Mapping Milhaud’s Reception

The concept of musical mapping was fairly new to my colleagues and at the beginning of our work. In reading the scholarship of Bethany Nowviskie and Mark Monmonier, we established a set of guidelines for our maps. We embraced Bethany Nowviskie’s suggestion put forth in her work How to Play with Maps, that digital humanists reclaim map-making’s ludic qualities. Doing so allowed us to constantly ask the question “what if”. Employing this in our methodology enabled us to use our data creatively to create maps and explore innovative ways of communicating our findings with others.

What kind of maps to make?

The first question geographers must ask is “What type of map should I make?” While searching for the map that will most effectively convey the knowledge we’ve gathered, we must consider three options:

Data-Focused Map

A data-focused medium is useful when one needs to present a large amount of information. For example, our map of all of Milhaud’s performances contained over 500 data points. Within each point is contained information about the date, performing forces, venue, description, citation, and more. In order to present such information, we decided that the pop-up tools available on Arc GIS’s mapping platform would provide a useful tool when presenting this map. Luckily, Arc GIS software contains many features, including a feature that allows you to layer information on to one map with different color coded points. When exploring large data-sets, having one succinct viewing platform in which you can access a wealth of different information and display that information is different ways is invaluable. This type of data-focused map was the ideal candidate to showcase the information we compiled in our database. We have created several data-focused maps that can be found in our map gallery.

To be truly effective, however, it is necessary to format maps in the correct way. The Arc GIS mapping software has lots of little quirks that need to be ironed out in order to produce an effective map. There are two options when embedding a map into a website. One can use a web map or a web map application. A web map only displays the bare bones of the data set that one is trying to collect. Users are unable to toggle between layers or see a legend. In a web mapping application, however, both of these things are possible. Open the boxes below to see a comparison of a web map and a web mapping application and further analysis of these maps as they apply to Darius Milhaud.

The map below is one example of a data-focused map that we made to examine the spread of Milhaud’s ten most popular pieces of music. This type of map was ideal for this situation. Not only is the map able to display the large amount of information we catalogued about each performance, but we were also able to include media. Each marker contains an embedded YouTube video containing a performance of the piece that is represented by the marker. Click on any marker to explore the performances of Milhaud’s top ten pieces.

What does this map tell us about Milhaud?

Several interesting patterns and questions arise from this research. First of all, why did these ten pieces yield the most performances in our research? And why are most of them stage or theater works? Does this suggest that Milhaud’s work was best received alongside a visual component?

The map itself also reveals some interesting trends. Right away it is clear that Milhaud’s work was performed in a very wide variety of cities, especially in Northern Europe. This map seems to suggest that the scope of Milhaud’s music stretched far beyond the boundaries of the French border. Whether this was as a result of the music’s popularity abroad, or of self-promotion, is a research question that may be worth pursuing from this map.

Some piece-specific trends bring us to other questions. For instance, there are no performances of Les Choëphores that appear outside of Paris, Brussels, and Antwerp. There are also eighteen performances of Salade in Paris, compared to only five outside of Paris. For the other eight pieces, the performances spread much farther outside of France. Why didn’t these two pieces spread further beyond the French border? Were they received less warmly abroad than others? Or was this simply a hole in our research? The map also shows that Le Pauvre Matelot was the most widespread of all the pieces on the map. So, was it the most popular abroad, or did Milhaud push to have it performed the most? In addition, a comparison of the reach of  Milhaud’s ballets Salade, Le Train Bleu, and L’Homme et son desir demonstrates that each work was performed in completely different places outside of Paris. This means that the three ballet troupes that premiered and toured the works, Les Soirees de Paris, the Ballets Russes, and the Ballets Suedois, must have had unique touring and performing history even though each company was incredibly similar in structure.  These questions and conclusions show how maps can emphasize new patterns and lead to new scholarship.

Why Data-Focused Maps?

Data-focused maps are useful tools when analyzing a large amount of data. As evidenced by the series of maps presented above, data-focused maps can take various forms and present a massive data sets in a way that is clear and easy to understand. Web mapping applications allow users a unique experience that can result in high-level data analysis that yields fascinating research opportunity. Data-focused maps, while particularly useful, are only one type of map that we used to present information about Darius Milhaud.

Narrative Maps

An additional type of map revolves around storytelling. Narrative maps’ primary purpose is to guide the user through a curated experience that tells a specific story.

Story Maps

One feature offered by Arc GIS that does this particularly well is the Story Maps feature. This allows geographers to curate a more specific experience regarding consumption of their data. Examples of such a maps can be found in our maps of Milhaud’s tours to the United States. The user is guided chronologically through Milhaud’s footsteps on the map by clicking arrows that move along the path of his tour. Story Maps are useful for creating a highly interactive, yet selective, experience for the map user. Enjoy exploring the maps below following in the footsteps of Milhaud’s first and second US tour.

Story maps are wonderful tools for research presentation, but they aren’t the only kind of narrative maps that are available through the Arc GIS mapping platform. However, there is an additional form of narrative map that we employed while researching Darius Milhaud.

Animated Maps

Animated maps provide the user with the ability to watch the progression of data over time. Below examine two examples of different chronological animated maps. Expand the toggle box to read more about the two animated maps we’ve created and what they can tell us about Darius Milhaud.

So, why narrative maps?

Maps that tell a distinct story are intriguing to all users, from scholars to the general public. Story maps created through Arc GIS allow for an immersive and highly curated user experience, while animated maps can enlighten unique data trends. To see more narrative maps, including those of Milhaud’s travels, visit our Map Gallery. As wonderful and immersive as digital maps are, there is still one more category of maps we would be remiss not to include.

Chronological Faceted Map of Milhaud’s Performances

Static Maps

While digitally interactive maps like those described above are certainly unique and innovative ways of demonstrating data, static maps can have an equally useful purpose. As a colleague of mine said in a blog post “While static maps can be problematic as they only show what the mapmaker wants to show, they are equally useful in concisely and easily establishing an argument” (Eli Baumgartner). Using a static map can make an undeniably clear argument. For example, see our faceted static map comparing the spread of Milhaud’s works over time. One can clearly see an Eastern shift in the performances of his music. When paired with an animated map of the same shift, users see the same argument in different ways. Static maps, however, cannot be made in Arc GIS. To create the map you see here, I first created a digital map in Arc GIS and took screenshots of the finished product, broken down by year. I then formatted the screenshots to ensure high image quality and placed them into a picture collage software, adding captions that specified the year. Creating a faceted map, in particular, can take a great deal of detailed work outside of the common digital mapping software.

What does this map tell us about Darius Milhaud?

Click on the map to expand the view. When examining the chronological spread of Milhaud’s works in such concrete detail, it is apparent that performances of his works move East over time. As I already explained when discussing the chronological animated map we had previously made, this trend presents an interesting argument about Milhaud’s performances in Germany. This static map, however, allows us to present that data as an argument. Our map clearly demonstrates (especially from 1928-1930) that Milhaud’s European performances are highly concentrated in Germany. Using a faceted map to display this information makes it far easier to present our argument clearly. The animated map we had made earlier while presenting the same data, did not present a defined argument in the same way a static map could.

Why Static maps?

Using a faceted map to display this information makes it far easier to present our argument clearly. The animated map we had made earlier while presenting the same data, did not present a defined argument in the same way a static map could.

Navigating Challenges in Mapping

While mapping can be an incredibly useful demonstrative and pedagogical tool, the process isn’t without its challenges. Below, i detail some of the most common issues geographers face when mapping.

Telling Lies

Minutia

Figure 1. The original Arc GIS map.

Challenges in mapping extend beyond the ethical quagmires of how to present the truth. They can be far more pedantic. As Monmonier points out, even the colors used in a map can affect the way viewers consume a map.When picking the colors for the map showing a break down of Milhaud’s performance by genre, for example, we needed to contend with nine colors, each color corresponding to a genre. We needed to ensure that the map remained aesthetically pleasing while simultaneously employing a color scheme that is diverse enough that each point can be identified clearly from afar. Figure 1 and 2 represent the different ways that colors can affect the way a viewer perceives a map. Figure 1 contains the colors that Arc GIS assigns to a map, while Figure 2 represents the way our team decided to present the map. This particular map is an animation of Milhaud’s performances over time. Using a black and white themed map allowed users to focus on the movement of the performances over time as a whole, rather than individual performances appearing in a bright red color, distracting from the pattern the data presents.

Figure 2. Arc GIS map re-colored.

Similarly, while mapping, we often found small mistakes in the Latitude and Longitude coordinates we’d assigned to a location. We learned that it was imperative to inspect each point closely, a process that would often devolve into our scouring of coastlines to make sure that points that were intended to land in Los Angeles didn’t end up in the Pacific. On occasion, errors were easy to fix. In our map of all of Milhaud’s performances, a performance that occurred in Denver populated on our map in the middle of Kazakhstan. Luckily, someone had simply forgotten to assign a negative sign to the longitude.

Software

Navigating the treacherous waters of Arc GIS can prove to be a challenge for even the most astute Digital Humanists. Often times, small details make can directly influence the display of a map. For example, when one of my colleagues was working on a map comparing the travels of Milhaud to his performances chronologically. The map is incredibly important to our project, however, when we tried to share the map publicly, only the travels appeared. After spending a day agonizing over what the problem may be, we discovered that it was a simple matter of forgetting to publish the layer of performances. It was a two-minute fix. Map making software is full of these detailed dangers, and they can be mind-numbing to overcome.

So, why maps?

All in all, though there are many challenges to the map making processes, the potential it promises are far outweigh any inconvenience. The ability gained to view data in a new and creative way, and the concept of a new visualization of musical ideas is far too enticing to ever imagine giving up on the idea altogether.  In fact, mapping offers untapped musicological potential. A potential that the Musical Geography project will continue to explore for years to come. To learn more about the maps we made for this project, visit our Map Gallery.