There is little doubt that the Mercator projection is one of the most popular world map projections of the modern time. For some people, it is the only world map they have ever known, the one that decorated their classroom wall at school. There is, however, also equally little doubt that it has been constantly criticized in the last few decades for being everything from ‘out of touch,’ highly misleading, to downright Eurocentric and racist. The crux of the argument has always been the same: The immense distortion of the size of Europe and North America, leaving the rest of the world (often represented by Africa) seemingly small and powerless. We have all heard the reveal: Africa can easily fit around 14 Greenlands, and Alaska is not more than one-fifth of Brazil, among many other comparisons.
These common misconceptions apparently bothered Boston Public Schools enough that in 2017, they entered the news as they announced retiring the Mercator map in favor of the Gall-Peters projection (see below), a map that they deemed free of imperialist agenda and capable of representing the Earth proportionately. Should we follow suit?
It would be careless to start comparing maps without first going through the basics of maps, including what a map actually is. A sidenote: What we often refer to as a world map is actually shorthand for a world map projection, because the method of portraying our spherical (or more correctly, ellipsoidal) Earth onto a two-dimensional plane is called a map projection. However, for the sake of brevity, I would also refer to it here and there in this article as merely world map.
Let us now determine what makes a map a map. We know a map consists of locations, and we know we often use maps to show directions. Judith A. Tyner weighed a couple of characteristics that are often used to define maps in her book The World of Maps: Map Reading and Interpretation for the 21st Century. Are maps always drawn to scale? No. When we draw one to show where the nearby bank is, we rely more on other elements on the drawing, such as intersections, parks or post office, to pinpoint the location. Can a map be of a photograph instead of a drawing? Yes. Think of event invitations with maps printed directly from Google Earth. Tyner then concluded that maps, simply put, are “spatial representations of information.” This information is coded in symbols, just like how treasure maps often have the X mark on the location where people have to start digging. The symbols go further — next to the “X” is a body of water, surrounded by trees, on the opposite side is a small hill — and they visualize the necessary information, as well as describe the spatial relation of one another. Eventually, a map can also be viewed as a means of communication: The information acts as the message delivered by the mapmaker, which will be decoded by the user at the other end.
However, it would be naïve to assume maps are purely spatial visualizations. In Rethinking Maps, Martin Dodge, Rob Kitchin and Chris Perkins express several arguments put forward by cartography scholars about how maps are powerful objects. One of them, John Brian Harley, a prominent figure in establishing the field of critical cartography, claims that in creating a map, mapmakers create knowledge rather than simply revealing them. After all, in our earlier communication scheme, the mapmaker is in charge of constructing the message, and we as the users will gain information as much or as little as the mapmaker decided to let us know. Yet, as sinister as this might sound, this effect is inevitable. Just like in news reporting, even the news items we believed to be rather neutral have undergone several subjective considerations, such as what information should be included, what this item aims to achieve and how its presentation should be. It is then imperative for a mapmaker to be aware of the context surrounding his creation and the political notions that may arise from his decisions.
“Every map is someone’s way of getting you to look at the world his or her way.”
— Lucy Fellowes, Smithsonian curator (quoted from Rethinking Maps)
This possibility of political undertone is what drove Boston Public Schools to denounce their Mercator projections. And on paper, there is a lot to admire about it. Why have our children been looking at a map that cannot even tell them the accurate size of countries and continents? Specifically, why are we giving our underrepresented students a world map that further reduces their existence one way or another? It is then logical to stand up, point out injustice and introduce change.
As teachers, you would not want your students to feel “lied to,” and yet, the Mercator projection successfully does that. Eric Gutstein, during his stint as a middle school teacher in a Mexican immigrant community in Chicago, Illinois wrote about his class project with the Mercator and the Gall-Peters map projections. The reactions from his students were particularly strong. A lot of them were quick to indicate history being “written by the white people,” with the Mercator map serving as a vehicle for white superiority agenda. And as the minority, they felt pushed to the side and being told they were insignificant and “didn’t deserve anything.” They were upset, and, judging by their assessment of the map, rightfully so.
One of Gutstein’s students, Rosa, was quoted asking him why teachers never told them how wrong the Mercator map was. There is one glaring problem with her question: It presupposed that the Mercator projection is wrong.
It is not wrong, despite all the area size inaccuracies and the negative reactions it has sparked over the years. J. Edward Britton wrote a piece defending the Mercator projection against racism accusations, saying, “Discrediting this claim requires only minimal knowledge of cartography.” And I agree. So here is what we need to know.
Every map in existence has limitations. Maps cannot portray every single detail of the area they are representing, so they have to filter out elements of little importance. They have to scale down buildings and land forms to fit, in which they reduce these into symbols. It is mathematically and practically impossible to create an exact projection of a sphere onto a two-dimensional plane, and thus distortions cannot be avoided. All these factors result in maps being developed for only one specific purpose at a time: A map may preserve angles, area sizes, distances or none at all. According to these properties, maps are categorized as conformal, equal area, equidistant and compromise. The following descriptions are summarized from the work of Fritz Kessler and Sarah Battersby, Working with Map Projections: A Guide to Their Selection. (Note: I decided to exclude azimuthal, due to the fact that some sources list it under projection class instead of property.)
- Conformal. A conformal map projection preserves angles, in which any angle on a conformal map is true to their corresponding equivalent on Earth. If city A, city B and city C create a 60° angle at city B on a globe, this will be shown as such on a conformal projection. When we look for directions, we would want the 90° angle between road X and road Y in real life to also be portrayed as 90° on our map for easier navigation. This is not possible on a non-conformal map.
- Equal area. As hinted by its name, an equal area projection preserves the sizes of areas as how they exist on the globe. It is therefore more reasonable to compare sizes of regions with an equal area map rather than with any other projection type, as it preserves landmass size most accurately. Shapes, however, will still be distorted.
- Equidistant. An equidistant map preserves distance, but only within a specific parameter. Some equidistant projections allow accurate distance measurement between two points on a meridian, which is the vertical line on our maps also known as a longitude (the equivalent of latitude, on the other hand, is called a parallel). Others require a central point, from which the distance is correctly measured. An equidistant map with London as its center, for example, accurately presents distances between London and any other point on the map, but not between Paris and Lisbon.
- Compromise. Compromise map projections do not preserve any property and are therefore not appropriate for any calculation. However, they are the most suitable for general reference purposes compared to any of the previous types. A compromise map does not need to strictly maintain its accuracy, and as such, its priority is on reducing visual distortions to achieve landmass shapes as similar as possible as these on the globe.
At this point, we might start to realize where our educators have gone wrong. But there’s a reason why the Mercator projection has endured centuries of fame, and the only way to understand it is to look into its history. I will detail this according to the work of Mark Monmonier, Rhumb Lines and Map Wars: A Social History of the Mercator Projection.
A student in Gutstein’s project remarked in disheartenment, “I just want to understand what [is] the point exactly of Mercator’s map.” (Note: I will quote Gutstein’s students multiple times in this article, not to prove how wrong they are, but simply because their arguments fully represent what most of us think about the Mercator projection.)
The answer, ironically enough, lies in the map’s name. What we now refer to as “the Mercator projection” was originally published in August 1569 under the name Nova et aucta orbis terrae descriptio ad usum navigantium emendata accomodata. In English, this translates into New and More Complete Representation of The Terrestrial Globe Properly Adapted for Use in Navigation.
Gerhard Cremer was born in March 1512 in Rupelmonde, Flanders to a family with little money. With the help of his priest uncle who had good connections in the city, Cremer attended a monastic school. Upon entering university in 1530, he latinized his name into Gerardus Mercator, with mercator being the equivalent of the German cremer ‘merchant’. He studied humanities and philosophy, while maintaining interest in mathematics. After graduating, Mercator was involved in practical projects under the supervision of the revered mathematician and astronomer Gemma Frisius. It was during this time that he started crafting globes, before eventually tried his hands at mapmaking. Mercator was around the age of 25 when he first published his own map — a map of Palestine, possibly influenced by his theological upbringing. The following year, his first ever world map followed suit. It had a double heart shape and provided a view of the Earth from both the North and South Pole. Mercator would then continue responding to regional mapmaking requests, first from merchants and later influential figures.
Mercator’s famous 1569 world map came in a set of eighteen paper sheets. Expanded, it reaches 202 cm x 124 cm (80" x 48") in size. It skillfully addressed the needs of mariners in the age of Earth exploration, in addition to making their journeys easier and more reliable. Prior to Mercator’s chart, there was a strong chance of being lost during long distance sailings, especially across the ocean. Early sailing maps only displayed approximate sailing directions and rough distance measurements. These were not accurate, but often they covered small areas only, and thus their mistakes were rather inconsequential. This was extensively improved by Mercator. With his projection, seafarers would first mark their start and end point, and then connect these with a straight line. They would measure the angle between said line and any of the meridians it cuts through. Finally, upon sailing, they would only need to maintain this angle from their starting point until they arrived at their destination — all within the level of accuracy that was unheard of before. Needless to say, the Mercator projection revolutionized navigation.
But is it possible for the Mercator projection to preserve area sizes?
The Mercator projection is a conformal map. Its only purpose is in the field of navigation, and it is not capable of portraying landmass sizes accurately, nor was it ever intended to do so. It has little to do in introducing children to our Earth. And as we will see, it has equally little to do with the accusations brought against it.
Now that we have known about map projection properties, it is hard to overlook the fact that we barely hear about the Mercator projection being criticized for not preserving distances. This shows how biased our perception can be due to limited knowledge. Unlike our ancestors, we do not normally use world maps as a navigational means. A lot us were also never taught about the properties of map projections at school, despite spending a substantial amount of time looking at one in Geography class. Thus, we are oblivious to the fact that a world map is not merely a portrayal of the Earth in regard to how our continents look and compare to one another. When a map shows the Earth’s landmasses just as how we wanted them to be, the map is right. When a map follows a valid approach that we are not aware of, and thus distorts the shapes we had in our mind, the map is wrong. Out of sight, out of mind.
But let us return to the technicalities of Mercator’s invention. The Mercator projection loosely follows an equirectangular projection, in which it displays its meridians and parallels as straight lines. (In true equirectangular projection, these lines are also always in equal spacing from one another. This is not the case with the Mercator map). Since the meridians and parallels of a sphere are naturally curved, ironing these flat would mean stretching or compressing certain areas. However, a mapmaker has the option to choose one particular area that they want to maintain as faithful as possible. On the Mercator projection, this is taken by the equator, which lies exactly in the middle between the North and South Pole. No other line shares similarity with the equator, and therefore, everything else has to be adjusted according to it. The farther north or south a line is from the equator, the more noticeable the distortions around it will be. This is because on a globe, meridians converge and parallels shrink in size as they depart from the equator, therefore becoming less and less similar to the equator. As a result, on any Mercator projection, the southern hemisphere is as distorted as the northern hemisphere.
And yet, all we hear about is how massively enlarged Europe and North America are on the Mercator map, without any acknowledgment that certain parts of Africa, too, are bigger than their actual size. This can be visually compared on the website The True Size Of …, whose purpose is to demonstrate how landmass sizes on the Mercator projection greatly depend on their latitude. If we drag, for example, Morocco and South Africa to the equator, these will also shrink, even if only modestly. But we never heard about how the Mercator map inflates northern and southern Africa, and we all know why.
The next issue to address is summed up perfectly in this quote from Gutstein’s student: “Because the map maker was from Europe, he put Europe in the ‘middle of the world’ and on top of other countries.”
Even if Mercator wanted to be a Eurocentrist, the Catholic Church as the bearer of high authority in Europe during the Middle Ages surely only made it harder for him to be one. In March 1544, extremist followers of the Church held Mercator in prison under the suspicion of joining Protestantism. Seven months later, he was released due to lack of proof, which saved him from being beheaded, burned at stake or buried alive. Only after he moved to Duisburg, Germany did he escape the tumultuous religious conflict and the constant threat of persecution. Still, his treatise on world history, titled Chronology (1569), was banned by the Church for mentioning leading figures of Protestantism in his effort to be as accurate as possible. As far as Mercator’s own historical records go, there is no solid evidence on whether he left the Catholic Church or not.
We may never know exactly whether Mercator thought Europe deserved to be the single ruling power on Earth. What we can prove, however, is that he did not even put Europe in the middle of his map. On both the original 1569 projection and the modern adaptation on the top of this article, if we place our sight on the center of the map, Africa shyly greets us from the right hand side. So where is this map that is supposed to have Europe at its heart?
On the hands of map publishers, according to Bob Abramms, founder of the map publisher and distributor ODT Maps. In his correspondence with NPR, Abramms explained that it is often the individual publisher’s choice to crop Antarctica off the map due to how enormous it appears on the Mercator projection. It does save ink and paper, but also results in the equator being on the bottom third of the map. As it turned out, this decision gave birth to a whole new argument from the same perspective, as is mentioned on a CNN article: The Mercator map lowers the equator to the bottom third in order to provide more detail on Europe. Thus, it is Eurocentric.
Here, we have to admit that the 1569 projection does have its equator placed suspiciously south from the center of the map. But this was most probably a logical decision — Mercator could not possibly put on his projection what he did not know in the first place. Based on how crudely drawn the lower ends of the map are, we could make the assumption that a large portion of the southern hemisphere had not been discovered yet. And this is true. Australia, New Zealand and Antarctica had not been sighted until 1606, 1642 and 1773, respectively.
But what about Mercator’s double heart-shaped world map? Did we not see a continent at the core of its southern hemisphere? Yes, Mercator placed a landmass there, but no, it was not Antarctica. For maps that were not strictly made for navigation such as the Mercator 1538 map, cartographers often illustrated regions merely to suggest the existence of a location. This is also apparent on, for example, a projection by Abraham Ortelius in 1570 (see below). Both maps depict a southernmost continent that clearly lacks details, signaling that not much was known about it.
I will now end our excursion with the final aspect of map limitations: User limitations. Tyner wrote, “A user who lacks map reading skills may misinterpret a map or ‘read into’ the map information that can’t be learned from the map alone by projecting his or her stereotypes and misinformation onto the map.” A politically nuanced map can trigger a reaction in us just as easy as our politically nuanced mindset can project itself onto an object. In an Al-Jazeera piece, Nick Danforth shared his observation regarding this subject. He noted that despite how extremely inflated Greenland, Antarctica and Canada are on the Mercator projection, we do not normally think of these as a superpower or a threat. A third factor is at play, and it is our own biases.
“Thinking that we were all raised with the Mercator map makes me feel kind of miseducated,” Gloria, another of Gutstein’s students, said, “because all these years we’ve been using the wrong map.”
Gloria was right, but probably not in the way she thought she was right. Her educators did not inform themselves properly about maps prior to passing their knowledge to their students, and as a result, they used a map that did not suit their intention at all. The real problem with the Mercator projection is us.
I would also add some words on Boston Public Schools’ new map. The Gall-Peters projection is admittedly an equal area projection. However, it is not and had never been the only equal area projection, as is incorrectly reported in this article. Originally named the Peters projection, the map was introduced by Arno Peters, a German filmmaker-turned-historian, at a press conference in May 1973. Yet, it proved to be essentially identical to a projection invented by James Gall in 1885. (This was a rather peculiar case: Peters was found providing a technical description that contradicted his projection. Two years later, a revised version of the map was published, which only made it exactly identical to Gall’s.) Regardless of whether this was a case of dishonesty or pure serendipity, the Lambert cylindrical equal area, announced by Swiss mathematician Johann Heinrich Lambert, had predated Gall’s Orthographic projection by 113 years. Additionally, in Flattening the Earth: Two Thousand Years of Map Projections, cartographer John Snyder suggested that both the Bonne projection and the Stab-Werner projection had their origins in the 1500s.
Despite the clarifications above, my aim was not to convince anyone that the Mercator map is a superior creation compared to Gall-Peters. Quite the contrary, I would encourage everyone to choose their own preferred map projection, because this means that an effort to understand maps is made. Those who wish to continue using their Gall-Peters projection should be allowed to do so, provided that all parties are aware of its controversies and reached said agreement together. Every world map is wrong as much as it is right, as long as its spatial information is faithful to reality. What we should not forget, however, is to always be mindful and transparent about the map we are using: Which property it preserves, why we use it, which events led to its creation, what connotation it possibly carries, and whether this is factually justified or not.
Our children need to learn about maps, but not so that they hold prejudice towards one another. Instead of limiting them to only one projection, show them different maps from different perspectives — let them learn that maps do not always have to be rectangular, west does not always belong to the West, and that north does not always have to be up. Explain to them that we have our distinct places of origin and that we are indeed different in ways more than one, but also remind them that a sheet of paper with territory markings should not determine how we treat each other. And finally, tell them that if we look at the Earth from its poles, we are much more connected than we think.
Thank you very much for reading this far! I would like to hear your thoughts on this topic, as well as corrections. I intend to have every content accessible without any paywall, so I welcome donations instead on Ko-fi. Alternatively, feel free to share this article if you find it useful :)