in literature. Nevertheless, one has to realize that there are many interpretations of the truth out there; readers will have to come to their own conclusions about mine.
My interest in this part of European history is, in part, personal. My father’s research into our family tree revealed the fact that, five generations ago, Gerrit Janz Kraak, the brother of my grandfather’s great-grandfather, took part in the Russian campaign as a soldier in Napoleon’s army. He died in Russia, at the battle on the Berezina River (now in Belarus), on November 27, 1812. This explains why the battle features prominently in so many of the book’s examples.
This book’s adventure began with my first sabbatical in 2011. The free time gave me the opportunity to visit the French military archives so that I could find proof of my ancestor’s misfortune on the banks of the river in Belarus where the French army crossed. A publishing deal with Esri Press luckily materialized at this time.
Although I am the sole author of this book, I know that we never really write a book on our own. The task requires many contributors. The greatest demand fell on my family, who had to get used to the phrase: “No time, have to work on the book.” Patiently, they endured. Thanks to Marijke, Eelke, Laura, and Emma.
Many of my students contributed to the datasets that I used to create the book’s illustrations. Staff members of my research groups contributed as well. Bas Restios created the software that made some of the illustrations possible, while Willy Kock processed the complex data. Discussions with Connie Blok and Corne van Elzakker helped to improve the chapter on maps.
Three people deserve a separate word of thanks. First, Irma Kveladze did most of the work preparing the Napoleon-related data for the flow maps and space-time cube. Although she used this dataset in her own doctoral research, my requests may have distracted her from her own work. We have discussed virtually all of the book’s figures from a design perspective. Above and beyond the call of duty, she willingly accompanied me to Belarus, where her skills in Russian helped me to understand our guide.
Second, Otto Huisman contributed to the quality of English in this book. Any awkwardness in language caused by the transition from Dutch to English cannot be blamed on him. As he read the book, Otto also became a fantastic sparring partner. His interests in analytical time geography have certainly enriched the book.
Last, but not least, I am indebted to Alan MacEachren. Our common research interests go back almost twenty years. His activities at Penn State University have always inspired me. He was willing to read the book and offer his critical comments, which focused less on language and style and more on fundamental concepts. Processing his remarks has certainly improved the reasoning in the book. This challenged me to sharpen and clarify statements and rethink and sometimes adapt viewpoints.
I hope the reader will find useful information on how to map time and, in the process, enjoy a bit of history.
Menno-Jan Kraak
Enschede, Netherlands
January 2014
Introduction
Maps tell time
All maps tell time. They portray a particular moment in the past, present, or even the future. For example, a map might show the boundaries of European countries in the early nineteenth century. It might depict a street plan or a future high-speed rail network in the present.
How maps represent time, however, can be ambiguous. The date 2012 on a street map, for example, may refer to the moment when the map’s dataset was collected, the date that the map was drawn, or the date that the final version was published. Provenance matters, too. A map depicting the spread of the Bubonic Plague in fourteenth century Europe might be a new map based on current knowledge or an old map rooted in the past. Maps depicting future railroad plans may have been composed today or at the end of the last century. References to time, like space, are also scale dependent. A map reader must pay attention to the temporal units used, such as weeks, months, or years. Sometimes, one is left to puzzle over which calendar system mapmakers have applied.
People today have high expectations for the maps they use. Whenever they look at maps, they expect real-time content, especially from maps displayed online. Of course, such expectations cannot be met in all circumstances, which vary according to both location and topic. The proliferation of in situ and human sensors promises to fully realize these expectations. For example, current technology allows people to observe in near real-time the water levels in Dutch rivers or the whereabouts of people through social media.
The maps featured here provide snapshots that display moments in time. Often, people look at maps in order to understand change itself, in processes and dynamics. Better-known examples of these kinds of maps come from social geography (such as migration, trade, or traffic) and physical geography (such as landslides, continental drift, or weather). In creating dynamic maps, cartographers ask questions such as the following: How have the borders in Europe changed since the nineteenth century? What is the traffic situation in the city during the day? When will the thunderstorm reach the recreation area?
The predominance of questions with a temporal component has grown, motivated by the increasing availability of information, which has stimulated demand for information still further. How do we best map change? More specifically, how do we design a map so that its temporal component properly narrates the story of change? This question, for me, owes its inspiration to Charles Joseph Minard’s map of the French invasion of Russia (see figure I-1). His map has enjoyed longstanding fame in both statistics (Funkhouser 1937) and cartography (Robinson 1967). Edward Tufte, an expert on information design, has helped to make it more widely known. In his book, The Visual Display of Quantitative Information (Tufte 1983), he analyzes Minard’s map in a section devoted to the “Narrative Graphics of Space and Time,” observing “how multivariate complexity can be subtly integrated into graphical architecture, integrated so gently and unobtrusively that the viewers are hardly aware that they are looking into a world of four or five dimensions” (Tufte 1983).
Tufte’s contention that Minard’s map “may well be the best statistical graphic ever drawn” (Tufte 1983) probably encouraged many readers to use this map to create their own variations of it using modern techniques (see chapter 2, section 2.2). Moreover, the quote implies a challenge: Can Minard’s map be improved?
Minard’s map originated as one of two that he drew in 1869 (see figure 1-1). The now-forgotten second chart depicts Hannibal’s Italian campaign in 218 BC during the Second Punic War. Together, the maps compare the huge losses that Hannibal’s and Napoleon’s armies suffered; Hannibal’s army of 96,000 soldiers shrank to 26,000 but fared better than Napoleon’s, whose 422,000 soldiers were decimated to a mere 10,000. Minard created his maps to protest the senselessness of war, something he personally experienced during the Napoleonic Wars at the siege of Antwerp in 1813 while he was posted there.
Figure I-1. Minard’s map: Top, Hannibal’s Italian campaign in 218 BC during the Second Punic War (“Carte figurative des pertes successives en hommes de l’armé qu’Annibal conduisit d’Espagne en Italie en traversant les Gaules [selon Polybe]”). Bottom, Napoleon’s Russian campaign in 1812 (“Carte Figurative des pertes successives en hommes de l’Armée Française dans la campagne de Russie 1812-1813”), published in 1869.
Minard’s map, by design a simplification and abstraction of reality, combines point, line, area symbols, and text. These symbols represent geographic objects, like houses, rivers, or administrative regions. Three components characterize each object: location, attribute, and time. Figure I-2a provides an example of Minard’s
