War is how Americans learn geography:

Visual literacy in a media savvy world

 

Annie Zeidman

IS 209 (formerly 246)

Fall 2001

 

Introduction

Most of us have spent quite a lot of time learning to be critical thinkers and readers of text.  We need to apply these skills to visual information as well.  The internet has challenged those of us who consider ourselves to be savvy consumers of information to apply the skills developed for printed material to a new format, namely the world wide web.  Just as we have done with books and journals, we look at online material and ask questions about the publisher, the author, the quality of the information, and much more.  How many of us do that when we are watching television, movies, or reading maps?  Certainly the widespread appeal and dissemination of visual media has prompted many people to examine what they are watching critically, although they are certainly still a minority.[i]  How many people ask critical questions about the maps they get from the Automobile Association of America (AAA) when they go on a road trip, or the statistical maps generated by the Census, or the graphic maps depicting the war in Afghanistan?  This essay will briefly discuss historical cartographic traditions.  Then, by examining a number of images, primarily ones that were generated by the print media during the U.S. bombing of Afghanistan in 2001, we will analyze some of their significant elements as a way of demonstrating cartographic conventions and how maps convey information.

History of maps

 

Maps are visual representations of things that are often hard to experience first hand.  We understand, at least theoretically, the idea of looking at a city or the world from high above it, especially those of us who have looked out of an airplane window.  But, we are rarely able to verify this information for ourselves.  Patrick Wilson raises the questions: how do we know what is trustworthy information, what do we trust, and what do we do with the information? (Wilson, 1977, 13)  Medieval maps of the world, called mappaemundi, among the oldest surviving example of cartography, portray a world that closely resembles the one described in the Bible, important in a world that was largely run by the Church.  For example, the “known” land (Europe, Africa and Asia) and the unknown (labeled Terra Incognita) are situated at the four corners of the rectangular map, since the apostles in the Bible were thought to have spread the gospel to the four corners of the world. (Wood, 91) 

The next innovation in map making, following Ptolemy’s texts in the 14th century, introduced elements like putting North at the top and using longitude and latitude lines.  These maps gained in popularity, in part, because they were invaluable in the expansion of trade.(Ibid)  The emerging nation states in Europe during the 18th and 19th century established more conventions still in use today.  In fact, Wood points out:

North is at the top, zero degree longitude runs through Greenwich, England, and the maps are centered on Western Europe, North America or the North Atlantic.  The resulting configuration has become so familiar that few people notice just how arbitrary it is.(Ibid) 

 

With this in mind, Wood cautions us against assuming that maps have progressed steadily and naturally towards an unbiased truth.  Where it is now obvious that the medieval maps (or any number of other maps from other traditions) contain inaccuracies and betray the subjectivity of the creator, it is a mistake to conclude that we are no longer dealing with the same issues.  Consider the GeoSphere world map (figure 1), composed entirely of images from the TIROS-N satellite operated by the National Oceanic and Atmospheric Administration (NOAA).  The author of the map readily admits that he selected only images that didn’t have clouds, or deleted cloud cover pixel by pixel, he enhanced the colors of vegetation, rivers were made thicker for visibility, and so on.

Wilson reminds us that our perception of the truth depends on where we, the observer, are situated personally.  (Wilson, 1983, 5).  How do we assign cognitive authority to something if we can’t see it or experience it, for ourselves?  Often we rely on other people’s accounts or interpretations of what’s happened, with a healthy dose of skepticism and our own understanding of the world to make sense of it all.  While we have been taught to think critically about words, we have rarely been exposed to the same sort of rigorous training when it comes to visual images.  Let us now focus on the use of maps in print media, especially those produced for the current war in Afghanistan.  Although this essay discusses content used primarily in print journalism, not television, many of the fundamental issues are the same.

The geography of war

Several reporters have noted recently that “war is the way Americans learn geography”.[ii]  Certainly, there is ample evidence of our geographic ignorance, one study demonstrated that 25% of high school seniors in Dallas, Texas couldn’t identify Mexico as the nearest foreign country.  Sadly, this doesn’t seem to be a recent trend, for just 2 years after the end of World War II, 28% of adults polled couldn’t identify Italy (Monmonier, 1989, 23-4).  In spite of our ignorance, or perhaps because of it, the news media uses maps as one of several ways to package the news.  And, although some cartographers feel that these maps are treated more like eye candy then news, it is interesting that they persist to this day (ibid, 176).  In our modern era, war seems to drive much of the cartographic applications and innovation.

It is a time honored tradition to use maps to bolster public opinion about the success of a military campaign.  And, conversely, maps, can be a powerful graphic tool used to discourage your enemies, even if you don’t speak the same language.  In 1943, Louis O. Quam wrote an article entitled “The use of maps in propaganda” in the Journal of Geography, about these applications.  In it, he explained the characteristics of “the propaganda map,” most of which are still quite relevant today.  Although most of his examples were taken from Nazi Germany, we will shortly see how the same cartographic tools are used in the war currently being waged against Afghanistan.

Before concluding that visual literacy is a purely aesthetic pursuit, it is worth keeping in mind that mapping needs that arise during war have demanded and supported the development of Global Positioning Systems and much of the map making that is now done electronically.  All sides need data, the lay of the land, its resources and hazards to fight effectively. (Clarke, 1992)  This information is so important the reminds us about the mapping subterfuge that characterized the Cold War.  The Soviet Union tried to maintain an advantage by deliberately misrepresenting natural features, transportation lines, boundaries, and villages in maps sold to the public.  They worried that if their enemies had accurate information they would be vulnerable to attack.  Maintaining a second set of detailed accurate maps is both expensive, and with the rise of satellite imagery, increasingly less effective. (Monmonier, 1996, 115-118)

 

Maps examined

 

As a way of seeing how cartographic practices are used let’s look at two maps of Afghanistan published by the Central Intelligence Agency, approximately 10 years apart.  Figure 2, (CIA 1972); Figure 3 (CIA, 1982).  Both show six cartographic images on a 45 x 58 cm sheet (approximately 18 x 23 inches)[iii].  The largest map shows terrain (but not topographic lines), political boundaries, names of the capitals, and railroad tracks, airfields, roads.  Five other maps show ethnolinguistic groupings, economic activity and land use, population, a representation of the earth and the relative size of the country compared to states in the US.  It is interesting to note that although the borders of the country did not change in those 10 years, the earlier map shows that the country could be contained by the state of Texas (figure 4), while the later map shows how the country would dominate the Southeast and the Gulf of Mexico (figure 5).  In the intervening decade, Afghanistan is a menace threatening to devastate cultural bastions like New Orleans, Memphis, Atlanta and the District of Columbia, our nation’s capital.  It is important to notice that the use of borders and labels can greatly effect our impression.  Neither representation is wrong cartographically speaking, and yet the crisp authoritative lines of the borders and familiar towns conveys a very different message in each of these two maps.(Monmonier, 1996, especially chapters 6 and 7)

Let’s examine the population maps shown in the upper right corner, in more detail.  The one published in 1972 uses a yellow to dark green color ramp. (figure 6)  The one from the next decade shows population on a scale from yellow to dark orange. (figure 7)  Using color intensity can be a helpful way of showing information that can be digested at a glance.  Such a map is called a choropleth, and it is commonly used to show population density in an area where the “relative darkness of the symbol shows the concentration of population on the land”(Monmonier, 1996, 22-23 ). The use of color can be helpful, but again, there is a long standing cartographic tradition of encouraging map readers to associate certain colors with both environmental and political factors depending on the context.  Green is associated with vegetation and wealth; yellow is commonly used to represent its environmental opposite, desert; blue is often used for water; and brown regularly marks mountains or features with elevation, the darker the color the higher the elevation.  Red has been used to represent power, fire, hazards, Communism, and much more (ibid, 170-1). These meanings may be unintended or deliberate, there is no clear indication that everyone reacts the same to every color.  The wary map reader should be careful to remember that while colors help distinguish borders, like those between land and sea, they can be deceptive.  One only has to remember that that mountains can support significant amounts of green vegetation even at higher elevations and tourist maps showing sparkling blue water where no such color exists in that location to get a sense of why it is worth being careful about your assumptions (Monmonier, 1996, 170-2).  With that in mind, it seems significant that during the decade that separates these two maps, the Soviet Union invaded and occupied the country from 1979 until 1989 (World Factbook, 2001, Afghanistan, “Background”).  And if color was going to be our sole determination of the impact of this change, Afghanistan appears to have gone from being a lushly vegetated and well populated country to a desert wasteland punctuated by potentially hazardous enclaves of possible Communist activity.  This change is apparent even though both of the color scales start with yellow on one end (the least populated), and get darker.

The authors of this map provide a clearly labeled map legend in the lower right hand portion of the map.  This is a common position for legends, these are easy to find, and appear to follow standard cartographic practices.  Given the notoriously destructive Soviet invasion and subsequent occupation the well informed reader would expect that this would take its toll on the population density.  Perhaps the 1982 map colors were meant to reflect the demographic shift that takes place during war?  However, the post Soviet occupation map clearly states in the legend that it is “Based on 1960 source” [sic] (figure 7).  Any data that is 22 years old, especially when there has been a significant change in conditions, like a war, seems suspect.  The earlier map doesn’t cite its source, but it seems likely that if they had fresher data in 1972, that they would have used it for the 1982 map as well.  But, the map clearly shows that there has been a downward shift in the population over time, hasn’t there?

In addition to the change in color, a careful comparison between the two maps shows that the earlier map divides the population into 5 categories, while the later one has only four.  As a result of this change, half of the scale in the 1982 map represents areas with 0-130 persons per square mile with pale and medium yellows.  Whereas, in 1972, the same number of people per square mile accounted for 3/5th of the population color ramp and was shaded yellow, light and medium green.  These sorts of changes are incredibly easy to execute on computerized maps.  Two of the most common ways of dividing data are into equal intervals or quartiles.  In equal-intervals, data is divided into equal parts without regard for the distribution of the data.  If the data is evenly distributed this method works just fine.  But, if the data is grouped around particular points, its possible that some of the divisions are empty.  This method’s most useful feature is that it is easy to calculate (Monmonier, 1996, 146-7).  Quartiles can use any number of divisions to divide the data into categories with roughly the same amount of the variable (in this case, persons) per square mile or kilometer or any other areal unit.  This scheme may make the map pattern seem more balanced, but it can also create breaks that are really very close or far too large to be descriptive (ibid, 147).  In the case of our Afghani population maps, to assign two classes to differentiate between areas with 2.6 and 26 people per square mile and then assign only one category to distinguish between 130 and 520 people per square mile shows how the quartile classifications can make data look smooth, when in fact, they are not. To further confuse the reader, in 1972, persons per square mile is at the top of the legend, kilometers on the bottom whereas, the 1982 map reverses these measurements.  These two maps show how similar (or possibly, identical) information can be manipulated to achieve dramatically different results.

 

The “selective” use of information

 

A population map from 1960s era data probably does not accurately display the current state of the Afghani people.  Since then, the country has faced the sorts of situations that change population dramatically.  Sadly, these same challenges, namely war, lack of food, uncertainty about the future, weakened infrastructure, and the like, also make it difficult to conduct accurate census counts.  The map in figure 8 shows population estimates for 2000 and is produced by the United Nations Development Programme (UNDP) and Programme Management Information System (ProMIS), based on data from the Oak Ridge National Laboratory LandScan Global Population 2000 Dataset.  This dataset estimates population using local census data and statistics.  Now that we have a newer population map, we find that comparing figure 8 to Figures 2 and 3 is problematic. The UNDP map does not label any cities, but rather the provinces, making all but the most rudimentary comparisons difficult.  The ten colors representing the data classifications make analysis more difficult, even if they are more accurate than the CIA maps, and we have no indication as to what the areal units might be.  It seems that according to this map, the south west of the country remains sparsely populated.  Population remains clustered in areas around the northern side of the country, although there seems to be an increase in the population south and west of Kabul. 

            Keep this new population map in mind while looking at maps generated by USA Today to show the progress of the Northern Alliance in the first 30 days of fighting.  A look at these maps (figure 9, figure 10 and figure 11) shows that although there were decided changes in the location of the bombings, the area controlled by the Northern Alliance, displayed so prominently, did not change until day 38 and 39, November 13th – 14th, 2001 (figure 12 and 13).  Moreover, the north eastern area controlled by the Northern Alliance, although it looks like an impressive piece of land mass, is fairly sparsely populated until its southwestern most part.  One of the other cities controlled by the Alliance, Changhcharan, located in the western part of the country, seems to be well populated, but the areas to its north east, are not.  Moreover, the USA Today maps fail to show clearly, although the city labels surrounding the map allude to it, that the areas controlled by the Northern Alliance were wholly surrounded by well populated areas that were not held by them. 

Use of symbols

Maps used in the news have a wider audience than any other form of cartography and must therefore be easy to understand without specialized training.  In our image saturated world, it seems that maps are used to break up text when there is “the absence of a suitable photograph.” (Monmonier, 1989, 14)  On the cover of the Los Angeles Times on November 15th, 2001, we have a clear example of just this phenomenon. (figure 14)  In the center of a lot of text, we have a map with crisp national borders (more on that below), areas indicating the shrinking amount of territory under the control of the Taliban and arrows showing “Taliban fighters fleeing.”  A quick look at the 2000 population map (figure 8) shows that the areas under Taliban rule are among the most heavily populated in the country.  The use of arrows is a time honored way of showing movement on a static map.  Any number of different kinds of arrows can be used to suggest broken blockades, attacks and retreats.  Certainly the simple, heavy arrows on this map are far less sophisticated then the gently fading “animated” flash sequences on USA Today’s online maps.[iv]  However, one of the advantages to using arrows on maps is that they “…give the impression of a fait accompli, whether or not it is true.”(Tyner, 143)

One of the hallmarks of German propaganda maps in World War II was their simplicity. Quam’s 1943 article about the use of maps for propaganda, “the less geography portrayed on the map, the more likely it is to convey at a glance the message of the map.”(27)  It is worth noting here that one of the characteristics of maps used on television newscasts, is that due to the technical constraints of the camera and the low resolution, the images must be kept simple.  Type and symbols have to be bigger, the map is usually a line drawing and colors must be carefully chosen to ensure legibility (Caldwell, 39-48, 105-24, 180).

Boundary lines

 

Drawing boundary lines is one of the most powerful aspects of any map.  Geography professor Bernard Nietschmann is states that “more indigenous territory has been claimed by maps than by guns…And more indigenous territory can be reclaimed and defended by maps than by guns.” (Stone, 2)  Consider the map painted on the wall behind a truck carrying guns and solders depicting the north eastern region of Afghanistan.(figure 15)  Not only does the scene link the map and gun metaphor quite literally, but it also shows how boundary lines can be used.  The northern borders are painted in heavy, definitive black lines.  The land beyond those borders, namely Uzbekistan and Tajikistan,  is identified by name and a half hearted boundary line.  By contrast, the south east boarder, the one shared by Pakistan is thinner and dashed.  What we can see of that land is festooned with ornamental decorations and Arabic words.  Writing of propaganda maps in 1943 Quam wrote that “Strong smooth lines are employed to denote stable, natural, and friendly borders, broken irregular lines suggest weakness, and disunity or stress of hostile frontiers.” (25)  Boundary lines are also frequently used to reinforce assertions of power and control over land, its inhabitants and natural resources, such as when European countries mapped and divided up parts of Africa (Monmonier, 1996, 90).

 

“selection” of information

 

            All journalism is subject to the demands and interests of its respective editors and publishers, maps are no exception.  News coverage of war in general, and this one in particular, has been heavily censored[v].  If limited access to data wasn’t enough of a reason of view maps made during war skeptically, it is worth pointing out that most maps that are trying to be persuasive must select what data they show, in order to make their case as convincing as possible.  As Tyner explains, the selection of information can be a limitation of the scale (like the degree of detail) feasible for a map, or it can be a deliberate attempt to distort the image created, if not technically untrue.  She uses the example of a German school atlas produced by the Nazi’s that depicted a world map showing only the ethnicity of Germans in the Americas, Australia and South Africa, select major cities and the only other place names were ones that sounded Germanic (141).  Consider what Time and USA Today does with what must be similar information about the Afghani caves.  Time’s depiction of the Afghani caves emphasizes how well stocked and busy they are.  (figure 16)  In this cross section we see supplies laid in, labels explain what the rooms are used for, and how they offer protection against the variety of military weapons also labeled and shown trying to penetrate the thick stone.  By contrast USA today’s caves and bunkers look utterly inhospitable, rudimentary and sparsely populated. (figure 17, figure 18)   Time seems to want to emphasize Bin Laden’s cunning, and the US military prowess in the face of such a well organized foe.  USA Today characteristically seems to want to emphasize just how simple and exposed the Taliban forces are in their caves and bunkers.  Moreover, CNN goes out of its way to tell its readers on a map of the Northern Alliance’s successes, that the information shown on the map is essentially useless, perhaps in a misguided attempt to silence critics worried about leaking defense secrets to the enemy. (figure 19)

Like other kinds of news, or information, maps need to be carefully edited.  This can be especially challenging when editors see maps as art and not representations of data in need of review.  Monmonier maintains that many blunders are done out of cartographic ignorance and poor fact checking (1986, 31-33; 1996, 43).  Other authors suggest that is a simple matter of insufficient resources.  An article by Strobel and Pasternak for U.S. News and World Report about the National Imagery and Mapping Agency (NIMA) reports that there is simply too much data and not enough trained personnel to monitor the changes that are taking place, and then in turn, update the maps as necessary.  They contend that when the US bombed the Chinese Embassy in Belgrade in May of 1999, it was because the under funded NIMA had provided a 2 year old map of the area that “failed to note the Chinese Embassy’s move across town the year before” (25).  Although, in this particular case, one would be hard pressed to sort through the political motives and intrigue to find out if those bombs struck their intended target or not.

Conclusions

 

Our world is full of seductive images attempting to cajole and insinuate themselves into our conceptualization of reality.  Now, as much as ever, we need to educate ourselves about the power of those images.  We need to be critical consumers of images, as well as of the words we read.  Maps, in particular, should be viewed with an eye towards understanding how their message is famed and presented.  Far too often, people mistake the data portrayed in them as the only way to present that information and confuse it with an objective reality.

Consider Monmonier’s provocatively titled tome How to lie with maps.  His work is cited extensively throughout cartography and geography literature.  And lest one thinks that his assertions are useful only to paranoid, conspiracy theorists, consider the article on the Persian Gulf War by Clarke (1992).  Clarke asserts that cartography was not only crucial in that war but that the technology, although untried, was “close to perfect” in its execution.  Perhaps his version of cartographic history fared better then the smart bombs and mis-fired missiles, which prompted former president George Bush to coin the phrase “collateral damage” when referring to the civilian casualties.  However, if cartographers and others who work with maps regularly are well aware of the pitfalls of spatially representing data, they have not publicized it as widely as they could have.  But, perhaps this duty should fall to those of us who care about keeping all of us, both well informed and thoughtful in this new millennium.

works cited


[i] See Jesse Drew’s chapter “Media Activism and Radical Democracy” in Resisting the Virtual Life, Paul Messaris, “Visual Literacy vs. Visual Manipulation” in Critical Studies in Mass Communication, Ronald Betting’s Copyrighting Culture, just to name a few.

[ii] I heard Ira Glass say this on his radio show, “This American Life”, show #167, during minute 32:51-56, although he did not take credit for saying it.  Two other radio shows, The NewsHour with Jim Lehrer on November 23, 2001 and National Public Radio’s Weekend Edition Saturday on November 24, 2001 used the same quote with no author attribution.

[iii] The images included here are tiled together from 11 x 17 inch photocopies of the original.  The copies were then scanned and pieced together in Photoshop ™.  They are included in this paper for reference, but should be used with caution as the scale and details have been distorted in the reproduction process.

[iv] I am providing a URL, but fear that any minute now, they will return 404 errors.

http://www.usatoday.com/graphics/news/gra/alliedstrikes39/flash.htm

[v] For more on this please refer to Janet Kaaya’s paper for this course “Effects of media access restriction and censorship during war” or any number of the articles she has in her bibliography, i.e. Felicity Barringer ‘s “Reporters Want More Access, but Are Careful to Ask Nicely”, New York Times, October 22, 2001 available online at URL: <http://www.nytimes.com/2001/10/22/national/22MEDI.html > [December 6, 2001].

 

works cited