Exiled But Not Forgotten: Investigating Commemoration of Musicians in Vienna After 1945 Through Visual Analytics Velitchko Filipov1 , Nathalie Soursos2 , Victor Schetinger1 , Susana Zapke2 , Silvia Miksch1 1 TU Wien 2 Music and Arts University of Vienna velitchko.filipov@tuwien.ac.at, n.soursos@extern.muk.ac.at victor.schetinger@tuwien.ac.at, s.zapke@muk.ac.at, miksch@ifs.tuwien.ac.at Abstract How is music functionalized to urban symbolic politics in the City of Vienna after 1945 until today? To address this question, we present a novel visualization for interactive analysis of multiple biographical timelines connected to Austrian music history. Instead of simply comparing the biographies of over 200 composers, musicians, and conductors in a linear fashion, we connect them with the urban processes that constituted the topos ‘Music City Vienna’, such as street-naming, awards, anniversaries, and exhibitions. The proposed approach intertwines people, events, and locations, allowing the interactive analysis of multi-modal data. Our main goal is to explore continuities and discontinuities, to find underlying trends and different narratives. Ultimately, this work is intended to stimulate the discussion on the challenges of performing visual analytics of historical data. Keywords: Timelines, Storytelling, Narrative Visualization, Music History, Urban Symbolic Politics 1 Introduction Vienna is associated with music more than any other In this interdisciplinary project we explore how Vi- European city. The Austrian capital is not only related to sual Analytics (VA) approaches can be used to embed iconic composers, among others the First Viennese School large amounts of historical and musicological data in space of the Classical period (Joseph Haydn, Wolfgang Amadeus and time. VA is defined as the science of analytical rea- Mozart and Ludwig van Beethoven) and in the 20th century soning facilitated by interactive visual interfaces (Thomas to the Second Viennese School (Arnold Schönberg, Alban and Cook, 2006) and provides a means to navigate and Berg, Anton Webern), but it is also well known for the interact with multi-modal data in different granularities Wiener Walzer and the Wiener Operette as well as for the and levels of abstraction. VA solutions can arise organ- Wiener Opernball and the Wiener Philharmoniker. The ically in an environment of interdisciplinary research in Austrian capital has a strong defined and longstanding an effort to both formulate and answer questions about data. identity constructed by means of music. Therefore, Vienna successfully claims since the 19th century to be named In IMMV we have designed and developed multiple Music City of Vienna (Nußbaumer, 2007; Barber-Kersovan interactive visualizations for exploring the data from et al., 2014). Besides the cultural, socio-political, and different perspectives. We have explored how we can economic potential for tourism and city marketing, the map events to a spatio-temporal overview visualization topos Music City was also used as an instrument of political and how we can embed those same events in a different power. Particularly, since 1945 the potential of its inter- context with a specific focus on the relational and temporal nationally well known musicians was recognized by the aspects of the data. Through VA we allow the development city and its cultural management. The connection between of narratives. The festivity becomes the unit of data in the city and its musicians was topographically manifested chronological and geospatial visualizations, allowing the by street-naming, monuments, plaques, memorials, and comparative revision of continuities and discontinuities, celebrated by anniversaries, concerts, exhibitions, and and contextualizing them within the city landscape. awards. In this paper, we explore the timelines of the musi- In the joint FWF-Peek-project “Interactive Music Mapping cians that were exiled from Austria between 1938 and Vienna: Exploring a city. 1945 up to the present day” 1945 and how the City of Vienna chose to recognize and (IMMV) we explore how music acts in the urban context commemorate them after 1945. The main contribution of Vienna as a social identification instrument and how of this paper is to present a novel visualization technique music is functionalized to urban symbolic politics. IMMV that allows the comparison between many different bi- aims to unravel the narrative of the Music City of Vienna ographies, in order to create new potential narratives topos in its entirety, by analyzing the basic types of music and to contextualize the visualization within a process of utilization (city tourism, city politics, image creation) historical research. We relay this process and show the (Zapke, 2017). The concrete subject of our research are the impact of different research goals in the development of events and festivities in the public space of Vienna during our visualization. the Second Republic (1945 up to 2018). Copyright © 2019 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0). 2 Related Work after males and females in the City of Vienna. Through the Related work falls mainly into two categories: the rep- use of geographic visualization and analysis techniques the resentation of temporal linearity, and of urban spatiality. gender disparities in street naming are made obvious to the Visually comparing timelines and time-oriented data is user. a well discussed topic within VA (Aigner et al., 2008). “TimeLineCurator” (Fulda et al., 2016) offers interactive 3 Data authoring of timelines using natural language processing In this section we describe the data model that we employ in to automatically parse, extract, and structure event data our research and visualization, give an overview of the ob- from a corpus of text. The timelines are visualized jects we have along with examples illustrating their struc- horizontally and color is used to denote different tracks ture and connectivity, provide details on how we process (events from different sets or documents), distinguishing the data to arrive to the final dataset that is used, and ex- multiple timelines. Furthermore, glyphs are employed in plain what sources were used to collect and structure the this visualization to depict the differences between events data. that span an interval of time or occur at a specific point in time. Glyphs and other visual variables, such as color, size, and shape, can be used to visually distinguish events with different characteristics or categorical dimensions. In a similar manner “TimeSets” (Nguyen et al., 2016) explores visualizing and comparing multiple timelines by using set relations among the individual events. In this approach events are associated with certain topics and this informa- tion is encoded by the use of colors. Gradient backgrounds that transition from one color to another are used to depict events that are shared between two topics. In the scope Figure 1: IMMV Data Model - Object types and rela- of our subject the focus lies in exploring and visualizing tionships. Dashed lines indicate intra-object relationships, the temporal facet of our data. In “Timelines Revisited” whereas the thick lines are inter-object relationships. (Brehmer et al., 2017), the authors review related work and literature, analyzing over 200 timeline visualizations, and propose a design framework based on three main 3.1 Data Model dimensions: representation, scale, and layout. This work represents the state-of-the-art in timeline visualization For the purpose of the Interactive Music Mapping Vienna and the design framework can be applied to biographical (IMMV) project, we have constructed a data model to and prosopographical data visualization. Representation describe large amounts of multifaceted information in relates to the general shape or disposition of an individual space and time, which are depicted in Figure 1. timeline, such as linear, radial, grid, or spiral. The scale deals with the disposition of information over the timeline, Our data model is comprised of six types of objects: and can be chronological, relative, logarithmic, among events, historical events, people/organizations, locations, others. Finally, the layout is the composition schema of the sources, and themes. visualization, how different timelines are integrated in it. Events or festivities play a central ontological role in our The second category, the spatio-urban dimension, project. They comprise our main data object and relate all was in similar projects mainly limited to the geolocation other objects. For this reason it is the only object type of streets and places. Street-names were analysed in that can be connected to objects of other different types. two recent interdisciplinary projects combining historical Other entities have to be related to each other through an research (historical, cultural and political dynamics of event. This approach simplifies the topology of our seman- street-naming) with VA. In the first project a commission tic network, and enforces scope. Events are considered to of historians (Autengruber et al., 2014)(2011-2013) be public happenings in the City of Vienna, such as: fes- examined, due to their connection to national socialism, tivals, celebrations, awards, and open-air concerts. Events historically critical personas that had streets named after are defined by a start and end date, a name, a description, a them in the City of Vienna. The project resulted in the location, and connected to people that have contributed to geolocation of these streets in an online map 1 , in a pub- it, other events, historic events, themes and sources. lication as well as in explanatory boards next to the street names. The second project, GenderATlas (Riegler et al., Historic Events are events of historical importance that are 2015)(2013-2015) is a platform for the visualization 2 and not directly related to music, but provide context, such as provision of information regarding the gender distribution the end of World War II or the period of the Second Re- in various fields in Austria. In a specific example, the public. Historic events are defined by a start and end date, authors examine the spatial distribution of streets named a description, a list of references, relations to other historic events and a list of important dates that have occurred dur- 1 https://www.geschichtewiki.wien.gv.at ing the timespan of the historic event as well as a list of 2 https://www.genderatlas.at/articles/strassennamen.html references and further literature. Themes serve as descriptors for events and characterize award ceremony). As a second step we proceeded to filter them based on the topics they are associated with, such people based on their roles, since we are only interested as “Musikalisierung der Topographie”, “Wienerlied”, or in composers, musicians, and conductors. Finally, we “Austropop”. They have a similar function to keywords and separate those musicians, composers, and conductors that provide meta-information to augment our semantic topol- have been exiled, in order to compare them later with the ogy. Themes are associated with a name, a list of refer- rest of the people in the dataset. ences, a set of categories (theme types), and relationships to other themes. We further processed the dataset and applied a cat- egorization of the events. The events can belong to Sources are the different primary and secondary sources one of the following five categories: street-namings, that are gathered about an event. They are an encapsu- anniversaries, exhibitions, awards/prizes, and memorials. lation of multimedia data, including visual sources (pho- tographs, posters, videos), audio sources (recordings, mu- 3.3 Historical Sources sical pieces), and textual sources (press articles, publica- The historical sources used for this paper were either tions). Sources are comprised of a name, a list of refer- related to the biographies of the musicians or to the ences, along with media attachments, provenance data and events. Regarding the biographies of the musicians, the copyright information, a description, and details about the quantitative collection of names and biographical data was creator(s), contributor(s), or publisher(s) along with rela- based on existing databases (Lexikon verfolgter Musiker tionships to other sources. und Musikerinnen der NS-Zeit, University of Hamburg), secondary literature (mainly the publications of the Doc- Locations describe particular places in the City of Vienna umentation Centre of Austrian Resistance, DÖW), and that are of musicological and historical relevance. Loca- online bibliographical lexica (Wikipedia, Wien Geschichte tions are defined by their current name along with a list Wiki, Österreichisches Musiklexikon). We are aware that of alternative names (reflecting the change over the course such sources often provide little more than a snapshot of a of history), location types (e.g., park, monument, official person’s life. Therefore, we refer to further relevant bibli- building, etc.), a list of references, geospatial data, and re- ography containing secondary sources, (auto-)biographies, lationships to other locations. or published correspondence. People/Organizations constitute the set of people that have Regarding the events, each type has different sources had a particular impact on the music history of the City of of information that were used. Street-names and awards Vienna. We differentiate between people and organizations were announced in the official gazette (Amtsblatt der but use an abstract model for both types. People and or- Stadt Wien) as well as in Wien Geschichte Wiki. Street ganizations are comprised of the following properties: a (re)namings during 1938-1945 and after 1945 were pub- name along with a list of alternative names, a list of ref- lished and visualized in a map by the aforementioned erences, a description, and relationships to other people or commission of historians (Autengruber et al., 2014). organizations. The main difference between the two is that Monuments and plaques regarding musicians are listed in people have roles and functions (e.g., composer, conduc- several publications (Nelson, 2006; Kretschmer, 1992) as tor, director, musician, etc.) associated with them, whereas well as in a more critical approach by the (Exenberger organizations have a set of organization types (e.g., NGO, and des österreichischen Widerstandes, 1998). The main political party, cultural organisation, music band, etc.). source of information for exhibitions were catalogues, press articles, and posters. Speeches by politicians at the 3.2 Data Processing inauguration festivities were also taken into consideration In the scope of this paper we chose to work with a subset as well as photographs picturing these events. In IMMV of data, specifically focused on musicians exiled between sources gathered about the events were collected, selected, 1938 and 1945 in order to find out which musicians the and developed with the help of our cooperation partners City of Vienna chose to recognize and commemorate after and archives (specifically, the Vienna City Library and the World War II. In addition we compare the timelines of Austrian National Library). the exiled musicians with over 200 musicians who were honored in the same time period by the City of Vienna. 4 Research use case: Exiled Musicians As a first step our selection criteria was to find all people We used the prototype to explore the narrative of musicians related to the theme “Musikalisierung der Topographie” and exiles after 1945 and to examine its effectiveness in (translated from German: musicalization of the topogra- conveying patterns and trends to our domain experts. phy). In our data model people are not directly related to the aforementioned themes, but via the event that honors After the annexation of Austria into Nazi Germany a musician an association is established. For this purpose on 12 March 1938 approximately 130,000 people fled we aggregate all people that are related to events belonging into exile (Neugebauer and Ganglmair, 2003). Among to the aforementioned theme, which initially resulted in them were expelled musicians, Jewish and non-Jewish, many people that were not of relevance to our subject of composers of “degenerate music”, but also composers of research (e.g., curators of an exhibition, vocalists singing “Wienerlieder” or “Operettas”, including icons as Arnold at an exhibition opening, politicians participating in an Schönberg, Alexander von Zemlinsky, Ralph Benatzky, Ernst Krenek, Robert Stolz, Hermann Leopoldi, Walter (Ehrenbürger and Bürger der Stadt Wien). Jurmann, and many more. Most of the musicians left Austria after March 1938 and went to the United States, Anniversaries or commemoration days, especially after South America, France, Great Britain, or Switzerland. one or two centenaries of the musicians birth or death, were celebrated in Vienna from the 1920s until today. After 1945 Until 1959 only approximately 8,000 returned. Those the years of Johann Strauß Jr. (1975), Wolfgang Amadeus who decided not to come back to Austria had manifold Mozart (1991, 2006), Joseph Haydn (2009), and Gustav reasons: the injustice and humiliation they suffered in Mahler (2010) were celebrated with extensive effort, in- their homeland, the general situation in post-war Austria cluding exhibitions, concerts, and conferences. with its half-hearted denazification, the anti-semitism Exhibitions are a form of presenting a musician or a group still practiced, the unresolved property issues (Knight, of musicians to the public by using visual sources, hand- 2000), but also personal reasons as the integration in a new written notes, or other objects connected to the exhibitions’ country. It is important to note that there have been no topic. Most of the music related exhibitions took place ei- official invitations expressed from the Austrian provisional ther in several Viennese museums, libraries and archives or government for the exiled to return. An exception was the in the apartments and houses of Beethoven, Strauß, Mozart, Viennese city councilor in charge of cultural affairs from Haydn, and Schubert administered by the Wien Museum. 1945 to 1949, Viktor Matejka, who sent letters appealing We gathered 110 exhibitions directly connected to Viennese artists and scientists to return (Matejka, 1983). musicians. The exhibitions Die Vertreibung des Geistigen aus Österreich(1982) at the University of Applied Arts and The gap in the cultural life and institutions of Vienna the series Musik des Aufbruchs (2004-2009) by the Jewish (orchestras, music academies, universities, etc.) left by the Museum directly refer to the topic of exiled musicians. The exiled musicians, composers, and conductors could never series Musik des Aufbruchs presented the life and work of be filled again, it is thus essential for the understanding of Hans Gál, Egon Wellez, Franz Schrecker, Eric Zeiss, Erich the Music City Vienna after 1945. But how to visualize this Wolfgang and Julius Korngold and Hanns Eisler to the Vi- “brain drain” or gap of musicians, who did not return to ennese public. Vienna after 1945? With the existing IMMV restriction on events, we were unable to make those visible, who were not 4.1 The Anatomy of Our Visualization present. Both the departure and the re-migration, were very Our visualization consists of different elements, as depicted personal for the exiled musician, seldom recognized by in Figure 2. On the left side (A), a short menu allows the public and not officially celebrated. As a consequence, filtering, searching for a specific person, and changing the we decided to treat street-namings, awards, anniversaries, ordering and grouping criteria of the data. On the right side and exhibitions as particular events. This allowed us to (B), the list of all people can be seen, with the count of their integrate the exiled musicians that were referenced by such associated events. This list is updated when a temporal events (even if they were not initially present), into our selection is made and reflects only those people who have database and visualization. been active in the time span selected. It can also be used to show additional information about the selected individual, Street-naming is a particularly powerful and “time- such as dates of birth, death, exile, and events. The core less” instrument for perpetuating public memory. The component, however, is the radial timeline (C) where the monopoly on the toponyms of a city is claimed by the gov- multiple timelines and associated events are displayed. The erning authorities. Therefore, street-naming is connected prototype also includes a timeline at the bottom (G), which to the ruling regime, which is using streets to inscribe its visualizes the events as a dotplot that are color-coded ideologies and to assert symbolic power and ideological according to their category and allows the user(s) to hegemony (Rose-Redwood et al., 2018). Especially, select and interact with individual items. We multiple the renamings during the Dollfuß-Schuschnigg-regime design iterations with different component configurations, (Austrofascism, 1934-1938) and during National Socialism including a line-chart, and came to the realization that for (1938-1945) are pivotal for the understanding of the the tasks of displaying, counting, and interacting with the symbolic relevance of toponymy (Autengruber et al., individual events in the timeline the dotplot performed 2014). Since 1945 the selection criteria of the Viennese better as a solution. The prototype is made available to the city council (Stadtrat) requires important personalities, general public and research communities 3 whose “greatness” is generally recognized in association with the location and importance of the named street and In the visualization time radiates outwards from the the naming has to be at least one year after the person’s center, with the innermost circle being the birth of Vivaldi death (Interkalarfrist). at 1678, and the outermost being the present date. The date of 1945 is highlighted (E) due to its importance in the Awards are an official act of how the city and its mayor scope of our subject, providing a key temporal reference are honoring a person for her/his outstanding achievements. for the visualization. A time period is mapped to a ring The most important and prestigious awards for musicians in this mode of viewing, and can be selected through in Vienna were: awards, rings and medals (Ehrenring, clicking and dragging for filtering a specific time range. In Ehrenmedaille, Ehrenzeichen), the prize of the City of Vi- 3 enna (Preis der Stadt Wien), and the honorary citizenship https://immv-app.cvast.tuwien.ac.at/biographical Figure 2: Overview of the visualization dashboard. A) Filtering controls. B) List of musicians being displayed. C) Main radial display component. D) Life timelines of musicians. E) Circle indicating 1945. F) Categorical data. G) Timeline visualizing post-death events Figure 2, a random period of 74 years (March 11th 1891 The focus of the visualization itself isn’t on the achieve- - September 11th 1965) is selected, showing a light blue ments of a person within their lifespan, but rather on their highlight and graying out those peoples lifelines and events superimposition with posthumous events‘ that happened that do not fall into the selected time span. in the City of Vienna. For this purpose we chose to classify the events based on the following categories: Every line radiating from the circle is the timeline of street-namings, memorials, awards/prizes, anniversaries, a person, with their life span highlighted in green with and exhibitions. Due to the importance of these events circles indicating birth and death. People who have been we chose to highlight them with a different colors as to exiled also feature an extra dark green timeline that is contrast them to the person’s life and draw attention to superimposed over their life lines. This extra timeline the patterns that they create when using certain ordering indicates the time of exile, from the point of their departure strategies, such as the date of homage (see Figure 4 C, D, from Austria to their return. It is important to note that E). estimating the exact duration of exile is difficult, as we seldom know the exact date, when the border was On the outer rim of the circle we encode the categor- crossed. Most exiled people left Austria after the 12th of ical information about each person (see Figure 2 F). In March 1938, some even during the authoritarian regime this specific use case we encode the profession (role) of of Dollfuß-Schuschnigg. But when did the exile end? the person. The prototype also allows the users to choose Was it on the 27th of April 1945, when the provisional a different categorization and grouping of the dataset, government announced that the exiled were legally allowed including exiled musicians, those that were born after to leave the host country (e.g., in the U. S. after spring 1945, and those that died before 1938 as these two dates 1946)? What if the exiled did return just for a short period play an important role in our subject of research, with of time? Such is the case of Erich Wolfgang Korngold, 1938 marking the starting point of exile and 1945 denoting who came back in 1949 for two years and left again the end of World War II. This element can be used to disappointed. In order to make the musicians decision not detect patterns in any sort of categorical attribute of the to return visible, we decided to visualize as ’exile’ the selected musicians, according to the grouping criteria period from crossing the Austrian border until the person’s used. Our color encoding reflects our efforts in balancing return. If a musician died without returning to Austria this the comparison between timelines, events, and displaying period after 1945 was also marked as exile leading up to categories. their death. At the bottom (see Figure 2 D) we have augmented the prototype with a timeline that contains only the events and visualized compared to other layouts. that are belonging to one of the five categories described in Section 3. The advantage of this timeline is that it easily One of the questions that this visualization was de- shows temporal trends, such as in which decade the most signed to answer is: which musicians came back after the street-namings or exhibitions happened. Furthermore, the exiled were officially allowed to return to Vienna? For timeline displays each event as a circle that is color-coded this reason we decided to represent time differently than depending on its category as to remain consistent with the the aforementioned approaches. As we are interested in rest of the visualization. The circles allow the user to inter- having a visualization where it is more obvious to notice act with individual events as well as count events belonging certain trends and patterns occurring at different decades to the same category in a given time interval in a more throughout Vienna’s history (with a special focus on the straight-forward fashion compared to other visualization years after 1945). techniques. Both the radial and timeline visualizations are linked, meaning that mouseovering an event in the timeline Figure 4 show three different ways to order the time- will highlight the corresponding person’s lifeline and event lines of people, and how they affect the visual subject. In in the radial part of the visualization and vice versa. each case, different trends stand out. In Figure 4A the biographical timelines are ordered by the birth date of the During our discussions with the domain experts (his- person and grouped by their role (composer, conductor, torians and musicologists) we realized the need for analytic musician), which is the first event on its timeline, marked tools to interact with the data in detail. To support them as a green circle. A clear spiral forms growing outward, we have added filtering features such as visualizing only which is complementary to that of Figure 4B, where exiled musicians (see Figure 3), and the selection of time timelines are ordered by death date and grouped by role. rings, which also provide a list of the people that have been In this case, the spiral signifies a ceiling, the limit of one’s active during this period sorted by the amount of events life. It is much easier to identify musicians that are alive they are related to. Furthermore, upon selecting any person in the second case, as all timelines with no death date will the visualization reorients itself and highlights the selected form a contiguous arc. person displaying her/his details in a side panel. This allows the users to also explore a person’s information and It is important to note that the data and individual links to external resources, as well as, the details about the timelines remain the same in all cases. By simply re- events of her/his homage by the City of Vienna. ordering and grouping them according to different criteria it is possible to obtain unique patterns that can tell us The visualization is developed to provide the domain something about the underlying behavior of the data. This experts with a tool to interact with the data in detail but is used in Figures 4C, D, and E to explore the subject of also gain an overview and realize patterns or trends in when musicians are recognized and commemorated by the how the City of Vienna recognized and commemorated City of Vienna. Each person is ordered according to the it’s musicians. The intent of the visualization is to also date of homage given by the city, resulting in a spiral, that provide casual users that are not acquainted with the can be clearly seen, formed by the events in a contrasting musical, political, and cultural history of Vienna with a color. tool to explore and investigate. The goal is to provide both the domain experts and casual users with an interactive 5 Discussion and Future Work exploration environment to unravel the narrative of the By exploring the data using the proposed overview visual- Music City of Vienna. izations certain interests and questions arise, leading to the main question of how the City of Vienna chose to remem- 4.2 Comparing Timelines ber it’s musicians by embedding them in the city’s topogra- Considering all possible combinations of representation phy and collective memory (Erll, 2005). We explored how proposed in “Timelines Revisited” (Brehmer et al., 2017), we can visualize this information and generate patterns that including scale, and layout, there are 5 ⇥ 5 ⇥ 4 = 100 might provide answers to the following questions related to different design possibilities. The proper design choice chronology, the person’s biography and the type of event: depends on the data, tasks, and users of the visualization. Is there a caesura in the visualization, which can be help Our work, however, cannot properly characterized in this us recognize patterns in the interconnection between events framework. Our main goal is to highlight events and remembering exiled musicians and political caesuras (e. g., patterns occurring after the deaths of the selected people, the announcement of the provisional government in 1945, while simultaneously allowing the user to see trends within the end of Allied-occupation in 1955, the fall of the Berlin their lives as within their posthumous commemoration. Wall in 1991), political and cultural turning points (e. g., Furthermore, most of the proposed approaches do come the causa Waldheim in the late 1980s or changes of gov- with significant drawbacks when considering the number ernment)? Is there any pattern or trend of the honored per- of timelines that they can simultaneously represent. This son identifiable relating to her/his popularity during life- is also the design rationale for our radial layout, because time and after death, any link to a political party or cul- of the amount of people that we are visualizing the radial tural institution, or the music style the person is associated layout generally has a better aspect ratio, area efficiency with? Was there any difference in chronology and quantity when drawing, and allows for more timelines to be encoded of events between those musicians, who returned from ex- Figure 3: Filtered visualization containing only exiled musicians. ile and those who remained in the host countries? Which of research, where both new information and new means to category of honoring events were preferred by the City of visualize it brings us closer to developing our questions in Vienna and have there been any changes during the time a nonlinear fashion. span analyzed? These questions opened up some interest- ing points and directions for future work that we would like In the future of IMMV the presented visualization to explore, including: will be connected with a map of the City of Vienna. Thereby the spatial dimension of the data, the distribution • Exploring the encoding of information in the outer of streets and events between the center and periphery ring, which provides an additional dimension for the of the city can be made immediately visible to the user. emergence of patterns; Zooming in place and time will enable the user to analyze • Visualizing group similarities through clustering, in a certain time period or location more detailed than the addition to ordering strategies; rest. The users have the possibility to create their own story depending on their research question. The possibility of • Inferring categorical dimensions such as political ori- comparing people by selecting multiple timelines and high- entation using information from organization affilia- lighting differences or commonalities is another direction tion, and involvement in particular events; for future work. Research questions regarding differences in periods of government could be answered by inspecting • Geographically mapping and visualizing the spatial and comparing multiple timelines, and exploring different facet of the events and implementing metrics that ways of encoding event data (moving from color coding to would take this information into account. glyphs or shapes representing the category of the events). Furthermore, our database is also in a continuous process of expansion and refinement. As new sources come to 6 Conclusion light, they are added to the database and are automatically We introduce a radial visualization supporting our domain included in the visualization. This entails a cyclic process experts in investigating the biographical data of musicians Figure 4: Six different orderings of the timelines around the central axis using different combinations of the grouping and ordering criteria. A) Ordering by birth date and grouping by role; B) Ordering by death date and grouping by role; C) Ordering by date of the first post-death event and grouping by role; D) Ordering by date of first post-death event and grouping by exile status; E) Ordering by date of first post-death event and grouping by people who died before 1938; F) Ordering by death date and grouping by exile status. In each case, a different spiral pattern forms, which allows for the detection of trends and outliers. exiled from the Vienna along means to find out how and continuities or discontinuities before and after the periods when they were honored or commemorated by the City of exile. of Vienna in terms of street-naming, exhibitions, awards, and anniversaries. In the interdisciplinary collaboration A central statement of this project is the relativity of of the IMMV-team the visualization helped to validate history and the plural forms of understanding and inter- and verify the correctness of the data in the database. preting historical facts. We show in this paper how VA Those exiled musicians, who neither received an award, approaches can support experts in the domain of histor- nor had a street named after them fall through the cracks. ical and musicological research to gain a chronological Furthermore, the importance of a musician or the quality overview and to combine music with political history and of their music cannot be implied from the fact that the the City of Vienna with its musicians. Historical research City of Vienna chose to honor them or not. Other ele- helped to define the events, people and to collect the infor- ments such as the concerts that happened in the Vienna mation that was needed to perform research on the subject. Opera, the Musikverein or the Konzerthaus, as well as The lack of potential interactions had to be tracked in ad- record sales figures or their popularity, were not part vance as spatial and temporal data extracted from texts can of our research. Furthermore, the research case exiled be ambiguous or uncertain. By using visualization as a tool musicians, enabled us to concentrate on a small selec- for data presentation we can recognize potential patterns, tion of persons, to see patterns, prove our hypotheses, and overlaps, and trends and offer new methods of research. open new research questions and directions for future work. Furthermore, the interactive elements of the visualization help the user to open additional information and sources Focusing on temporally anchored events, will allow about the selected event and person or about neighbouring us to extract salient events or periods of time which can events and people. This offers on the one hand a selective, be further used to identify other elements in a person’s subjective reading and interpretation, which can be on the timeline. Further academic research also has to find traces other hand combined with combinatorial and interactive of musicians in Vienna before 1938 in order to recognize reading, offering new ways for interpreting historical data. The multimediality and complexity of historical data Anfang? Oder: Chancen eines Anfangs, pages 39–47. with various sources and media connected to events and Salzburg/Wien. biographies can be diversified and concretized via VA. D. L. Nelson. 2006. Vienna for the music lover. The com- This opens up new elements of narration, substituting the plete guide to Vienna’s musical sites and performances existing methods by a new and unique visual narrative. today. Vienna. W. Neugebauer and S. Ganglmair, 2003. Dokumenta- By remembering the exiled musicians via several tionsarchiv des österreichischen Widerstandes (eds): commemorative events, the City of Vienna in a way Jahrbuch, chapter Remigration, pages 96–102. Wien. re-emigrated those who left Vienna after March 1938. P. H. Nguyen, K. Xu, R. Walker, and B. W. Wong. 2016. Today they are again a part of the city’s memory. Of Timesets: Timeline visualization with set relations. In- course ambiguities and ambivalences of each fortune and formation Visualization, 15(3):253–269. the complexity of each biography cannot be simplified M. Nußbaumer. 2007. Musikstadt Wien: Die Konstruktion via visualization. Their collective displacement thus eines Images. Wien. bequeathed a gap, which we tried to fill at least visually. M. Riegler, M. Wenk, E. Aufhauser, F. Ledermann, M. Schmidt, and G. Gartner. 2015. Genderatlas österre- Acknowledgements ich - Entwicklung eines zielgruppenorientierten online- This work was conducted within the framework of the tools. Mitteilungen der Österreichischen Geographis- project "Interactive Music Mapping Vienna" (AR384-G24) chen Gesellschaft, 157:323–339. funded by the Austrian Science Fund (FWF) and lead by R. Rose-Redwood, D. Alderman, and M. Azaryah. 2018. Susana Zapke (Music and Art, University of the City of Vi- The Political Life of Urban Streetscapes. Naming, Poli- enna). We also thank our cooperation partner Vienna City tics, and Place. London and New York. Library for providing us important sources. J. J. Thomas and K. A. Cook. 2006. A visual analytics agenda. IEEE Comput. Graph. Appl., 26(1):10–13. 7 References S. Zapke, 2017. Y. Franz and C. Hintermann (eds.): Unrav- W. Aigner, S. Miksch, W. Müller, H. Schumann, and elling Complexities. Understanding Public Space, chap- C. Tominski. 2008. Visual methods for analyzing time- ter Mapping the Music City in the 20th century. Theoret- oriented data. IEEE Transactions on Visualization and ical Approaches, pages 71–76. Wien. Computer Graphics, 14(1):47–60. P. Autengruber, B. Nemec, O. Rathkolb, and F. Wenninger. 2014. Umstrittene Wiener Straßennamen. Ein kritisches Lesebuch. Wien. A. Barber-Kersovan, V. Kirchberg, and R. Kuchar, 2014. A. Barber-Kersovan, V. Kirchberg and R. Kuchar (eds.): Music City. Musikalische Annäherungen an die „kreative Stadt", chapter Music City. Musikalische Annäherungen an die kreative Stadt, pages 9–30. Bielefeld. M. Brehmer, B. Lee, B. Bach, N. H. Riche, and T. Munzner. 2017. Timelines revisited: A design space and consid- erations for expressive storytelling. IEEE Transactions on Visualization and Computer Graphics, 23(9):2151– 2164. A. Erll. 2005. Kollektives Gedächtnis und Erinnerungskul- turen. Eine Einführung. Metzler, Stuttgart. H. Exenberger and Dokumentationsarchiv des österreichis- chen Widerstandes. 1998. Gedenkstätten zu Widerstand und Verfolgung, Exil, Befreiung. Eine Dokumentation. Gedenken und Mahnen in Wien 1934-1945. Wien. J. Fulda, M. Brehmer, and T. Munzner. 2016. Timelinecu- rator: Interactive authoring of visual timelines from un- structured text. IEEE Transactions on Visualization and Computer Graphics, 22:300–309. R. Knight. 2000. "Ich bin dafür, die Sache in die Länge zu ziehen": Die Wortprotokolle der österreichischen Bun- desregierung von 1945 bis 1952 über die Entschädigung der Juden. Wien. H. Kretschmer. 1992. Wiener Musikergedenkstätten. Wien. V. Matejka, 1983. J. Jung (eds.): Vom Reich zu Österre- ich. Kriegsende und Nachkriegszeit in Österreich erin- nert von Augen- und Ohrenzeugen, chapter War 1945 ein