or images, is that character motivation and intention are communicated in a way that engages the audience. Reciting events dispassionately just to get information out is not engaging. Engaging the audience is done through any communications channel (visual, auditory) by engaging the experiences, attitudes, desires, fantasies, imagination, metaphors, knowledge, and categories, of the individual. Sometimes this can be done better with images and actions, and sometimes with words. (See, Engaging The Audience).
A good story teller, whether telling the story aurally (think Homer and the Odessey), or visually (movie), or in print (novel), can conjure up images, situations, and ideas that captivate an audience.
Compare these sentences:
John enters quickly through the door, obviously distraught. "There is a monster out there. Marsha is running through the woods. Get your gun and come with me."
John smashes through the door like a human canonball, braking just in time to avoid running over Jim. "God almighty, it's huge! It has teeth like railroad spikes and claws like nails. It's on a rampage, chasing Marsha through the woods on two legs. Bring the biggest gun you have, and make your peace with God - we're all going to die. He storms back out."
Either scene might be lines in a book, or a description expressed by a character in a movie. The second scene above is visual, the first is not, but the visual one uses only words to conjure up images and situations, summoning existing experiences and feelings within the person.
Compare these images:
Against the deep crimson sunset, we see the profile of Marsha as she stops, screams, and then cringes in terror, and then we see the profile of the monster as it rises five feet above her, momentarily poised for the kill. John and Jim rush past, obscuring our view. John and Jim bow their heads, and then all is silent.
In the darkness of the woods, we see fleeting images of brown fur, a mouth full of large teeth, claws tearing at the ground, and small trees and limbs crashing and whipping all around. Marsha rises from where she has fallen. The monster is on her and Marsha is lost in a flurry of legs as large as trees and a gaping mouth full of huge teeth.
John and Jim follow the sound of the monster through the woods. We see them arrive too late to help Marsha. The monster attacks her, killing her.
The first two images are effective, but in different ways. In the first, everything is stripped away from the setting except Marsha and the monster. That's all we need to see. We don't even need to see the actual kill for the scene to be effective in chilling us. John and Jim step in front of the camera so we don't see the gruesome details.
In the second scene, the focus is on the details. We see just enough of the monster's terrible qualities to instill chills in us. And then we see more horrible attributes of the monster as it attacks Marsha. Still no blood and guts - just plenty of monster.
Scenes one and two summon experiences and feelings within the individual. The third scene, although it presents images, does not tell the story visually. It has no visual impact on us. The reason is because there is nothing in the scene to touch our feelings and experiences, and we see no character reactions to which we can relate.
2. Fiona N. Newell, Marc O. Ernst, Bosco S. Tjan & Heinrich H. Bülthoff (March 2000). Viewpoint dependence in visual and haptic object Recognition. http://www.kyb.tuebingen.mpg.de/bu/techr/list6.html Report 80.
3. Fiona N. Newell & Heinrich H. Bülthoff (March 2000). Categorical Perception of Familiar Objects. http://www.kyb.tuebingen.mpg.de/bu/techr/list6.html Report 79.
4. Philip Barker, Paul van Schaik, Spencer Hudson and Check Meng Tan. (~1998). Mental Models and Their Role in The Teaching and Learning of Human-Computer Interaction. http://wheelie.tees.ac.uk/groups/isrg/papers/sedawc97/
5. P.N. Johnson-Laird, Vittorio Girotto, and Paolo Legrenzi (1998). Mental models: a gentle guide for outsiders. http://www.si.umich.edu/ICOS/gentleintro.html.
6. Harnad, Stevan (2002) Symbol grounding and the origin of language http://eprints.ecs.soton.ac.uk/archive/00006471/01/harnad02.symlang.html
Various perspectives on mental models: A House Of Horizons And Perspectives. http://www.ceptualinstitute.com/uiu_plus/isss98/house-of-eyes.htm
Visual Semiotics And The Production Of Meaning In Advertising. http://spot.colorado.edu/~moriarts/vissemiotics.html Visual Communication Division of AEJMC, Washington, DC, August 1995.