APPLIED THEORY SUMMARY

䉬 Presents a framework for analyzing imagery in

multi-modal print documents and Web sites
䉬 Demonstrates how images and text work
together to make meaning for readers/users
䉬 Provides analytical tools and tips to help choose
still images to enhance textual messages

Visual Social Semiotics:

Understanding How Still
Images Make Meaning
CLAIRE HARRISON

P
rofessional communicators are increasingly in- help, promotional materials—that are designed to be rhe-
volved in developing documents, both print and torical in the classical Aristotelian sense, that is “the faculty
online, that include still images such as photo- of observing in any given case the available means of
graphs, illustrations, and diagrams. During my persuasion” (1954/2001). In other words, we manipulate
20-year career as a writing consultant to government, I words (our available means) to persuade readers/users that
have seen the ratio of visuals to text per page in public our particular message is credible, meaningful, and useful
documents such as brochures, pamphlets, and reports in- in a world flooded by a torrent of messages. Many of us,
crease steadily. In the case of online documents, rapid however, are not trained and practised in the use of images
advances in browsers and other technologies have trans- for rhetorical purposes.
formed the Internet, formerly a text-only medium, into the Do we understand sufficiently how visuals persuade read-
World Wide Web, a highly visual publishing environment ers/users about messages? Unless we are also trained as artists or
in which Tyner finds an historic echo: “The way that pic- art historians, how can we acquire this knowledge? The world of
tures and texts work together in multimedia interfaces is still images and their analysis is vast. As O’Toole notes,
reminiscent of the visually stunning illustrations of Biblical
texts seen in the illuminated manuscripts of medieval Books are written about individual works, groups of
times” (1998 p. 40). works, the artist’s whole oeuvre, schools, movements,
It is well beyond the scope of this article to explore the centuries of art. A great number of journals are devoted
historical, social, political, and technological reasons be- to nothing else. . . . Television films, lectures and tape-
hind the re-emergence of the visual as an important mode slide shows attempt to enrich our knowledge and per-
of communication within written documents. The impor- ceptions. Art is taught in school, in art college, in uni-
tant fact for professional communicators is that readers/ versity. Art is discussed in committees and boardrooms.
users no longer rely solely on written text for comprehen- All these involve verbal discourses about art and about
sion; they absorb and process all that they see within a individual works, so people are finding words to talk
document to create meaning for themselves. Horn calls this about art. However, these are diverse and competing
multi-modal mix visual language: discourses, with their own historical, biographical, eco-
nomic or technical preoccupations and they don’t, for
. . . the tight coupling of words, images, and shapes into the most part, help us as we stand in front of the art
a unified communication unit. “Tight coupling” means work, lost for words. (1994, p. 4)
that you cannot remove the words or the images or the
shapes from a piece of visual language without destroy- Yet, professional communicators cannot afford to be “lost
ing or radically diminishing the meaning a reader can for words,” because many of us work on production teams
obtain from it. (1999, p. 27)

However, as writers, we are trained and practised in Manuscript received 22 July 2002; revised 14 September 2002;
the use of words. We create documents—reports, online accepted 26 September 2002.

46 TechnicalCOMMUNICATION • Volume 50, Number 1, February 2003

 APPLIED THEORY
Harrison How Still Images Make Meaning

along with managers, designers, programmers, illustrators,

photographers, and so on.
I have, for example, often sat in meetings (as the
writer) where the designer passed around stock photos that
could potentially be used on a Web site. As a team, we had
to decide which image(s) would best represent the intent
of the Web site and contribute rhetorically to the text. Like
my colleagues, I generally relied on my “gut” feeling about
an image to make a judgment and struggled, as they did, to
explain my likes and dislikes to help the team reach a
consensus.
What I needed was an informed vocabulary that would
enable me to articulate my reactions when visuals were
being selected, tested, and evaluated during print produc-
tion phases such as design and layout, and Web production
phases such as development of the graphical user interface Figure 1. René Magritte’s painting makes a striking and
(GUI), asset creation, and prototype/beta testing and eval- explicit statement about the relationship between content
uation. (the signified) and its visual representation (the sign).
It became clear to me that those of us who communi-
cate primarily as writers face three significant challenges in WHAT IS VISUAL SOCIAL SEMIOTICS?
this new multi-modal communication environment. To en- Semiotics
sure that our documents are most effective for readers/ Semiotics is generally described as the “study of signs.” For
users, we must a sign to exist, there must be meaning or content (the
1. Understand how text and still images work to- signified) manifested through some form of expression or
gether to make meaning together for readers/users. representation (the sign). Figure 1 is a well-known painting
2. Know when still images enhance or detract from by René Magritte that demonstrates this relationship in a
text, and vice versa. striking and explicit manner. By putting the sentence This
3. Be able to effectively discuss the issues of multi- is not a pipe below a highly realistic representation of a
modal communications with other members of the docu- pipe, Magritte reminds viewers that the image is not reality
ment’s production team. but artifice—in other words, a representation or sign.
The purpose of this article, then, is to present a frame- A thoughtful viewer might note that the word “pipe”
work—visual social semiotics—that can help professional itself is an arbitrary combination of four letters that conveys
communicators who need practical tools for image analysis the concept of “pipe” through the form of written expres-
and who may not have the time or inclination to immerse sion— once again, a sign.
themselves in a new field of study. Although visual social Signs exist within semiotic systems. For example, the green
semiotics is not the only theoretical framework for exam- light in a traffic signal is a sign meaning “go” within the semiotic
ining how images convey meaning (others include Gestalt system of traffic control; words are signs in the semiotic system of
theory, art history, psychoanalytical image analysis, and language; gestures are signs within the semiotic system of non-
iconography, to name a few), it is unique in stressing that verbal communication; and so on. Because semiotic systems
an image is not the result of a singular, isolated, creative encompass the entire range of human practices,
activity, but is itself a social process. As such, its meaning is
a negotiation between the producer and the viewer, reflect- Semiotics provides us with a potentially unifying con-
ing their individual social/cultural/political beliefs, values, ceptual framework and a set of methods and terms for
and attitudes. use across the full range of signifying practices, which
Because professional communicators must persuade include gesture, posture, dress, writing, speech, photog-
readers/users of one particular viewpoint among many raphy, film, television, and radio. . . . As David Sless
competing and conflicting perspectives, I believe they will notes, “we consult linguists to find out about language,
find that visual social semiotics can help them better un- art historians or critics to find out about paintings, and
derstand the rhetorical, meaning-making potential of still anthropologists to find out how people in different soci-
images in relationship to text, provide them with tech- eties signal to each other through gesture, dress or dec-
niques to analyze such images, and contribute to their oration. But if we want to know what all these different
ability to effectively discuss imagery within a team setting. things have in common then we need to find someone

Volume 50, Number 1, February 2003 • TechnicalCOMMUNICATION 47

 APPLIED THEORY
How Still Images Make Meaning Harrison

with a semiotic point of view, a vantage point from 2. The meaning of signs is created by people
which to survey our world.” (Chandler 2001) and does not exist separately from them and the
life of their social/cultural community. Therefore,
It is this cross-cutting vantage point that allows professional signs have different meanings in different social and cul-
communicators to compare and contrast objects from two tural contexts—meanings that can range from very differ-
different semiotic systems—language and imagery—and ent (for example, different languages) to subtle and
make a valid, useful analysis. nuanced (for example, spoken English in U.S. versus
spoken English in India). This principle has profound
Social semiotics implications for professional communicators who must
Social semiotics is a branch of the field of semiotics. Lemke write for international audiences. The growing number
notes that of books and articles on this subject attests to the diffi-
culties writers face when trying to create messages for
social semiotics is a synthesis of several modern ap- people whose semiotic systems are different from
proaches to the study of social meaning and social theirs.
action. One of them, obviously is semiotics itself: the 3. Semiotic systems provide people with a vari-
study of our social resources for communicating mean- ety of resources for making meaning. Therefore, when
ings. . . . Formal semiotics is mainly interested in the they make a choice to use one sign, they are not using
systematic study of the systems of signs themselves. Social another. As Lemke adds,
semiotics includes formal semiotics and goes on to ask
how people use signs to construct the life of a commu- These are the contexts of “what might have
nity. (1990, p. 183) been”. . . . In the same sentence, what other words
could have been used? At the same point in the
Because every community is different, the signs used game, what other plays might have been made? For
by one community may be different from those used by the same detail in the painting, what other colors
another, For example, the color red indicates mourning for could have been used? (1990, p. 188)
people in Ivory Coast, whereas, in contrast, it represents
procreation and life for people in India. The ability to choose gives communicators a certain
Social semioticians apply three important principles amount of power to use signs in unconventional ways and,
when analyzing a semiotic system such as language or therefore, affect and even alter meanings.
imagery—principles that have significance for professional
communicators. Visual social semiotics
1. Semioticians believe all people see the world Visual social semiotics is a new field of study (originating in
through signs. As Chandler explains, the 1990s) and has been defined by Jewitt and Oyama as
involving “the description of semiotic resources, what can
Although things may exist independently of signs we be said and done with images (and other visual means of
know them only through the mediation of signs. We see communication) and how the things people say and do
only what our sign systems allow us to see. . . . Semioti- with images can be interpreted” (2001, p. 136).
cians argue that signs are related to the signifieds by Here is an example of how visual social semiotics can
social conventions which we learn. We become so used to be used as a tool in analyzing an image on a Web site to see
such conventions in our use of various media that they if it enhances, or detracts from, the text. Figure 2 is a
seem “natural,” and it can be difficult for us to realize the photograph from the Web site home page for the Supreme
conventional nature of such relationships. (2001) Court of the U.S.
The visual social semiotician would note a significant
Schriver suggests that successful professional commu- aspect of this photograph—its point of view. The photo-
nicators use intuition to “imagine the audience and draw graph is taken from an ant’s-eye perspective, placing the
on their internal representation of the audience as a guide Court building at a high vertical angle from the viewer. This
to writing . . . ” (1997, p. 156). I would add that this intuitive angle allows the photographer to glorify the Court by
“internal representation” includes a highly sensitized un- emphasizing the grandeur of its architecture and its classi-
derstanding of the sign conventions in a communicator’s cal elegance. The perspective elongates the columns and
particular language semiotic system. This sensitivity con- makes the portico more imposing. Moreover, the high
tributes to the skills that enable writers to replicate their vertical angle compels the viewer to look up at the build-
communities’ discourse in ways that attract interest or ing—a statement about the pre-eminent power of the
please readers. Court.

48 TechnicalCOMMUNICATION • Volume 50, Number 1, February 2003

 APPLIED THEORY
Harrison How Still Images Make Meaning

䉬 Deletes the people who do things and the people to

whom things are done
䉬 Eliminates tense (that is, past, present, and future)
䉬 Omits modal verbs (for instance, can, might, should)
When these elements are cut out from a text, the
process (for example, to judge) becomes depersonalized
and is rendered as an object (for example, a judgment).
Such textual entities are not only impersonal, they also take
on the quality of being timeless and fixed, even though
arguments can change and judgments have been over-
turned. On this Web page, the nominalizations contribute
substantial weight to the importance and influence of what
nine people do for a living.
Let us imagine the team putting this Web page to-
gether. They have already created the text for the title and
links and are now choosing a photograph. Say they have a
choice of three photographs: the ant’s-eye view, a bird’s-
eye view, and a “level” view. Which would provide them
with the best “tight coupling” to make the page’s visual
language most effective? For example, how “supreme”
would the Court appear if the bird’s-eye view photograph
was used? Or how timeless and fixed would the Argu-
ments, Guides, Rules, and Opinions seem if the perspective
Figure 2. Glorifying the Supreme Court through a high of equality were used on the page? Clearly, the team would
vertical angle (http://www.supremecourtus.gov). choose the ant’s-eye perspective to enhance the impor-
tance of the highest court of the U.S. judicial system, as
already implied by the text.
Note that I have not discussed the size or placement of
It’s important to remember that the producer of this the photograph on the Web page. Although such features—as
Web page had other perspective choices that could have well as the relationship of the image to other elements of the
altered this “power statement” about the Court. For exam- page, both printed and online— can have a significant impact
ple, we could have seen the building from a bird’s-eye on visual language, they are part of page design, an aspect of
viewpoint, making it smaller, putting it within the context visual communication not covered in this article. To help
of its surroundings, and reminding us that the U.S. govern- readers focus on the content of images analyzed here, most of
ment is “of the people, by the people, and for the people.” the images I use in this article are excerpted from Web pages.
Or we could have had a photograph of the building taken Professional communicators should be aware that vi-
on its steps so that we would be looking horizontally sual social semiotics may not be able to answer all the
toward the interior—a perspective that would imply equal- issues that an image may raise. For example, 10 different
ity between citizens and the judicial system. creators could be given the same image to produce and
Linguistic social semioticians would look at the text that, would do so in 10 different ways, creating different effects
in addition to the title “Supreme Court of the U.S.,” includes a through aesthetics and style— effects that are not ad-
list of links such as Oral Arguments, Case Handling Guides, dressed through this particular framework. Nor can visual
Court Rules, and Opinions. They would note that the docu- social semiotics help in understanding the rhetorical role
ment really contains ellipsis—a form of cohesive textual tie that text itself plays as a graphical image through color,
that acts by assuming that the reader is able to presuppose typography, and placement. (See Schriver 1997 for a discus-
meaning despite omitted words—in this case, verbs and ad- sion of how people “see” text.) However, as I hope the
jectives. The full thought behind the title and links is: “The discussion of the photograph of the U.S. Supreme Court
Supreme Court of the U.S. hears legal Oral Arguments, sets demonstrates, visual social semiotics can be an extremely
court Case Handling Guides, establishes Court Rules, and useful tool for analyzing images and their relationship to text.
gives legal Opinions.” The linguistic social semiotician would
note that this full thought makes heavy use of nominaliza- THE THREE CATAGORIES OF IMAGES
tions—arguments, guides, rules, and opinions. Nominaliza- The first step in understanding visual social semiotics is
tion, the process of creating a noun from a verb, knowledge of the three categories of images: the icon,

Volume 50, Number 1, February 2003 • TechnicalCOMMUNICATION 49

 APPLIED THEORY
How Still Images Make Meaning Harrison

Figure 3. When a house represents the home page, it is an Figure 4. When the upward-point arrow represents the top
iconic image. of the Web page, it is an indexical image.

index, and symbol. (The following explanation is adapted

from Hammerich and Harrison 2002, pp. 140 –142.)
䉬 Icon An image is iconic if it bears a similarity or re-
Figure 5. When a line appears beneath a word or phrase
semblance to what we already know or conceive
on the Web, it is a symbolic image.
about an object or person. Icons include paintings,
maps, and photographs and can range from very
realistic to very simplistic. On the Web, a common and to analyze how they are used to produce meaning by
icon is the house used to represent the home page, contemporary image-makers” (p. 1). Kress and van Leeu-
as shown in Figure 3. wen’s book was a first step in establishing visual social
䉬 Index An image is an index if it is recognizable, not semiotics for imagery in Western culture, and analysts in
because of any similarity to an object or person, but this field are now adding to the framework through applied
because we understand the relationship between the research and additional theory. (See Kress and van Leeu-
image and the concept that it stands for. A weather- wen 2001 for the development of an overarching social
vane, for example, has no resemblance to any as- semiotic theory for all types of multi-modal communica-
pect of weather, yet it stands for the concept of tions.)
wind. A common Web index is the upward-pointing Essentially, the Kress and van Leeuwen framework
arrow to indicate the top of a Web page as shown in recognizes that an image performs, simultaneously, three
Figure 4. Indexical signs can be confusing and, kinds of meta-semiotic tasks to create meaning. These tasks
therefore, often require text to accompany them. are called the representational metafunction, interpersonal
This is why the upward-pointing arrow on a scrol- metafunction, and compositional metafunction.
lable page is often also labeled, “Top of Page.” This section of the article addresses the major elements
䉬 Symbol An image is a symbol when it has no visual of each metafunction, briefly analyzes different Web site
or conceptual connection to an object or person. We images according to each metafunction, and provides a list
know the meaning of the image only because of of useful questions for basic analysis of the metafunction.
convention; that is, it’s something we’ve learned. A At the end of the article, one of the images is used to
word, for example, is a symbol because it does not demonstrate how the three metafunctions work together to
resemble what it stands for, nor does it have any create visual meaning for viewers.
indexical relationship to what it signifies. Take the I have included a variety of types and styles of images
word rose. It doesn’t look like a rose or bear any to help professional communicators working in different
relationship to the concept of a rose. A common fields who wish to undertake visual analyses. It is important
Web symbol is the line beneath a word or phrase to note that these analyses apply visual social semiotic
used to indicate a link, as shown in Figure 5. theory from the perspective of Western culture in general
Although indexical and symbolic images play impor- and North American culture in particular. As such, it may
tant roles in a Web site’s visual language, I will focus not be relevant to other traditions that have developed
primarily on icons in this paper because professional com- different conventions of imagery and reading.
municators are likely to find that these images are the most
contentious during team discussions. The representational metafunction
The representational metafunction is about the people,
ANALYZING ICONIC IMAGES places, and objects within an image—the represented par-
In their seminal work, Reading images: The grammar of ticipants (RPs)—and answers the question “What is the
visual design, Kress and van Leeuwen (1996) set out the picture about?” Table 1 outlines basic structures and pro-
first social semiotic framework for analyzing images, noting cesses of this metafunction.
that “We intend to provide inventories of the major com-
positional structures which have become established as An action image According to Kress and van Leeuwen,
conventions in the course of the history of visual semiotics, “When participants are connected by a vector, they are

50 TechnicalCOMMUNICATION • Volume 50, Number 1, February 2003

 APPLIED THEORY
Harrison How Still Images Make Meaning

TABLE 1: BASIC STRUCTURES AND PROCESSES OF THE REPRESENTATIONAL

METAFUNCTION

Structures Processes

Narrative: Narrative images allow viewers to 䉬 Action: The narrative is created by vectors that can be bodies,
create a story about the RPs because the limbs, tools, weapons, roads, and so forth.
images include vectors of motion. 䉬 Reactional: The narrative is created by eyelines (acting as
vectors) between RPs.

Conceptual: Conceptual images do not 䉬 Classificatory: RPs as “kind of” something or some group (that
include vectors. Rather, RPs tend to be is, they are members of the same class). Advertisements for
grouped together to present viewers with beauty products often have classificatory images such as a group
the “concept” of who or what they of models (for instance, Revlon models).
represent. 䉬 Analytical: RPs are displayed in terms of a “part-whole”
structure. The “whole” is a Carrier who possesses “parts” called
Attributes. The Supreme Court building in Figure 2 is a Carrier,
and its architectural components are its Attributes. A pie chart is
an analytical image in which the chart is the Carrier and its
segments are Attributes. Diagrams are also analytical processes.
䉬 Symbolic: RPs are important for what they “mean.” A motorbike
in an advertisement can, for example, be analytical (that is,
asking the viewer to check out its attributes), but it is also
symbolic of virility. Abstract shapes such as triangles, squares,
and circles also fall in this category.

represented as doing something to or for each other. . . .

These vectors are formed by depicted elements that form
an oblique line, often a quite strong diagonal line . . . ”
(1996, pp. 56 –57).
The photograph in Figure 6 demonstrates the effective
use of vectors to create action and, therefore, narration. It
is the main image on the home page for the American
Podiatric Medical Association, and the text following is the
first paragraph appearing beneath it.

Paying Careful Attention to Details

The foot is a complex structure made up of 26 bones,

thirty-three joints, 107 ligaments, and 19 muscles and Figure 6. Strong vectors create a meaningful narrative
tendons. They deserve the focused attention and care (http://www.apma.org/).
that the members of the American Podiatric Medical
Association (APMA) can provide. For feet of all ages, newborn’s leg. Both vectors begin at a corner of the pho-
from the young foot to older feet the members of the tograph and meet approximately in the middle, creating a
APMA provide high quality healthcare. Details are care- powerful interaction between the major RPs—the foot and
fully attended to in meeting the requirements of the hand. The smaller vectors created by the toes and fingers
healthcare needs of your feet. Good foot health is vital to provide additional action elements so that the picture has a
your overall health. naturally dynamic rather than static feel. Given the text,
viewers have no problem understanding the narrative: that
Several vectors appear in this image. The two strongest the newborn is a patient and that the hand belongs to a
are the diagonal lines created by the adult’s hand and the caring podiatrist who pays careful attention to detail, no

Volume 50, Number 1, February 2003 • TechnicalCOMMUNICATION 51

 APPLIED THEORY
How Still Images Make Meaning Harrison

matter how tiny.

The representational metafunction can include embed-
ding (that is, containing more than one process) in the
same way that a sentence can have a main clause and
embedded relative clauses such as “The man who wore the
hat was walking with an umbrella.” In Figure 6, the out-
of-focus circle in the background—the hand entrance to an
incubator—is a conceptual structure, that is, a symbolic
process, embedded within the action process. An incubator
is found in hospitals and is, therefore, symbolic of medical
intervention to promote human health.
This embedded image supports the text statement
that “For feet of all ages, from the young foot to older
feet the members of the APMA provide high quality
healthcare.” Further, Kress and van Leeuwen point out
that “Circles and curved forms generally are elements we
associate with an organic and natural order. . . . The Figure 7. This “slice” of a faceless crowd encourages the
world of organic nature is not of our making, and will viewer to consider breast cancer in the abstract
always retain an element of mystery” (1996, p. 53). Note (http://www.thebreastcancersite.com/cgi-bin/WebObjects/
how the text demonstrates practitioners’ specialized CTDSites).
knowledge of the “organic and natural order” by unveil-
ing some of its mystery: “The foot is a complex structure
made up of 26 bones, thirty-three joints, 107 ligaments,
and 19 muscles and tendons.” trations in their projects would be well served by becoming
acquainted with the extensive literature on the subject.
A conceptual image. Conceptual images do not involve (Kostelnick 1998 provides a good introduction to the dif-
action or reaction on the part of RPs but represent “partic- ferent standards for data displays as well as identifying
ipants in terms of their more generalized and more/less leaders in the field of technical illustration.)
stable and timeless essence, in terms of class, or structure,
or meaning” (Kress and van Leeuwen 1996, p. 79). Useful questions for analysis of the
Figure 7 is the main illustration from the home page of representational metafunction
The Breast Cancer Site and is accompanied by the text: 1. Who are the represented participants (RPs) in the
“43,000 will die this year.” The use of a faceless crowd, a image? Include both human and non-human objects.
classificatory process, encourages the viewer to think of 2. Are there are any vectors in the image that indi-
breast cancer in the abstract, that is, about the many who cate action? If so, what kind of story does this action tell?
will die and the social, political, and economic conse- 3. Are the human RPs looking at each other, creat-
quences of the disease. The figures, which are only out- ing eyeline vectors? If so, what does this tell me about
lines, are similar to those seen in graphs, giving a technical/ the history of these people?
scientific resonance to the image. (It is also interesting that 4. If there are no vectors, what is the image trying
this facelessness of the figures allows the viewer to insert to tell me in terms of social/cultural concepts? What
his/her face, thereby adding a personal, and frightening, types of conventional thinking do different objects evoke
resonance.) The calendar in the right-hand corner is an in me?
indexical sign, serving as a type of “death-watch count- 5. Is the image a complex one with more than one
down.” The rectangular shapes of both the image and this process embedded within it? If so, how do these embed-
sign add to the scientific/technical quality of this image ded processes add to my overall understanding of the
since “In our society, squares and rectangles are the ele- image?
ments of the mechanical, technological order . . . ” (Kress 6. In terms of the overall document/Web site, does
and van Leeuwen 1996, p. 52). the choice of image RPs best enhance its intent and that
Although visual social semiotics assisted in the analysis of the text?
of Figure 7, the theoretical framework is not sufficient to
deal with the many issues involved in the design of appro- The interpersonal metafunction
priate and accurate charts, diagrams, and so on. Profes- This metafunction is about the actions among all the par-
sional communicators who must deal with technical illus- ticipants involved in the production and viewing of an

52 TechnicalCOMMUNICATION • Volume 50, Number 1, February 2003

 APPLIED THEORY
Harrison How Still Images Make Meaning

TABLE 2: BASIC FEATURES AND PROCESSES OF THE INTERPERSONAL METAFUNCTION

Features Feature Processes

Image Act and Gaze: The image act involves the 䉬 Demand: The RP is looking directly at the viewer. A
eyeline of the RP(s) in relation to the viewer. demand generally causes the viewer to feel a strong
engagement with the RP.
䉬 Offer: The RP is looking outside the picture or at
someone or something within the image. In this case, the
RP becomes an object of contemplation for the viewer,
creating less engagement than that of the demand.

Social Distance and Intimacy: Social distance is The viewer can see an RP in six different ways.
determined by how close RPs in an image appear 䉬 Intimate distance: The head and face only
to the viewer, thereby resulting in feelings of 䉬 Close personal distance: The head and shoulders
intimacy or distance. 䉬 Far personal distance: From the waist up
䉬 Close social distance: The whole figure
䉬 Far social distance: The whole figure with space around it
䉬 Public distance: Torsos of several people

Perspective—The Horizontal Angle and 䉬 The frontal angle: When an RP is presented frontally to
Involvement: This angle refers to the relationship the viewer. This angle creates stronger involvement on
between the position of the RP(s) and the viewer. the part of the viewer as it implies that the RP is “one of
us.”
䉬 The oblique angle: When an RP is presented obliquely to
the viewer. This angle creates greater detachment since it
implies that the RP is “one of them.”

Perspective—The Vertical Angle and Power: There 䉬 High angle: The RP “looking up” has less power.
are two possible vertical-angle relationships: 1) that 䉬 Medium angle: The RP “looking horizontally” has equal
of the RP(s) and the viewer, and 2) that between RPs power.
within an image. 䉬 Low angle: The RP “looking up” has less power.

image (that is, the creator, the RPs, and the viewer), and From early childhood education and the schools to af-
answers the question “How does the picture engage the ter-school programs and health services, we take on the
viewer?” Table 2 outlines basic features of this metafunc- issues affecting our kids.
tion.
The human face is one of the most powerful resources
Creating strong viewer involvement. Figure 8 is the in visual imagery because people are “hard-wired” from
only image on the home page of the Stand for Children infancy to study faces and their expressions. However,
Web site and appears below the following text: faces can be presented in many different ways. This pho-
tograph effectively combines four aspects of the interper-
Stand for Children is America’s only nationwide grass- sonal metafunction to engender strong viewer involvement
roots voice for children. Our members take action at the with the image: visual demand, intimate distance, frontal
national, state, community, and neighborhood levels. angle, and medium vertical angle.

Volume 50, Number 1, February 2003 • TechnicalCOMMUNICATION 53

 APPLIED THEORY
How Still Images Make Meaning Harrison

have chosen to bring the child as close to the viewer

as possible (even cutting off the top of her head) to
create a strong sense of viewer affinity.
䉬 Frontal and medium vertical angles Kress and
van Leeuwen note that “The horizontal angle en-
codes whether or not the image-producer (and
hence, willy-nilly, the viewer) is ‘involved’ with the
represented participants or not. The frontal angle
says, as it were: ‘what you see here is part of our
world, something we are involved with” (1996, p.
143). Simultaneously, the medium vertical angle be-
tween the viewer and child indicates equality. The
combination of the two angles promotes intense in-
volvement with the child and, by association, with
American children. The angles suggest not only that
children are equal to adults but that their issues and
problems are equally as important.
When analyzing any image, it is important to remem-
ber that the producers had other semiotic choices. For
example, the producers of this site could have chosen to
photograph the child in profile (at an oblique angle pre-
senting a visual offer), or with a group of children (public
distance), or placed at a low angle to the viewer (less
equality). It is questionable whether any of these choices
would have resulted in an image as rhetorically supportive
Figure 8. This head shot creates strong viewer engagement of the accompanying text.
(http://www.stand.org).
An interesting visual trend As discussed in Table 2, the
vertical angle is associated with power relationships. I have
noted an interesting trend in imagery when producers wish
to encourage readers/users to incorporate new technolo-
gies into everyday practices. The images representing the
䉬 Visual demand Kress and van Leeuwen say that a use of the technology are frequently at a low visual angle to
demand has two functions: “In the first place it cre- the viewer. In semiotic terms, this means that the viewer
ates a visual form of direct address. It acknowledges holds the power over the technology, not only in terms
the viewers explicitly, addressing them with a visual of choosing to use the technology, but also of being able
‘you’. . . . [Also,] the participant’s gaze (and the ges- to understand, control, and manage it. For example,
ture, if it is present) demands something from the Figure 9 from the U.S. Postal Service encourages users to
viewer, demands that the viewer enter into some go online to change their addresses, whereas Figure 10
kind of imaginary relation with him or her” (1996, p. tells us to “Eliminate check writing and pay anyone
122). The child’s direct look and lack of smile anytime online.”
strongly beseech the viewer to acknowledge the Figure 11, from a site promoting telemedicine, is inter-
needs of children. That the child is female and non- esting because it demonstrates two different vertical angles
Caucasian is interesting, suggesting that the produc- and power relationships. The first is the user-over-technol-
ers wished to add gender and racial meanings not ogy relationship because the perspective of the photo-
included in the text. graph ensures that viewers look down on the scene itself.
䉬 Intimate distance Spatial distances are related to The second is implicit in the vertical angle between the
emotions of intimacy and distance. Kress and van doctor (higher; more powerful) and the patient (lower; less
Leeuwen explain that “The relation between the powerful). This angle plus the embedded symbolic repre-
human participants represented in images and the sentation of a figure in a white coat and stethoscope is
viewer is once again an imaginary relation. People are designed to reinforce beliefs about the important knowl-
portrayed as though they are friends, or as though they edge possessed by healthcare providers and their resulting
are strangers” (1996, p. 133). In Figure 8, the producers high power status in our society.

54 TechnicalCOMMUNICATION • Volume 50, Number 1, February 2003

 APPLIED THEORY
Harrison How Still Images Make Meaning

Figure 9. The low visual angle is intended to encourage

users to go online (http://www.moversguide.com/).

Figure 10. The low angle suggests that online banking can
be easily learned and managed
(http://moneycentral.msn.com/banking/home.asp).

Figure 11. The low visual angle of the total image suggests
that patients should not be overawed or intimidated by new
Useful questions for analysis of medical technologies, while the vertical angle between
the interpersonal metafunction physician and patient reinforces the status of the medical
1. Does the image include human RPs? If so, what profession (http://www.mgh.org/telehealth/vision.html).
type of image act is taking place, a demand or an offer?
Theoretically, an object can create a demand—for exam-
ple, a car placed so that its headlights appear to be look-
ing at us. However, demands and offers seem most pow-
erful when they involve an actual human face. The compositional metafunction
2. If the image act is a demand, how does it affect The compositional metafunction answers the question
me? And is it accompanied by any gestures or expres- “How do the representational and interpersonal metafunc-
sions that make it more forcible? tions relate to each other and integrate into a meaningful
3. If the image act is an offer, why has the producer whole?” Composition in imagery is the equivalent of syntax
of the image chosen to make the RP an object of study? in language—a set of rules that enable the signs of lan-
4. How close do I feel to the RPs in the image? guage (that is, words) to be arranged grammatically so that
Does the closeness make me feel as if the RPs are friends they make sense to the reader.
or strangers? In either case, why has the producer of the In other words, the composition of an image or the
image chosen to evoke these feelings within me? lay-out of a print or Web page represents visual syntax. If
5. What do I notice about the perspective in the the signs are not put together in a rule-oriented way,
image? What horizontal and vertical angles have been viewers will see a hodge-podge of images rather than a
used? coherent whole. Table 3 outlines the basic systems and
6. How does the horizontal angle affect my sense of elements of this metafunction.
involvement with the RPs?
7. How does the vertical angle add to my knowl- Revisiting Figure 6
edge of power relations between myself and the RP and Let us revisit Figure 6 and its text to examine how the three
between the RPs themselves? metafunctions work together to make a visual message for
8. What other semiotic resources could the producer the viewer, and why the image and text work so effectively
have used to create a different impression? together as a rhetorical unit. On the Web page, the image

Volume 50, Number 1, February 2003 • TechnicalCOMMUNICATION 55

 APPLIED THEORY
How Still Images Make Meaning Harrison

䉬 Salience Size, focus, and foreground/background

are the processes of salience that contribute to the
rhetoric of the photograph. The hand is the largest
RP, emphasizing the importance of the doctor. The
vectors are in focus and in the foreground, whereas
the circle of the incubator and the rest of the back-
ground are out of focus. The viewer is, therefore,
compelled to pay attention to the action inherent in
central image of the clasp.
䉬 Framing Although the photograph has no actual
lines as a frame, the use of color (which of course
cannot be seen in the black-and-white version of the
photo reproduced here) sets it apart from the white-
ness of the Web page on which it is placed and
frames it as a separate item. Also, the colors
within the image fall within the white-cream-
brown spectrum of the palette, creating a continu-
ity that holds the pictorial elements together. The
text sits directly below the photograph with the
slogan “Paying Careful Attention to Details,” acting
as the photograph’s caption and the lead-in to the
Figure 12. The image plus text create a strong rhetorical unit.
paragraph itself.
䉬 Modality Photographs, by the nature of their tech-
nology, suggest a reality that is far stronger than that
of drawings, illustrations, and paintings. As Shapiro
appeared above the first paragraph in the text, as shown in notes, “Of all modes of representation, [photogra-
Figure 12. phy] is the one most easily assimilated into the dis-
I have already discussed the representational meta- courses of knowledge and truth for it is thought to
function of the photograph, and the way in which the RPs be an unmediated simulacrum, a copy of what we
and their vectors create a strong narrative structure (that is, considered the ‘real’” (1988, p. 124). In fact, we find
a podiatrist taking good care of a newborn). The interper- photographs to be so real that Shapiro believes they
sonal metafunction for this image is “realized” through the pacify us to the point of unquestioning acceptance
use of intimate distance, the horizontal frontal angle, and of societal norms and conventions. In Figure 12,
the medium vertical angle, which heighten viewers’ sense therefore, the producer of the image uses the photo-
of personal involvement and identification with the RPs. graph’s high modality so that the ideal will be de-
The compositional metafunction effectively integrates the picted, not as “what might be,” but as “what is” (that
first two metafunctions through a variety of elements. is, the truth).
䉬 Information value The photograph uses all three In sum, the systems of the compositional metafunction
types of information value systems to get the rhetori- in Figure 12 play a significant role in integrating the two
cal message across to viewers. Within the image, other metafunctions so that the rhetorical messages of the
there is a clear left/right system. The foot on the left image and text combine almost seamlessly and come
represents the given (that is, a patient requiring care) across “loud and clear” to the viewer.
while the hand on the right represents what the
Web site wishes to offer as a new idea to users— Useful questions for analysis of
that podiatry care is a significant aspect of personal the compositional metafunction
healthcare. The image also has a center/margin sys- 1. How have the RPs been placed to provide infor-
tem with the hand holding the foot as the nucleus of mation, and why has the producer of the image chosen
information. And, finally, the image combined with this placement?
the text creates an ideal/real system. The ideal of the 2. Which RPs are more salient than others, and how
positive, caring doctor-patient relationship is visual- does this salience affect the impact and meaning of the
ized in the photograph above the text while the re- image?
al—facts about feet and practitioners—is incorpo- 3. How are the RPs held together or separated
rated in the text below the image. within an image, and why?

56 TechnicalCOMMUNICATION • Volume 50, Number 1, February 2003

 APPLIED THEORY
Harrison How Still Images Make Meaning

TABLE 3: BASIC SYSTEMS AND ELEMENTS OF THE COMPOSITIONAL METAFUNCTION

System Elements

Information Value: The placement of RPs allows them to 䉬 Left/Right: RPs on the left side of an image have
take on different information roles. the value of being “given” knowledge while RPs
on the right are “new.”
䉬 Given ⫽ familiar, commonsense
䉬 New ⫽ an issue, a problem, a solution
(Note: This value is based on how we read in
Western cultures, that is, from left to right. This
does not necessarily apply to cultures in which
reading occurs from right to left or in columns.)
䉬 Top/Bottom: RPs at the top of an image have the
value of being “ideal” while RPs below represent
the “real.”
䉬 Ideal ⫽ emotive, imaginary, what might be,
often the pictoral elements of an image
䉬 Real ⫽ factual, informative, down to earth,
practical, often textual elements in an image
䉬 Center/Margin: RPs in the center provide the
nucleus of information to which surrounding
elements are subservient.

Salience: Salience refers to the ability of an RP to capture 䉬 Size: The larger the RP, the greater the salience.
the viewer’s attention. 䉬 Sharpness of focus: Out-of-focus RPs have less
salience.
䉬 Tonal contrast: Areas of high tonal contrast have
greater salience.
䉬 Color contrast: Strongly saturated colors have
greater salience than “soft” colors.
䉬 Foreground/Background: An RP in the foreground
has greater salience than an RP in the background.

(continued)

Volume 50, Number 1, February 2003 • TechnicalCOMMUNICATION 57

 APPLIED THEORY
How Still Images Make Meaning Harrison

TABLE 3: (Continued)

䉬 Framelines: The lines within the image that

divide RPs or hold them together.
䉬 Pictorial framing devices: The stronger the
Framing: How RPs are framed affects whether they are lines around the image, the greater the
seen as connected or separate. connection.

Modality:䉬 Modality refers to how we feel about the visual Modality markers: The visual cues that indicate
message’s validity and reliability. Images with higher “realness” generally run along a spectrum of
modality appear more real than those with a lesser possibilities.
modality. However, the “realness” of imagery can be 䉬 Color saturation, differentiation, and modulation:
problematic. For example, although Figure 6 has very low 䉬 Full color ⫽ high modality
modality according to the framework (that is, the figures
are not real, but abstracted), the message behind the 䉬 Black-and-white ⫽ low modality
image may hold great validity for viewers. 䉬 Contextualization:
䉬 Fully conceived background ⫽ high modality

System Elements

䉬 Background completely absent ⫽ low

modality
䉬 Depth:
䉬 Deep perspective ⫽ high modality
䉬 No perspective ⫽ low modality
䉬 Illumination:
䉬 Fullest representation of light and shade ⫽
high modality
䉬 Absence of light and shade ⫽ low modality

䉬 Although Kress and van Leeuwen (1996) do not include modality in any of the metafunctions, I have followed the lead of Jewitt and Oyama
(2001, pp. 151–153) by placing it in the compositional metafunction.

4. How does the use of color or lack of it affect the understanding many conventions found in Western imag-
rhetorical message of the image? ery that, despite people’s differences in age, ethnicity, gen-
5. How real does the image appear to the reader, der, and so on, evoke generally uniform reactions. How-
and does this sense of reality affect the validity of its ever, learning to use the framework en masse is a
message and that of the accompanying text? formidable, daunting process because it is complex and
6. Are there other ways this image could have been introduces a great deal of new terminology. I have found
organized that would strengthen its message and more that the best approach is not to attempt to use the whole
effectively enhance its accompanying text? framework immediately, but to focus on one or two aspects
of the framework when starting to analyze images. Here
FINAL WORDS ABOUT GETTING STARTED are my suggestions for getting started.
Kress and van Leeuwen state that 䉬 Begin with simple images rather than complex
ones—for example, the icons chosen for a GUI or
Social semiotics is an attempt to describe and under- head shots that involve only one person. Move to
stand how people produce and communicate meaning more complex images and then page designs as you
in specific social settings, be they “micro” settings such develop your skills.
as the family or settings in which sign-making is well 䉬 Use only one metafunction in early analyses. I sug-
institutionalized and hemmed in by habits, conventions gest the interpersonal metafunction because it is the
and rules. But social semiotics, sign-making in society, one whose effect you and your colleagues will feel
is so varied an activity that any attempt to capture it in most immediately on viewing an image.
a general theory must look crude by comparison with 䉬 Use only one or two elements of the chosen meta-
the richness of the actual semiotic world. (1996, p. 264) function in early analyses. Add elements as you gain
the confidence to undertake more in-depth study.
Given this caveat, I believe that professional commu- 䉬 When analyzing the representational metafunction,
nicators will find visual social semiotics an effective tool for dig deep. The rapidity of human visual perception is

58 TechnicalCOMMUNICATION • Volume 50, Number 1, February 2003

 APPLIED THEORY
Harrison How Still Images Make Meaning

such that you understand what an image is about REFERENCES

immediately on sight. Chances are you will not be
accustomed to taking an image apart to analyze its Aristotle. 1954/2001. “Aristotle’s rhetoric.” W. R. Roberts,
separate elements. trans. http://www.public.iastate.
䉬 Your visual virtuosity also means that you are likely edu/⬃honeyl/Rhetoric/index.html
to understand many of the elements of the composi-
tional metafunction at an intuitive level. Make this Chandler, Daniel. 2001. “Semiotics for beginners.”
knowledge explicit by explaining it to others. http://www.aber.ac. uk/media/Documents/
䉬 Remember that not every aspect of every metafunc- S4B/semiotic.html
tion is important in analyzing an image for its impact
on a reader/user. For example, the most notable fea- Dragga, Sam, and Dan Voss. 2001. “Cruel pies: The
ture of the photograph of the U.S. Supreme Court in inhumanity of technical illustrations.” Technical
Figure 2 is the interpersonal aspect of power “real- communication 48:265–274.
ized” through the vertical angle.
䉬 Continually ask yourself why a producer chose a par- Hammerich, Irene, and Claire Harrison. 2002. Developing
ticular semiotic resource. Just as you instinctively know online content: The principles of writing and editing for the
how to use words to evoke certain feelings in readers/ Web. New York, NY: John Wiley & Sons.
users, producers often rely on instinct in their image
choices and may not be aware at a conscious level of Horn, Robert. E. 1999. “Information design: Emergence of a
the sign conventions they have employed. new profession.” In Information design, ed. R. Jacobson,
䉬 Reflect on the development of images and design as Cambridge, MA: The MIT Press, pp. 15–33.
a project moves through its production phases, not-
ing how colleagues discuss images and how illustra- Kostelnick, Charles. 1998. “Conflicting standards for
tors and designers respond to their suggestions. And designing data displays: Following, flouting, and
if usability testing is undertaken, note how users re- reconciling them.” Technical communication
act to the rhetorical messages embedded in the com- 45:473– 482.
bination of text and imagery. The more you learn
about others’ social/cultural “lenses,” the more sensi- Kress, Gunther, and Theo van Leeuwen. 2001. Multimodal
tive you will become to the different potentials for discourse: The modes and media of contemporary
meaning-making with images. communication. London, UK: Arnold.
䉬 As you gain confidence and can handle a variety of
images and page designs, consider ways in which Kress, Gunther, and Theo van Leeuwen. 1996. Reading
you might, as Dragga and Voss suggest, propose images: The grammar of visual design. London, UK:
that a team humanize technical illustrations by add- Routledge.
ing imagery to pie charts, bar graphs, diagrams, and
so on, and “genuinely integrating words and pictures Jewitt, Carey, and Rumiko Oyama. 2001. “Visual meaning: A
instead of simply juxtaposing the two on the page or social semiotic approach.” In Handbook of visual analysis,
screen” (2001, p. 270). eds. T. van Leeuwen and C. Jewitt. London, UK: Sage
䉬 Practice visual social semiotics to “get the hang” of Publications, pp. 134 –156.
it. Magazine advertising is a fertile field for analysis
as marketers rely on either meeting our expectations Lemke, Jay. 1990. Talking science: Language, learning and
regarding sign conventions or breaking them to values. Norwood, NJ: Ablex Publishing Corporation.
attract our attention.
Throughout my schooling, I was told that “a little bit of O’Toole, Michael. 1994. The language of displayed art.
knowledge is a dangerous thing”—the implication being London, UK: Leicester University Press.
that people making judgments without in-depth knowl-
edge of a particular subject will make mistakes. My expe- Shapiro, Michael J. 1988. The politics of representation:
rience in visual social semiotics is the reverse. Even a small Practices in biography, photography, and policy analysis.
understanding has been extremely useful in my work. Madison, WI: University of Wisconsin Press.
Therefore, I believe that professional communicators will
find value in the framework whether they decide to focus Schriver, Karen. 1997. Dynamics in document design. New
on one aspect or delve deeply into theory. TC York: John Wiley & Sons.

Volume 50, Number 1, February 2003 • TechnicalCOMMUNICATION 59

 APPLIED THEORY
How Still Images Make Meaning Harrison

Tyner, Kathleen R. 1998. Literacy in a digital world: Teaching completed an MA with Distinction in applied language studies at
and learning in the age of information. Mahway, NJ: Carleton University, Ottawa, ON, specializing in writing, rhetoric,
Lawrence Erlbaum Associates, Publishers. social semiotics, and systemic functional linguistics. She is the co-
author with Irene Hammerich of Developing online content: The
CLAIRE HARRISON, president of CANDO Career Solutions, principles of writing and editing for the Web and has published
Inc., is an award-winning writer and editor with 25 years’ experi- papers on hypertext, professional writing, and Critical Discourse
ence working for government and corporate clients. She recently Analysis (CDA). Contact information: [email protected]

60 TechnicalCOMMUNICATION • Volume 50, Number 1, February 2003