3. From Parodies to Pages: The Many Faces of (Detecting) Text Reuse

Jutta Ernst and Oliver Scheiding argue that scholarship of newspapers, magazines, and other publications

requires a conceptualization of periodicals not so much as objects of print culture, but rather as participating in relational and dynamic webs—both on a local and translocal level—of socioeconomic conditions, legal and ideological frameworks, institutional organizations, action networks, and communicative environments. (1)

Such a reconceptualization is shown in the current interest in textual exchange between publications. In the nineteenth century, reprinting a text was a common practice to demonstrate that editors were actively involved in the news network system. In some cases, this was explicitly mentioned in the texts to emphasize the authenticity and significance of the information. In other instances, it was omitted, leaving it up to the scholar to identify the source(s). Editors of German-American newspapers in Minnesota, Ohio, or Pennsylvania also made use of this exchange practice to quickly and diversely fill their publications with material from the New World and their old homeland. Most of the content found in newspapers consisted of reprinted material. Reprinting as a method allowed smaller newspapers in rural areas of the U.S. to present their readers with news of political, economic, or criminal nature from the big cities, and vice versa. They used a variety of other German-American newspapers through subscriptions or relied on American papers that were specifically translated for a German-speaking audience. Unlike editors, readers did not have subscriptions to several newspapers. Through the exchange system, readers in Minnesota, Ohio, or Pennsylvania, however, were reading the same material, which was sometimes modified and adapted to the local environment, but very often it was left intact depending on the nature of a text. Legally, this exchange was accepted. Newspapers and magazines were carved out from intellectual property laws due to diverse pragmatic reasons. First, the objective of the American newspaper system was to spread news across the entire territory of the new nation to link people and places from the East to the West. Second, there was a long-standing agreement that the abundance of material made it practically impossible to conduct large-scale tracking of reprints, especially regarding intensive textual edits.

In this system, all kinds of texts, short and long, were published, circulated, reprinted, or rewritten. Nevertheless, the genre of a text had a tremendous impact on copyright legislations and enforcements that had to be developed and hammered out in the nineteenth century on a national and international level. Copyright negotiations, particularly concerning book publications of fictional material, became numerous and people repeatedly debated over how to shape and change the legal situation for the national and international publication landscape. On May 4, 1892, Der Deutsche Correspondent was among the first German-American newspapers in the U.S. to announce on its front page that the German Empire and the U.S. would enact a new transatlantic contract on copyright:

Die Bürger der Vereinigten Staaten sollen im deutschen Reich den Schutz des Urheberrechts bezüglich der Werke der Literatur und Kunst, sowie den Schutz der Photographie’n gegen unbefugte Nachbildung auf derselbe Grundlage genießen, wie solcher den Reichsangehörigen gesetzlich zu.^[1]

The article titled “Der Verlagsrecht-Vertrag mit Deutschland” (“The Publication Law Contract with Germany”), which was preceded by local news from Washington and followed by international news about China (“The New Anti-Chinese Bill”) and Belgium (“Belgium Dynamiters: Another Explosion”), and an ad posted by the German Healing Institute in New York that promoted a book about regaining virility and “Geschlechtskrankheiten” (“Men and Youth”), stated that the objective of the contract was to protect the citizens of the United States and their copyright of literary works in Germany.^[2] The contract was a response to the activities of publishers in Europe, who increasingly translated and published American best-sellers by Mark Twain or Henry James in many cases without the consent or knowledge of their literary creators prior to this enactment (Stokes, “Copyrighting American history” 292). Many people, on both sides of the Atlantic were eager to consume reading material and the market for fiction started to boom accordingly (Birkle 51). The mass production of newspapers, magazines, books, and other periodicals went hand in hand with increasing literacy rates, enabling more people to read and write. At the same time, newspaper publishers were experimenting with different ways of copying material from other sources that fell under other copyright legislations. It took until the turn of the century to establish far-reaching copyright enforcements as announced in Der Deutsche Correspondent in 1892. “While supporters of the international copyright chiefly sought to bring order to the transatlantic trade,” as McGill writes, “those who defended the system of reprinting sought to protect emergent mass culture that circulated in uncopyrighted newspapers, zines, and pamphlets as well as in books” (“Copyright and Intellectual Property” 407).

Seen from a national perspective, the United States has had copyright legislations since its founding, but the specifics of copyright law have evolved over time. Throughout the century, the growing mobility of texts and new technical advances in print, in other words new opportunities to reuse texts, led to new discussions on ownership of people and their ideas (see Stokes, “Copyrighting American history” 293). These developments brought in their wake discussions on how to define intellectual property that resulted in several U.S. Supreme Court cases. In Wheaton v. Peters, the Supreme Court decided its first ruling on copyright in 1834. The Court held that the author of an unpublished work could control the first publication but did not have the same control over reproduction after the initial publication (see Baker).^[3] Next to reproductions of the same text, there was no solid legal basis in the context of rewritings including translations, which was not only beneficial to German editors in Europe, but also to German migrant editors in the U.S. They did not only reuse American material by translating news such as the report on “The New Anti-Chinese Bill” but also texts that fell under the category of the feuilleton. There was one notable court case that brought national attention to the reprinting and translation work of German-American editors decades before the international copyright legislation in 1892. In 1853, Harriet Beecher Stowe sued F.W. Thomas, publisher of the Democratic German-language newspaper the Philadelphia Freie Presse (1848–1887).^[4] Thomas had published a German translation of Stowe’s Uncle Tom’s; or, Life Among the Lowly (1852). In Onkel Tom’s Hütte, oder: Leben unter den Verstossnen. Nach dem Englischen der Mrs. Harriet Beecher Stowe, frei bearbeitet von Adolf Strodtmann (1853), the German-American editor listed her as the author but sold it without her permission.^[5] Stowe brought the case before the court. The Judge was Robert Grier of the United States Supreme Court in his capacity as a Circuit Court judge. According to Melissa Homestead, he was an outspoken anti-abolitionist (201). He proclaimed that Stowe could not prevent Thomas from publishing a translation of her work. According to the Court, the author had given up her absolute control and ownership of the characters she created, and anyone could reuse them through imitation or adaptation:

Uncle Tom and Topsy are as much publici juris [public property] as Don Quixote and Sancho Panza. All her conceptions and inventions may be used and abused by imitators, playwrights, and poetasters. They are no longer her own—those who have purchased her book, may clothe them in English doggerel, in German or Chinese prose. Her absolute dominion and property in the creations of her genius and imagination have been voluntarily relinquished. All that now remains is the copyright of her book; the exclusive right to print, reprint and vend it, and those only can be called infringers of her rights, or pirates of her property, who are guilty of printing, publishing, importing or vending without her license “copies of her book.” A translation may, in loose phraseology, be called a transcript or copy of her thoughts or conceptions, but in no correct sense can it be called a copy of her book. (qtd. in Homestead 201)

As this passage illustrates, Uncle Tom’s Cabin, a fictional testimony to the horrors of slavery and a textual embodiment of human rights activism, could no longer be called Stowe’s own once it had been rewritten. Grier stated that translations were new and merely copies of her ideas but not copies of her book. The passage does not only highlight the loss of exclusive control over intellectual property in the case of rewriting a novel but simultaneously displays derogatory remarks about literary taste. One can interpret the equation of “English doggerel,” which was not considered high-quality poetry as it had an unusual rhyme and was said to lack artistic merit, and “German and Chinese prose,” as an indirect expression of anti-migrant sentiments. Looking at the historical development of copyright laws in the United States, Siva Vaidhyanathan argues that they were not just influenced by economic interests but also reflected cultural values regarding race, class, ownership, free speech, and democracy (48). Grier’s ruling and the line of argumentation he chose reflected his opinions on Nativism and human rights. Four years after this case, Grier was still on the Supreme Court during the landmark case Scott v. Sandford, in which the majority opinion stated that enslaved people were not citizens of the United States.^[6] This case and its outcome advanced the U.S. a step closer to the Civil War. In addition to revealing the cultural values of the judge, the Stowe v. Thomas case illustrates the social life of a text once it had left the confines of its author’s hands.

The case also reflects the contemporary multilingual print landscape in the United States in the middle of the nineteenth century. German migrant papers created their own world of “text reuse” based on its special feature: the German language. By printing literary works in another language, migrant newspapers specifically benefited from insufficient copyright legislations and enforcements as translations did not count as reprints. This act of translation mirrored the migrant newspaper’s two-fold mission that Kamphoefner describes as an “ambivalent, Janus-faced role”: “One the one hand, it looked backward, serving to preserve the mother tongue” while educating “immigrants in American ways” (Germans in America 107). The translation of a novel such as Uncle Tom’s Cabin displays this double role, shedding light on the role of fiction, specifically illustrating that this genre was not only about entertainment but likewise functioned on the level of information by educating migrants on specific American topics such as race, democracy, citizenship, and freedom. Editors legally reused these works and profited from these processes without having to compensate the “original” authors. The network of transregional exchange contributed to the popularity of these texts among different German-American communities.

American bestsellers were not only rewritten by German-American editors in publication hubs such as Philadelphia, but also in rural areas of the Trans-Mississippi West. Thomas was not the only one to provide fresh translations of Uncle Tom’s Cabin. In 1853, even the small-town paper Hermanner Wochenblatt (1845–1854), the first German newspaper west of the Mississippi, printed its own translation of Onkel Toms Hütte in 26 issues.^[7] Hermann, situated on the Missouri River eighty miles from St. Louis and the Mississippi River, was a smaller community in Gasconade County, but “a center for German intellectuals and radicals” (Wittke 55).^[8] Christine Graf’s husband bought the Wochenblatt one year later because its owner Eduard Mühl had died of cholera. The Graf family renamed it to Hermanner Volksblatt but continued the abolitionist tradition of the paper and advocated that policy despite the dangers it entailed. “The classic accounts of the German-American press attempt to explain its role by examining the manifest political content” asking, “for example, what stand did the editors take on the question of slavery or America’s entry into World War I?” (Peterson, “The Political” 41). With this statement, Peterson, who above all criticizes Wittke’s work when it comes to the neglect of literary works, advocates for “expanding our notion of the political to include the social reproduction of norms and values within the aesthetic realm” (41). Despite its small size, the newspaper had a significant impact. According to Kamphoefner, Gasconade County was the only one in a slave state to have a majority vote for President Abraham Lincoln in 1860 (Germans in America 110). Anti-slavery ideologies were not only shared by the editors of the paper but also by its readers emphasizing the informational value of both editorial content and other forms of communication such as poetry and prose.

Stowe eventually lost her case against Thomas. But she continued to write other fictional and non-fictional texts including novels, essays, and articles on social and political issues throughout her life and became a supporter of the temperance movement and the women’s suffrage movement. According to Wittke, “[m]ost novels which appeared in the German-American press were of mediocre quality, directed toward the lower-middle-class readers who preferred entertaining newspaper fiction to serious problem literature,” and within the same paragraph, he adds another German-American editor to the list of translators of Stowe’s work: “In 1853 Ludwig N. Demnitz, editor of the Beobachter am Ohio of Louisville, furnished his readers with an original translation of Uncle Tom’s Cabin, entitled Onkel Tom’s Hütte, oder das Leben unter den Niedrigen” (228). The many rewritings that the novel fostered do not only show the popularity of it and the political power of fiction. The transregional ecosystem of textual exchange, which involved several people and groups on a local and translocal level, helped in spreading the word about these emancipatory movements. Through translations into various languages at home and abroad, Uncle Tom’s Cabin became an international bestseller in the nineteenth century. These many rewritings illustrate the popularity of the novel and even European readers’ interests in American topics during the era of transatlantic mass migration. In Germany, it similarly fostered further translations such as, for instance, Onkel Toms Hütte: eine Erzählung aus dem Negerleben in den amerikanischen Sklavenstaaten nach Harriet Beecher-Stowe für die Jugend geschrieben, translated by Andrea Fogowitz under her pseudonym “Heinrich Herold” (1858), and Onkel Toms Hütte. Für die Jugend bearbeitet (1899), translated by L. Kheim.^[9] These two titles already display changes in the act of translation. First, both were translated and rewritten for a young audience. The former also added “A Narrative of N-Life in the American Slave States” as a subtitle to mark a political distinction within the U.S. territory and it mentioned Stowe as the author. In contrast to the translation in Germany, Stowe’s title, however, did not include the N-Word, but “Lowly.” Thomas’ translation in the Philadelphia German-American newspaper used neither of these terms, but Verstossene (“castoffs”). The numerous publications of these novels, as evidenced by these title changes, demonstrate that various adaptations of the text were produced. These adaptations offer valuable insights into local debates, ideas, and ideologies concerning the representation of enslaved people in the United States across different periods and locations. Such adaptations serve as useful resources for examining how German Americans interpreted American culture and how editors and writers translated it for a German-American audience. Like translations of novels, when examining similar fictional and non-fictional texts even within the same language published in newspapers in different locations, however, one will also not always find “identical” texts: “[r]eprinted news does not imply identical copies. Text often gets reframed, especially when translated across languages” (Keck et al., “Lajos Kossuth” 6). Editors changed titles, introductions, and conclusions. They modified, deleted, and added information. Even if editors simply deleted or added words of negation, they could reshape possible interpretations of a text.

In the context of nineteenth-century German-American newspapers, reprints could range from exact repetitions of text to significantly altered versions particularly when a change of language was involved. “Scissors-and-paste journalism was about replicating content, but it was also about making the old appear new” (Pigeon, “Scissors-and-Paste” 14). In the following subchapters, I will discuss the types of sources that German-American editors used to make “the old appear new,” highlighting the significance of translation for the German-American press and how it sets it apart from its American counterparts. To structure this chapter, I will provide case studies of two popular mobile and editable genres that illustrate the blurry lines of fact and fiction in nineteenth-century newspapers: lists and poems. Looking at how texts from demographic datasets to nationalistic verses changed through their travel and were embedded into the newspaper layout emphasizes the merging of poetry as news content as well as the importance of creating and publishing data in the form of lists for constructing imagined communities. This chapter, therefore, also argues for examining non-narrative texts that were reprinted within the same newspaper, as they show that editors regularly changed their own material providing insights into their own understandings of race, gender, and ethnicity. This study complements Cordell’s idea about expanding notions about reprints as fillers and to critically reflect on the role of German-American editors as translators of content from the Old and the New World showing their active role in modifying texts, and thereby shaping public opinion(s).^[10] Such an approach further supports Trish Loughran’s idea, who argues that the nineteenth century was characterized by an information landscape in which “fragmented pieces of text circulated haphazardly and unevenly,” displaying a plurality of “local and regional reading publics scattered across a vast and diverse geographical space” (xix). According to Sebastian Herrmann, the publishing landscape of the U.S. in the nineteenth century was more akin to today’s world of data-driven “filter bubbles” and fragmented, isolated, “tribal micro-audiences” than to the 1980s Andersonian vision of a (still relatively) unified nation bound by the circulation of (still relatively) “coherent national narratives” (19):

It is the material, and the material practices of production and circulation, work in print culture studies regularly asserts, that played at least as much of a role in shaping the literary culture of the nineteenth century as the “authors” and their “works” did, and this assertion again often helps complicate the boundaries between more narrowly literary production and textual production in more general terms. (21–22)

This book is inspired by the current body of revisionist research that aims to decenter the author as the focal point in literary and cultural studies. In contrast, it highlights the roles of publishers, editors, sub- and co-editors to show that textual practices can be considered as “information work” in order to emphasize creative origination. To account for the people in newspaper publishing who contributed to the many interferences in the social life of a text, such as introductions to stories, articles reviewing or referring to these stories, translations or other types of adaptions, Cordell coined the term “network author.” This view expands the model of authorship in the nineteenth century as it embodies a communal rather than individual process, which is distributed rather than centralized (“Reprinting” 418). In this view, Stowe, Thomas, Strodtmann, Mühl, and many more German, German-American, and American editors, writers, and publishers were all connected in a net of authorship. According to Cordell, newspaper selections, in other words, reprints, may appear as insignificant snippets or mere filler (“Reprinting” 37–38). When these selections are identified, grouped, and analyzed as a collective corpus, however, they transform into essential indicators of the period’s print culture, revealing its key priorities, concerns, and sources of entertainment (ibid.).

These first results enabled through the reprint detection algorithm, allowing us to look at texts at scale and more closely, are numerous but they represent only a tiny fraction. Creating a full critical bibliography for all digitized newspapers used in this study, including a complete tabulation of all texts they reused, would require far more time and effort than the scope of a monograph could allow. I do not here propose a system for a wholesome approach of identifying reprinted texts in digitized newspapers, as I will not have space to describe in detail the computational steps that were taken. My aim is more practical by offering what Lincoln Mullen calls a “braided narrative for Digital History,” moving between method and interpretation (382). The procedures described below can help scholars, who are less familiar with computational methods, better understand how text mining works, and why the combination of quantitative and qualitative approaches may provide a better understanding of the nineteenth-century German-American newspaper system. “The rise of an analytical practice of text mining in history bears the marks of a hybrid endeavor that draws from many fields” (Guldi 522). In contrast to the previous chapters, this section introduces more computational terminology; however, I aim to present these concepts in an accessible manner for Humanities scholars in order to highlight both their potential and their pitfalls. Even though the computational detection of text reuse provides for the first time the opportunity to trace mobile texts across time and space, it has limitations for studying text reuse and gender. First, TDR has limitations when it comes to the study of text reuse that goes beyond the repetition of similar sequences of words. It cannot identify rewritings such as translations or parodies. Second, it is difficult to operationalize social constructs such as gender, as the absence of information about people can also be used for analyses revealing gender dynamics and ideologies of that time. To read these stories across genres, however, one needs to (gender) mine (and genre classify) the corpus further, as will be shown in the next chapter. The goal of this chapter is not only to highlight instances of circulating texts that were characteristic of the German-American press but also to illustrate what types of material were specifically valued for reuse or to identify who copied from whom. It aims to introduce the method for computationally identifying text reuse and outlines steps for a “Critical Search,” in which one “alternates between reading primary sources, consulting secondary sources, testing algorithms, and using the algorithms as an index to discover new primary sources that might disrupt the scholar’s original conclusions” (Guldi 536).

3.1. The Computational Study of Text Reuse¶

In the last few years, technological innovations in digitization, that is turning newspapers into machine-readable data, have fostered computational periodical research on text reuse with diverse objectives. For instance, Paju et al. use computational text reuse analysis on a large corpus of digitized newspapers from the National Library of Finland spanning nearly 150 years to examine temporalities of reprinted texts. Oiva et al. deploy this method on a multilingual corpus to examine the routes of transnational news flow regarding the assassination of the Russian Governor-General of Finland, Nikolay Bobrikov, in 1904 (an event which would later feature in James Joyce’s 1922 novel Ulysses). Other studies use computational text reuse analysis to study transnational and multilingual news events to detect vagaries and silences within the system prior to the laying of the transatlantic cable (Keck et al., “Lajos Kossuth”), to theorize the concept of authorship in nineteenth-century newspapers (Cordell, “Reprinting”), to uncover paratexts of literary works (Cordell, “‘Taken Possession Of’”), or to find popular texts that have been neglected by literary historians (Cordell and Mullen). These projects have begun to exploit the full potential of cultural heritage data to unfold how texts were widely reused across the globe “working toward collaborative and integrative approaches to get closer to the shared vision of ‘finding meaning’ in digitized historical newspaper data” (Oberbichler et al. 226). Together, they display that the system of text reuse changed throughout the nineteenth century and that national and regional newspaper landscapes functioned on different modes of operation based on editors’ political affiliations and motivations, technical infrastructures, censorship, and copyright legislations and enforcements. The computational method used to discover reprinted texts is called text reuse detection (TDR). This approach is a type of text mining, where researchers employ automated techniques to identify, extract, and examine patterns and trends within vast digital collections (see Berti et al., Bryan et al., and Potthast et al.). As part of digital text analysis, text mining has been established as a special case of data mining to account for the peculiar characteristics of textual data (Althage 247).

In recent years, text reuse detection has become an expanding field of research that provides the opportunity to explore “large text corpora and the circulation of texts and text passages” (Paju et al. 254). The computational identification of instances of text reuse has fostered projects to examine intertextuality, scissors-and-paste journalism, plagiarism, paraphrasing, literary and scholarly citation, as well as the spread of discourses in historical text corpora such as newspapers.^[11] One of the tools for modeling text reuse was developed by David Smith in the Viral Texts project.^[12] Similar to a plagiarism software for academic papers, he developed an algorithm that would be relatively robust to OCR noise, and could compare different digitized historical newspapers from Chronicling America, detect similar passages of texts in different newspaper editions, extract them and cluster them together in a new database. The software tool that he developed for computational text analysis and document clustering is called PASSIM (Parallel Automated Source Support for Improved Manuscript Studies). PASSIM is freely available for download and use.^[13] It was designed to aid scholars in identifying and analyzing text reuse and similarity in large collections of documents such as newspapers, but it can also be used with other digital editions. This has helped scholars to identify potential sources of influence and inspiration, track the circulation of texts, and trace the evolution of ideas and concepts over time as briefly indicated above.

While data mining processes structure data such as metadata, text mining focuses on unstructured data. Natural language texts are unstructured data. From the computer’s perspective, they lack discernible structure and appear merely as a sequence of characters. However, even text data are measurable. PASSIM compares large sets of texts to identify similar or identical passages.^[14] It uses an algorithm that breaks texts down into small units (such as words or phrases) and then compares these units to identify those that appear in multiple texts. The software then clusters these shared passages together and presents them to the user for further analysis. To do this, PASSIM uses a combination of techniques from natural language processing, machine learning, and information retrieval. It employs various measures to determine the degree of similarity between passages, such as the number of shared words or the frequency of shared n-grams (sequences of n words).^[15] PASSIM has the ability to scale to large collections of texts working with millions of documents and detecting matches across different languages and writing systems. It breaks down the text into overlapping sequences of 5-grams:

die Herausgabe einer Zeitung ist à indexed as “die Herausgabe einer Zeitung ist”: [1, 2]
Herausgabe einer Zeitung ist wirklich à indexed as “Herausgabe einer Zeitung ist wirklich”: [1, 2]
enthält die Zeitung zu viel à indexed as “enthält die Zeitung zu viel”: [1, 2, 3]
die Zeitung zu viel Politik à indexed as “die Zeitung zu viel Politik”: [1, 2, 3]
(Index of 5-grams using sentences of the reprinted text “Freuden eines Redakteurs”)

An inverted index is created for the 5-grams, which maps each unique 5-gram to the locations in the texts where it appears. The next step is to compare the 5-grams between the documents to find matches. Finally, it calculates a similarity score based on the number of matching 5-grams. If passages from separate pages contain at least five matching phrases, each five words long, and these phrases overlap by at least 80%, they should be considered identical and clustered together (see Cordell, “Viral Textuality”). The numbers in the index shown above seek to exemplify these steps. They illustrate the page numbers, where the phrases were found. This inverted index allows the program to find all the pages where each phrase appears.

Using PASSIM, 44,750 clusters of reprinted texts were identified in 36 newspapers from 13 states. While this approach has produced impressive results, as it allowed for the identification of various similar passages and texts, there is always the likelihood of missing some. Shorter texts are difficult to detect because they do not contain enough 5-grams. Moreover, texts will be missed if the raw data that underlies the collection is messy due to OCR, that is when all 5-grams contain too many OCR errors. Texts can also be missed for the very simple reason that editors not only modified some words but large parts of the texts. PASSIM will not find them if they do not share a reasonable amount of n-grams. At the core of every quantitative study lies the determination of frequencies, that is how often a particular value appears within a given distribution of characteristics. Frequency is not a perfect proxy for popularity or cultural importance, but “when taken as the locus of scholarly inquiry,” as Cordell writes, “it produces a hybrid, decentered set of bibliographies that highlight the prevalence of understudied or ignored genres of everyday reading and writing” (“Reprinting” 432). Frequency refers to the number of occurrences of a reprinted text. The larger a newspaper database the more likely it is to find a higher number of reprinted texts. Based on the size of the corpus, the number of similar texts is impressive; however, it represents only a sample of all circulating and reprinted texts in the nineteenth century. In the nineteenth century, routes of transport were changing constantly, which was influenced by and simultaneously had an impact on changing demographics and settlements, wars and riots, and technological innovations of ship and rail systems. These developments make it difficult to track down who copied from whom since editors did not always indicate their sources. The objective of this study is not to find out who copied most extensively or who were the “influencers.” Instead, the aim is to find out what kind of texts were shared between the migrant communities (and represent the German-American press as a network), as they allow us to better understand readers’ interests, the social life of texts, and the editors’ roles as translators of “traveling cultures” (Clifford 316).

How to start making sense of this corpus of reprinted texts? Given the magnitude of the dataset, it was impractical to manually analyze each individual entry. As a usual step in Digital History, I preceded by turning data into visualizations. Network analyses are often used in the course of evaluating large quantities of intertexts (Gruber 274). Once the algorithm converged, I first examined text reuse on the corpus level by means of network visualizations, such as can be seen in figure 7. The figure shows a network graph that visualizes the dataset of reprint clusters. Node size and color represent centrality and community clustering, respectively, while edge thickness indicates the frequency of shared reprints between nodes. The visualization provides a first overview of connections and disconnections between different nineteenth-century German-American newspapers and their editorial exchanges.^[16] Because Chronicling America comprises smaller, state-level digitization efforts, it reflects a wider notion of publishing than much scholarship, which has focused disproportionately on publishing centers like New York, St. Louis, Philadelphia, or Chicago. Yet when the text reuse data are modeled as a network, as in figure 7, it is striking how papers across the U.S., including the regions that go beyond the German belt, are revealed as, in the words of Cordell, “central nodes in the wider network” (“Reprinting” 435). Network research can be a useful tool here to open up new perspectives on the sources through quantitative analysis instead of a qualitative examination of individual newspapers, which in turn can lead to new questions. “The network idea, which has seeped into periodical studies also from other disciplines, for example, science technology studies (STS), and which has given rise to projects relying on large, digitized data and forms of ‘distant reading,’ seems to be one of the most promising new developments in the field” (Ernst and Scheiding 3). Historical network analysis has become a well-established method within the Digital Humanities, particularly in Digital History (Rehbein 259). It seeks to incorporate social science network research techniques into historical studies, beyond using the concept merely as a metaphor. “Network analysis may include the analysis of the positions of nodes to assess relational power or the structure of a network to explain social capital and performance, where the network serves as a proxy for social structures” (Romein et al. 307).

Figure 7: Network graph of newspaper reprints, showing connections between newspapers that reprinted content from one another.

Geoffrey Rockwell and Stéfan Sinclair refer to visualizations as “analytical tools,” which should be seen as “instantiations of interpretive methods” as they allow us to see certain aspects and features of texts or other objects in new ways (4). Graphs like figure 7 are helpful because they provide evidence for connections between publication locations.

The rapid geographic expansion of the ever-vaster young nation met with complementary movements of a massive urbanization and of powerful communicative integration, while an unparalleled explosion of textuality was facilitated by new printing technologies and was powered along by the commercial mass-distribution of printed matter. (Herrmann 231–232)

Network graphs provide a different view or reading of the press system. They allow us to see it as a network shaped by an “unparalleled explosion of textuality” instead of imagining it solely as a composition of printed letters on paper that are hosted in a physical archive. “On a more abstract level, visualizations like these ask us to question and redefine our concepts of community and what it means to be part of a social network with others in a specific context” (Ringler). “Considering reprints as signals of social interconnection among publications, weak in the singular but powerful in the aggregate, we can use shared texts to graph conversation and influence across nineteenth-century periodical culture” (Cordell, “Viral Textuality” 47). The Viral Texts project also uses graph visualization of text reuse at corpus level for American newspapers published during the ante-bellum era. With an interactive network visualization, they provide “a bird’s eye-view of millions of text reuse passages, distilled into a graph that shows how newspapers formed a network of reprints and content reuse” (Düring et al. 3). Although it covers a shorter time frame (1836–1860), the dataset is significantly larger than mine and offers valuable insights into different textual practices. Magazines tend to form tighter clusters, and distinct regional dynamics become visible, particularly the way urban newspapers often initiated or influenced text reuse, which then spread inward to rural areas. Figure 7 presents a much simpler graph than the interactive visualization produced by the Viral Textsproject. Yet it similarly offers an alternative way of reading the German-American press, distinct from the tabular view in the previous chapter. Rather than emphasizing subscription numbers or chronological periods, it focuses purely on the textual exchanges between newspapers. In this way, both the Hermanner Volksblatt in rural Missouri, Der Deutsche Correspondent in Baltimore, and the Freie Presse für Texas are in the center despite their geographical distance. In addition, we can also observe geographical connections such as, for instance, in the case of the newspapers from South Dakota and Nebraska or the ones from Pennsylvania.

But what other purposes does it serve beyond reimagining the press as a nineteenth-century social network, and, just as importantly, what remains invisible through this lens? Network analyses face challenges in data collection from historical sources. These challenges include common issues in historical research like gaps in sources and specific difficulties related to quantitative methods, such as the mismatch between the precision of structural network measures and the sometimes-ambiguous nature of historical sources like newspapers. Because a model emphasizes certain attributes, it inevitably carries subjective bias and represents just one possibility among countless other potential models (Rehbein 265). Aware of the inaccurate and sometimes misleading metadata discussed in the previous chapter, I am hesitant to present these findings as definitive facts. Instead, I regard them as part of an interpretive process that brings me closer to understanding the dynamics of text reuse.

Figure 8: Temporal distribution of reprint activity in the corpus, highlighting periods of increased and reduced reuse.

As every scholar in Digital History knows, no single solution can address all questions. We, therefore, may experiment with various tools revisiting our artifacts through multiple lenses until compelling questions begin to emerge, along with strategies for how to explore them (Rockwell and Sinclair 169–187). The dataset of text reuse clusters can also be visualized in various other ways, including approaches that integrate temporal and quantitative dimensions (line graphs). According to figure 8, most reprint activity in the corpus occurred between 1870–1880. At this point, one might ask whether these peaks correspond to specific events such as technological innovations or shifts in migration flows. This increase in text reuse between 1870–1880, for instance, directly correlated with the period of most rapid growth during the mid-1870s according to circulation and subscription numbers (Bergquist, “German-American Press” 142). The press’ growth continued to a peak in 1892–1893, which also had an impact on text reuse according to figure 8. These two peaks occurred within the third and fourth wave of migration from Germany that brought a readership of newly arriving German immigrants. Moreover, the high reprint activity also correlates with new copyrights enacted in 1870 and 1874. The 1870 Act incorporated the term “fine arts” into copyright law, and the newly established categories, which included paintings, drawings, and statues, were perceived as encompassing the realm of fine arts (Brauneis 586). These advancements marked a significant departure from previous copyright categories, which primarily encompassed works intended for reproduction in multiple copies such as maps, charts, books, prints, musical compositions, dramatic compositions, and photographs (ibid.). Bringing these different numbers and historical developments into dialogue, we could argue that the time frame of the dataset provides an ideal vehicle for a longitudinal study of the way texts were shared within a developing community of readers that represent a significant segment of the millions of German-speaking migrants who streamed to the United States in the second part of this century.^[17] In the 1880s, migration from Germany to the U.S. reached its climax with 1.4 million and in 1890, 2.78 million German Americans of first generation were living in the U.S. (50 million in total) (Bungert 241). This time frame and these peaks suggest that the dataset allows for an analysis of texts about a variety of American topics such as the abolitionist movement, reconstruction, women’s suffrage, and populism, and German issues such as nationalism, socialism, and imperialism.

Tables, graph networks, and histograms entail their own set of limitations. They obscure the textual aspect that reflects the editorial practice of exchange. This means they do not allow for the examination of text reuse at the passage level, focusing on the differences between individual instances of text reuse among pairs of witnesses.^[18] Examining the witness, that is analyzing the reused passage on the newspaper page, is an important aspect for the contextualization and the characteristics of intertextuality. I present these visualizations because they serve an interpretive purpose: they help move beyond close reading and allow us to see the corpus from a distance, prompting new questions. For example, we can observe that newspapers were interconnected, but what kinds of texts did they share? Did they primarily circulate news or (short) fiction? Were entire pages reprinted? Who copied from whom most frequently? And how were texts adapted to reflect local or regional dynamics? “These questions are also interpretive questions, but of a different kind than the interpretive screen the tool provides” (Ringler). These hermeneutical questions reflect our humanistic training as they are messy, not statistically verifiable, and rely on contextual interpretation of the artifacts (Ringler). While such questions cannot be fully answered from a distant reading perspective alone, they are essential for reassessing the corpus in terms of its potential for transregional and longitudinal studies. They also invite us to consider its relationship to broader developments, such as innovations in printing technology and shifts in migration patterns.

Despite its limitations as outlined in the previous chapter, this corpus presents a novel perspective to study the transregional editorial practices of the “network author” and to analyze the interactions between a community of readers, writers, and editors within a network as shown in figure 7. Mark Algee-Hewitt suggests that we should treat authorship as a pattern of word frequencies rather than linking it to the intentions or psychology of an individual. In this way, we can reconceptualize the author and better account for the diverse forces that shaped a text’s creation in the nineteenth century. As I aim to show in this chapter, we can gain a new understanding of the central role of newspapers in German-American publishing culture through the practice of textual exchange. The objective of the following newspaper texts that have I selected for this chapter is, on the one hand, to broaden the role of German-American editors as influencers, brokers, and translators, offering insights into textual practices and their connections to accuracy, cultural sensitivity, and consistency. One the other hand, this perspective seeks to illustrate how genres beyond news, including non-narrative texts, shaped migrant conceptions about race, gender, and processes of datafication.

Many reprinted texts were assigned to different clusters by PASSIM, often due to OCR errors or substantial modifications introduced by editors. Methodologically, this chapter demonstrates the robustness of the detection method in handling minor variations such as altered or omitted words and numbers while also revealing its limitations in identifying translations and parodies. The selected texts range from those reprinted within a single newspaper to those that appear to have traveled the farthest across time and space. Moreover, by focusing on texts not explicitly addressed to women, I aim to underscore the challenge of operationalizing gender. Gender ideologies often surface in texts that do not directly mention women or girls, highlighting the need for interpretive approaches beyond keyword-based methods.

3.2. Lost in Translation: Lists as Popular Non-Linear Narrative Forms¶

As Sebastian Herrmann’s study elucidates, the middle of the nineteenth century saw not only the rise of debates about the value of literature for representing reality, but also of data. “[D]uring romanticism, US culture negotiated the outlines of the literary—what literature is, what literary value consists of, and what literature can do—in relation to the outlines of another representational project that was gaining sharper contours and a stronger foothold in public perception at the time: data” (13). Until today, data is not seen as information, but rather as fact or evidence, which he traces back to the 1850s in his study, to serve a rhetorical purpose. Herrmann who focuses on literary figures including Stowe points out that nineteenth-century abolitionist writers increasingly used data because “they recognize that narrative is an imperfect form when it comes to imagining communities of the size and diversity of the young United States” (243). One of the—what we would now call—data repositories that writers used for their texts was the United States Census: a longue durée example of how people have been comprehensively datafied (and classified). It is in the words of Dan Bouk an “old-fashioned Big Data” repository (5).^[19] Abolitionists utilized the data to quantitatively illustrate the geographical aspects of slavery, including its economic, age-related, and gendered histories.

Throughout the century and particularly after the Civil War, the political landscape of the U.S. evolved, influencing the census’s size and structure due to the nation’s demographic expansion and shifting methods of representing and classifying people. The questionnaire for the 1890 Census, for instance, collected more information than any census before it. To name only a subset, it collected data, both quantitative and qualitative, about people such as the number of persons in a house, names, military service during the Civil War, race (described as white, black, mulatto, quadroon, octoroon, Chinese, Japanese, or Civilized Indians), age, and place of birth (if foreign-born, number of years in the U.S.). They also collected data about sex including information about women, such as if they were mothers and how many children they had (including the ones that had died), their marital status, or if they were a widow of a man who had served in the Civil War. The 1890 Census no longer collected information about enslavement but instead designed new categories to account for Native Americans, Black Americans, and the diverse migrant groups that entered the country during the era of transatlantic mass migration. Both the structure of and (metadata) descriptions in classification systems were outputs for acts of constructing hierarchies among different groups.

As a comprehensive historic source that sought to capture the diversity of the United States, the census also offers the opportunity to trace migratory settlement patterns. According to Kamphoefner, in the nineteenth century, German Americans “[mainly] avoided the South, except for Texas, were lightly represented in New England as well” but predominantly settled in “the urban and rural Midwest” followed by “northeastern cities between New York and Baltimore” (Kamphoefner, Germans in America 66).^[20] These settlement arrangements show that migrants “could spend their lives largely in a German-speaking world if they chose to” and “[i]n some respects, it appears that German Americans lived in a parallel society, linguistically speaking” (ibid. 113). The German-American press in these settlements contributed to this monolinguistic environment. Thanks to editors’ translation efforts, readers were well-informed about national concerns and developments. A reprint cluster of the U.S. Census report of 1890 offers the opportunity to scrutinize this translation work. To translate the United States for a German-reading audience, editors not only reused and translated serialized novels by American women writers but also non-narrative forms of information. These ranged from lists of newly arrived migrants, train schedules, weather forecasts to demographic reports that included both qualitative and quantitative data about people, goods, nature, or time.

Edward Raine, editor of Der Deutsche Correspondent in Baltimore, and before him his brother Friedrich, regularly covered demographic news about the nation by publishing data from the census and catered it to a German-reading public as the examination of one reprint cluster illustrates. A closer examination of similar texts within this cluster reveals insights into local dynamics of knowledge production and circulation, portraying the German-American editor as a translator. At the same time, this reprint cluster draws attention to the translator’s practice in (not) accurately and consistently representing the information extracted from the data repository. When Raine created his own subset by publishing a summarized translation of the 1890 Census report, he decided to cut out data about women, which existed in the “original,” altogether. In the form of a list, he published and translated parts of the information from the report on February 23, 1897, under the title “Flächenraum, Einwohnerzahl u.s.w. der Ver. Staaten” (“Territory, Population and so forth of the U.S.”).^[21] After listing geographical data about the United States, he provided the following categories and population figures: entire population (62,622,250), male population (32,067,880), negro population including the Indians and so forth (7,638,360), male persons over 21 (16,940,311), male persons of military age (13,326,168), and school age population (22,447,392).^[22] Raine’s report ended with listing the ten biggest cities before concluding that the sixteen southern states had 15,549,358 White citizens and 6,898,806 Black citizens.

Data as well as the absence of data provide insight into gender ideologies of the time. Prevailing ideologies of male superiority during that period highlight the fact that even though there were available data about women (including on fertility and marital status), the processes of reprinting, including selection, revision, and translation, often prioritized data about men. Consequently, these acts of rewriting perpetuated a narrative that prioritized male experiences and perspectives over female ones. In Digital History, ongoing discussions center around the fact that what is available online (as illustrated in chapter two) directly shapes what can be analyzed using computational and digital tools. Visual theorist Johanna Drucker argues that the etymological root of the term data, “that which is given,” does not adequately represent its meaning, and prefers to speak of “capta,” which literally means “that which has been captured or gathered” (“Humanistic Approaches”). In other words, each act of collecting data, whether textual or numerical, is inherently aimed at specific goals, using instruments, and selectively focusing on a small portion of what could have been gathered with different objectives and tools. “[D]ata matters for how it mobilizes cultural imaginations: how people think about data, how they think about their world through data, what they associate with this form, and what representational desires they express through it” (Herrmann 41). Raine’s census summary followed a different hierarchical path of data structure in contrast to the original census report. In other words, he did not simply translate the data, but he rearranged it in a way that fit his idea of grouping people. He actively turned data into information that shaped the way German Americans perceived themselves in terms of race and gender but also other groups. First, even though massive data about women existed in the 1890 Census, women’s bodies, irrespective of age, race, or class were eliminated in this process of “copy and paste” while figures about the male population were inserted directly after the number of the entire population. In addition, the editor published incorrect figures about the Black population and mixed-up categories. He used the term “negro population” and the numbers of the entire “colored” population to account for Black Americans even though this also included other ethnic groups. Instead of listing “Chinese,” “Japanese,” and “Civilized Indians” as distinct entries in the category “colored,” he first inserted them into the category of “negro population” and only added “Indian” as another category. He deleted the adjective “civilized” that preceded “Indian” and summarized all other categories—some of them were newly created race categories in the 1890 Census—as “u.s.w.” (and so forth). Such an example shows how translations and abbreviations represent cases of wrongfully (re-)publishing statistics about people through editorial practices. “Statistics, like newspapers, novels, and encyclopedias, have the aim of representing entities too large for an empiricism based on the individual’s senses” (Peters 14). Statistics such as census reports were highly mobile pieces of information in the nineteenth century. They were also a tool to make the invisible visible, as Peters also points out, having the power of rendering some people and groups invisible. Information about native and foreign-born as well as native and foreign White people did not feature in this reprint at all. In the last part of the summary, Raine added information about the racial diversity of the binary system White versus Black represented in the sixteen southern states. Such a conclusionary statement emphasized the educational aims of the newspaper to make the distinction between races, as well as the northern and southern states, visible for its readership.

Raine established hierarchies about people from different locations, races, age, and gender by using different modes of representation. Examining the practice of reprinting from one publication to another is one way to trace (intentional) acts of publishing (dis-)information. Another way to identify how editors wrongfully spread false statistics about groups is to analyze reprinting practices in the same outlet across time. Even though some words were exchanged in different samples of the text, the slightly different versions were identified as reprints and clustered together accordingly. This shows the robustness of the method for minor differences between reprints. Der Deutsche Correspondent did not only publish the census report in 1897. Studying instances of reuse in the same publication can indicate that editors just used texts to fill pages. However, even if the editor of Der Deutsche Correspondent reprinted the subset simply for the purpose of filling pages, he nevertheless regularly reshaped the subset and its content. One year earlier, on February 12, 1896, the summarized report used the same title; however, what differs between the issues of 1896 and 1897 is yet another different hierarchical structure of data.^[23] While the other subset displayed the order “entire,” “male”, and “negro” population, this time the editor had also extracted “Eingeborene” (Native born) (53,372,703) and “White” (54,983,890) from the census report and inserted these between “male” and “negro” population. While in this instance he decided to make “White” visible, he had already printed the wrong figures and categorical entries for “negro.”^[24]

Sociologist Patricia Hill Collins coined the concept “matrix of domination” to think about the organization about oppressions in society. Although her research was designed with Black women in mind, it offers valuable insights into the existence of multiple forms of classification and how power structures shift when these categories are added, compared, or intersected. Kimberlé Crenshaw expanded her concept by introducing intersectionality to study the existence of multiple categories and how they influence one another. Intersectional feminism functions as a framework that analyzes the intersecting systems of oppression and discrimination, such as sexism, racism, and classism, among others, and how they uniquely impact individuals based on their overlapping identities and social categories, including gender, race, class, ability, sexuality, and more. We can examine these intersectional dimensions by examining both visible and invisible data. By first listing Whites and simply bringing other racial groups together, Raine first established a hierarchy among races. This perspective shifts, however, when we bring in another category such as gender. While Black men are visible in Raine’s list, white women do not appear at all. In other words, the parameters of the matrix shift when we bring in other categories. Similar observations can be made in historiographical accounts such as in Wittke’s biographical sketch of Der Deutsche Correspondent. He dedicates several paragraphs to the paper as it “was widely read in Maryland before the Civil War,” became “the state’s leading German paper,” and whose “staff over the years included outstanding German journalists” (214–215). Even though its staff included at least one woman who would later also become the owner of this paper, Wittke’s account, however, further silences the contributions of women as it ends with the lines: Eduard Raine “remained with the paper until his death in 1911” without even mentioning that Annie V. Raine took over after his death (215). The story of the paper did not end at this point; however, this information is not shared by Wittke to his readers. Only two pages later, when Wittke summarizes the main topics that can be found in the press by close reading 30 titles, he argues that by the end of the nineteenth century, “[e]ditors and publishers still expressed strong views on woman’s suffrage” (217). At a meta level, the women’s suffrage movement represented not only the fight for voting rights but also a demand for broader recognition and visibility within public, professional, and political spheres. Acts of non-representation in official reports on national demographics correspond with a time of great societal change for both men and women. As regards newspaper coverage on women’s suffrage, Wittke does not provide examples of newspapers, articles, or specific dates that could illustrate these sentiments. Given the topic’s prominence and contentious nature, it would be reasonable to expect that it sparked additional scholarly initiatives.

Which newspapers held a strong view on gaining women the right to vote? Did their views change over time? Were there geographical differences in who supported women’s emancipation? Just the fact that Wittke puts women’s suffrage on top of the list of popular topics in the German-American press (that also set them apart from American newspapers) arouses curiosity to dig deeper into the pages and uncover how these strong opinions were textualized in the newspapers and in lists. In contrast to prose, non-narrative information (such as the census list) provides further material to study the role of women in the press. These lists cannot only be read as an increasing use of data practices to represent a reality that was already underway in the middle of the nineteenth century, but also as a response to a crisis of masculinity in which participants sought to silence certain voices by rendering them invisible. In this way, we can begin to trace the antagonistic sentiments towards the women’s movements that questioned women’s position in society, their role as citizens, and which led to their exclusion in reports on the country’s progress. Examining reprinting, in other words tracing how editors reused material from other locations, but particularly their own, is a systematic approach to detect, which data were not extracted, and if these acts of selection, in other words, classifying important data from alleged unimportant data, changed over time. As this example illustrates, scholars may not initially consider text reuse within a single newspaper to be a valuable area of study, often assuming that it merely served to fill space or that the reused content consisted primarily of advertisements. However, this overlooks the importance of examining other repeated formats such as lists, tables, and graphs that appeared in the news. These forms of data also warrant closer examination, particularly in terms of their origin and circulation. Much like the digital source criticism discussed in chapter 2, this calls for a critical approach to understanding how information was compiled, reused, and framed within the press. “There are stories in the data. You just have to know how to read them” (Bouk 3). Data is not just a material fact but an expression, one that can give linguistic and quantitative form to the power of inclusion and exclusion. Thinking about data and gender in the nineteenth century allows me to, in the words of Herrmann, “focalize debates about the contours” (14) of women’s representation that played out in American culture and how it contrasted to the representation of migrant communities. The neglect of the extensive information about mothers, single women, and widows in the translation of the census report becomes even more symbolic when considering that these data were processed by women themselves. Ever since the U.S. Census Bureau started using professional enumerators in the 1880s, it has employed women in that role.^[25] Histories of innovation and counting, however, have predominantly hidden and devalued feminized labor.^[26]

Lists including data about people, goods, commodities, or locations were highly mobile and editable forms of information published, reused, and translated in German-American newspapers that represented gender dynamics of this period. While data about women were rendered unimportant by German-American editors, as shown in the examples above, there were other lists ranging from arrival lists of migrants and marriage announcements to obituaries that prominently featured women. With the reprints of lists of demographic data, we can begin to sketch how individuals and groups were represented in the newspapers.

Around the turn of the century, more women had entered the workforce above all because the country needed these laborers: from women as computers—who manually performed calculations—and teachers to household servants. As a result, the last decades of the nineteenth century saw not only the rise of German-American papers, but also of a different migration pattern. Specifically starting in the 1870s, German migration to the U.S. was not only characterized by family or single men migration patterns any longer but instead constituted of a larger group of single migrant women. A keyword search of “Weib” (woman) in the reprinting dataset allowed me to identify another translation of a census report shedding light on how German-American editors utilized census data to address gender-related realities and conflicting viewpoints. Decades earlier in 1872, Der Deutsche Correspondent reprinted “Das Weib im amerikanischen Zensus” (“Woman in the U.S. Census”), a narrative account which used the data extracted from the 1870 Census that outlined areas in which women worked.^[27] The data revealed that amongst other things women worked as cattle herders (45), trappers (2), pilots (1), and gun smiths (33). The text, however, did not state these figures to emphasize emancipatory changes and women’s representations in traditional male professions, but to pursue two goals: it emphasized the hardships of labor and criticized American women’s (non-existing) work ethic. It argued that while the average American woman considered herself too good to do hard work, many women particularly in the rural areas—i.e. regions with a strong concentration of German migration—were forced to work hard “des lieben Brodes Willen” (to make a living).^[28] As this example elucidates, Raine reused data about women not in reports to imagine the contours of the entire nation but for narratives designed to react to current events, to raise anti-emancipatory sentiments, and to respond to men’s discontent about women’s progress in the labor force. Even though Raine had access to data about women, he opted not to (re-)use the extensive census data; rather, he repurposed it for different aims. While the previous examples depicted a narrative of progress, this reprint elucidates how data transformed into a narrative of complaint, aiming to shape a distinct identity for men and women, particularly focusing on the different nature of American and German-American women. The matrix shifts once again when the category of ethnicity is introduced, producing narratives of “us” versus “them.” From an intersectional perspective, this illustrates how systems of power such as gender, race, and ethnicity interact to shape exclusionary discourses and reinforce social boundaries.

In the nineteenth century, women’s suffrage evolved into significant social movements in both the U.S. and Europe, as activists and reformers organized rallies, petitions, and advocacy campaigns to demand equal voting rights for women. These movements gained momentum and widespread support, challenging entrenched social norms and inspiring legislative change across various countries. Pioneers in the U.S. and Europe played crucial roles in mobilizing people and raising awareness about the importance of women’s enfranchisement not only through public events but also through media outlets. At the same time, however, such advocacy prompted reactions, which differed among certain groups and societies. As Allison Clark Efford argues, the majority of German-American editors could still not bear the thought of granting women equal rights (48). The reprint of the translation of the 1890 Census report elucidates how editors were able to textualize their male superiority and female inferiority beyond articles on national and international news, discouraging the idea of equal rights for women without explicitly targeting German-American women as objects of discussion in their publications. Reprinting reflects the perceived interest in news while it also amplifies that interest through its wide circulation, “a feedback loop” that Cordell describes as virality (“Reprinting” 418). Cordell uses the metaphor of virality, which is nowadays widely associated with information flows on the internet, to describe reprinting practices in nineteenth-century American newspapers. The different arrangements and retellings of the report illustrate this feedback loop even within one newspaper outlet. By constantly reprinting the report, Raine used the genre of the list instead of prose to educate readers about the changing demographics in which race explicitly played a crucial role, while gender was silenced. Even though women had become such an important readership for the German-American press due to the increased migration of German women in the last decades of the nineteenth century, editors continued to shape public opinion and reinforce the societal resistance to women’s emancipation through articles and lists. Lists, however, were not the only short forms that editors used to spread political opinions. They also used other, covert or disguised, strategies to attack changing traditional gender roles through the pages of the press, particularly in the feuilleton. Such political voicing through the allegedly non-political part of the newspaper was rooted in the failed revolutions of 1848 in Europe. In the immediate aftermath of the failed revolutions, editors in Germany could no longer openly express their radical views in newspaper articles. Increasing censorship forced many publications to shut down or alter their content. As a result, editors increasingly turned to the feuilleton section as a space for more veiled political and cultural commentary. Many of these texts, ranging from short and long book reviews, travelogues, studies of customs, popular science reports, biographies and obituaries, to articles on art, theater, and music, as well as novels, novellas, humorous sketches, anecdotes, jokes, and above all poetry, also crossed the Atlantic. They were published, reused, and commented upon extensively in the German-American press, circulating widely and contributing to broader currents within American cultural life.

3.3. The Music of Migrants: From Poetry to Parody¶

“May let these words our watch-cry be, / Where’er upon life’s sea we sail: / ‘For God, for Country and for Yale.’” These are the last three lines of “Bright College Years,” which up to World War I was the official alma mater of Yale University. This tune that celebrates American ideals about Christianity, unity, and loyalty is a standard element of commencement, sports games, and alumni get-togethers (see Goldblum). “Bright College Years” is nowadays considered the unofficial alma mater, as Judith Ann Schiff, Chief Research Archivist at the Yale University Library writes. “But though the song—an instant hit when [Henry] Durand penned it during his senior year—has become a permanent part of Yale tradition, there was a time when the future of ‘Bright College Years’ was very dim. The problem was the music,” or more specifically, its roots: which lie in the rise of German nationalistic movements. In 1881, ten years after German Unification, Henry Durand, who graduated from Yale the same year, wrote the lyrics to a musical piece by German composer Karl Wilhelm. In 1854, Wilhelm had composed this melody to a poem titled “Die Wacht am Rhein” (“The Watch on the Rhine”) by Max Schneckenburger, a Swabian poet, who had previously penned the lyrics in 1840 while residing in Switzerland.

In the poem, the speaker makes a thunderous call that seeks to animate all German men to unify, rush, and defend the German Rhine against their archenemies: the French. The poem uses the symbolic significance of the Rhine River to explore the complex issue of national identity—and masculinity—influenced by the historical shifts in the territories along the Rhine and their changing national affiliations.^[29] Throughout the decades, the text shifted its form influenced by various particular events that helped in increasing its visibility and popularity. As a song, it was performed on the silver wedding of the Crown Prince of Prussia, later Emperor Wilhelm I, in Düsseldorf the same year, and subsequently became the unofficial German national anthem reaching its climax of popularity during the Franco-Prussian War (1870–1871). Historian Sabine Giesbrecht observes another peak at the beginning of the twentieth century. She argues that “Die Wacht am Rhein” was one of the most popular songs on postcards distributed across Germany from 1914–1918 and advanced as a national anthem during World War I (72), where it was instrumentalized as a musical “propaganda weapon” (88). In December 1919, the Yale College Council suggested to ban “Bright College Years,” as Ella Goldbaum writes, due to its German nationalist background and Germany’s role in and responsibility for World War I. However, it failed. While “[f]ive alumni clubs voted against the tune,” twenty clubs voted in its favor arguing that “despite the perceived political implications of the era, the anthem’s history was too precious to change.” To this date, not only the tune of “Die Wacht am Rhein” has found its way into American popular culture being reused by American educational institutions. As a patriotic song, it has also been featured in mass media productions such as the popular American romantic drama film Casablanca (1942). There, in Rick’s Café Américain, it is sung by a group of German officers during World War II to symbolize German militarism and nationalism. However, it was not only used because of its perceived symbolic value for Germans during that time but also because the directors could freely use it as it did not fall under any copyright restrictions. In Germany, it has seen several peaks in popularity during the Franco-Prussian War, World War I and II, and it has become popular among Neo-Nazi groups. In the 1990s, it was rewritten by Landser, a successful German right-wing rock band, into “Die Wacht an der Spree.”^[30] “[B]y highlighting the transformative relations between texts, the specific relationship between literature and memory becomes clear” (Henke 971). “[I]ntertextuality offers a model for how memory travels in literature” (ibid. 973).

Since it was penned by Schneckenburger, “Die Wacht am Rhein” has undergone many transformations—anthem, melody, song, lyrics, poem, stanza, or simply text—serving as a powerful mnemonic device for almost two hundred years to evoke and reinforce German strength and unity. Historians of the Franco-Prussian War (1870–1871), German Unification (1871), and World War II (1939–1949) have likewise illustrated within their time frames that “Die Wacht am Rhein” was performed by singing societies and military troops alike in Europe in times of conflict and war. They all argue that the song was more popular than, for instance, the much older “Heil dir im Siegerkranz” (1790), which was the former Prussian hymn and in 1871 became the official anthem of the first German Emperor Wilhelm I. Unlike the latter it emphasized the nation instead of the monarchy. According to Bungert who examines festivity culture of German Americans, “Die Wacht am Rhein” not only gained popularity in Europe, but also quickly conquered the singing societies in the U.S. around 1860 (224). In the 1850s, a million Germans migrated to the U.S. not only due to economic hardships but also due to political unrest and failed revolutions. The poem portrays a response to migrants’ anti-monarchy sentiments influenced among other things by Bismarck’s derogatory judgment of German migrants as traitors, as Bungert further elucidates (221). While studies like these emphasize and illustrate the popularity of the song, they place it predominantly in the context of European war history and make the general spirit of battle and dissemination of news responsible for its success. Such a view illustrates the performative act of the poem to account for its popularity but nevertheless limits the written or published history of the text, and the value of its form.

In the nineteenth century, poems constituted one of the most viral phenomena in newspaper publishing because they were “fluid, adaptable for new contexts and audiences” (Cordell and Mullen 29). Poems became a popular and versatile genre not only due to their brevity and adaptability but also because of their immediate emotional resonance. “Die Wacht am Rhein” was extensively performed by migrant communities at various events. Simultaneously, it circulated in written form across states and decades through the network of German-American editors and, thereby, transformed into a hymn for German Americans from Maryland to Minnesota to celebrate their long-distance nationalism and express their senses of (be-)longing. Due to its size, adaptability, and migrants’ conflicting relationship with nationality, the poem constituted one of the most viral texts identified in this corpus. As I will show in the following paragraphs, it was widely printed, reprinted, edited, and translated in German-American newspapers “at least” hundreds of times from Ohio to Texas. Looking at the temporal distribution, it appeared almost in every year from 1866–1900, which suggests the popularity of the text among the German-American newspaper editors and readers, adding a quantitative layer to Bungert’s observation. I use the term “at least” because I could identify 259 search results for the string “Die Wacht am Rhein” in this corpus to find texts about the poem that were classified into different clusters. In addition, I might have missed some texts due to OCR errors and, most importantly, there are many more digitized newspapers now available that might allow me to increase this number. A search query of the phrase “wacht am rhein” with the filters “German” and “1830–1900” in Chronicling America inFebruary 2023, for instance, returned 1,476 results. “At least,” therefore, implies that since digitization continues, the likelihood of finding more search results will continue to increase in the following years.

Numbers alone, however, cannot account for a text’s popularity and virality. Researching such a text in the migrant press allows us to trace the reuse and adaptability of poetry in the context of long-distance nationalism and how it blends the literary with the informational while constructing and mobilizing an ethics of consensus and remembering (Bungert 220). In the German migrant press, the reuse of the poem was not necessarily in fidelity to the text or to textual consistency. Through its travel across states and decades, it transformed into different textual forms, from poetry to news, whereby it changed its language, form, style, and motives. As a poem, it was, for instance, published in its entirety on September 15, 1870, in the Minnesota Staatszeitung (1858–1877), during the Franco-Prussian War, which lasted from July 19, 1870, to January 28, 1871.^[31] Here it was printed on the first page beneath the section “Belletristisches” even with author attribution.^[32] According to the census of 1890, the German population in Minnesota consisted of 41,364 persons born in Germany making up approximately “a fifth of the State’s population” with most of them living in Ramsay County, particularly in St. Paul (Arndt and Olson 220). The weekly Minnesota Staatszeitung from St. Paul was one of five German-American newspapers. This changed, however, in the following years and before World War I the number increased to 19 with “1 daily and 18 weekly” papers (ibid.). According to the metadata in Chronicling America, Samuel Ludvigh was the publisher of the paper. Before Ludvigh, a Hungarian migrant, took over the paper in 1858 (until 1860), he had worked as editor for the Philadelphia Alte und Neue Welt. Like many other male editors, he changed locations to pursue his profession, a pattern that was far less common among women editors. In 1858, Ludvigh was touring the country to promote his own publication Die Fackel (1843–1869), a weekly “Literaturblatt für Ludvigh’s Reden, Vorlesungen, Gedichte, und prosaische Aufsätze” (Subtitle: A Literature Journal for Ludvigh’s Speeches, Lectures, Poems, and Essays). First published in New York and Baltimore before moving to St. Paul, Die Fackel was “a highly personal journal,” which he used “as a platform for spreading his rationalistic views” and was “read all over the nation” (Arndt and Olson 189). Ludvigh became known as “Fackelträger” (torchbearer) or “Fackel-Ludivigh” because he was “travelling extensively” to secure more subscribers which he needed to continue with the publication. Ludvigh’s example elucidates how German-American editors toured the country to promote their publications and simultaneously displays the value of literature for advocating political thought. It also sheds light on the complex regional identities encompassed by the German-language press, which also served Hungarian German readers among its diverse audience. Die Fackel was published weekly until 1851 and afterwards quarterly until publication under Sarah Ludvigh, his widow, ceased in 1869. As one might expect, his wife’s name is also a missing element in Chronicling America’s metadata. I may be repeating myself by adding more and more women’s names to the record of German-American publishers and editors. However, naming them is intentional: it draws attention to the gaps in data collection and invites us to imagine what a comprehensive metadata catalogue of the German-American press might look like if it fully reflected the diversity of those who produced it.

In contrast to the Baltimore newspaper, the Minnesota Staatszeitung, whose name was altered by Ludvigh (Minnesota Deutsche Zeitung, 1856–1858) to mirror German-American state newspapers in New York and Illinois, changed its owner several times. After Charles Reuther, Christian Exel, and A.R. Kiefer, the paper was published by Forty-Eighter Albert Wolff, exiled from Germany, and his business partner Theodor Sander. Wolff had acted as editor when Reuther and Exel were in charge of the paper. As owners, he and Sander changed the 4-page and 7-column weekly to an 8-page and 6-column newspaper and also launched a tri-weekly 4-page and 6-column paper under the same name that mainly covered national news with the aim to expand their readership across the state. “Die Wacht am Rhein” was published in the Minnesota Staatszeitung when it had approximately 1,800 subscribers (ibid.). At this point, Wolff had become State Commissioner for Germany and Sander became the main publisher and editor. According to the masthead, the paper could not only be sent across the country ($ 2.50 per year) but also even to Europe for $ 4.65 (postage free). Publications did not only travel westward but German-American newspapers were also read by the people on the other side of the Atlantic. Editors in Germany used these papers to inform the people back home about political, cultural, and social life in the New World.

In Minnesota, the poem was published shortly after the Battle of Sedan (September 1–2, 1870), a city near the river Maas. Victory in this battle was decisive for the outcome of the Franco-Prussian war because it also resulted in the capture of Emperor Napoleon III. Other texts on this front page provided further reading material that depicted both the political tensions in Europe. The poem was followed by a serialized historical novel titled Kaiser Alexander und seine Zeit (Emperor Alexander and His Time) by L. Mühlbach.^[33] The extract of the novel published in this edition portrayed anti-French sentiments—what we would today call hate speech—evidenced by the lines “wir hassen die Franzosen” (“we hate the French”) and “Napoleon, dem feindlichsten Barbaren” (“Napoleon, the most hostile barbarian”). L. Mühlbach, also known in Germany as Louise Mühlbach, was the pen name of popular novelist Clara Mundt. Mundt was active with her husband in the 1848 revolution. Before writing historical fiction, she predominantly published works on emancipation and the right to divorce (e.g. Frauenschicksal in 1839 or Nach der Hochzeit 1844).^[34] Both Forty-Eighter men and women turned to fiction, and specifically the feuilleton, to express their political ideals, personal experiences, and critiques of society following the failed revolutions of 1848. Texts of the feuilleton served as a tool for people to express their worldviews, disguised as fiction. On this page, “Die Wacht,” a historic text of nationalistic and nostalgic nature, was put into dialogue with Mühlbach’s novel. Together, they created a fictional unit that can be interpreted as a response to the ongoing war. This connection illustrates how historical texts were repurposed to comment on contemporary events, offering readers a deeper understanding of the cultural and emotional landscape during the war. They make us aware that we should pay attention to the prose and poetry that was printed adjacent to news articles and editorials. “Die Wacht” continued to be reused and printed, though not always in its entirety. The title, stanzas, or half-stanzas were embedded in news reports ranging from American celebrations such as the President’s Day in Washington DC to the military recruiting of German migrants in 1870, or factual texts about its similarity to the “Star Spangled Banner” and “La Marseillaise.” By the end of the nineteenth century, even editors themselves commented on the success of the “tune” and published information on its history such as on August 1898, in Der Deutsche Correspondent, under the section “Briefkasten” (“Letterbox”).^[35] The article explained that “Die Wacht am Rhein” was not written because of the outbreak of the war in 1870, but already decades earlier in 1840, having been triggered by discussions about changes to Rhine territories. It continued to argue that the poem did not become famous until Carl Wilhelm composed the tune in the 1850s.

The example of “Die Wacht am Rhein” not only provides insight into the role of poetry in shaping what Anderson calls “long-distance nationalism,” but also serves as an illustration to address the role of authorship in the context of popular verses (58). According to Henke, to study intertextuality, “scholars have to lay bare the texts’ references to other texts, but they do not have to ask about the authors’ conscious purposes” (972). What I mean by authorship here is the question of whether popular verses were accompanied by the names of their authors, whether authorial authority or ownership remained visible during their circulation, or whether this aspect was omitted in the course of their transmission. Such observations raise important questions for future quantitative analyses of author identification, as popular texts, particularly poetry, were often published without attribution to their “creator” in the press (see Keck et al., “Who’s in the News?”). Even though the article in Der Deutsche Correspondent mentioned Max Schneckenburger as the author, his name was only published in brackets in comparison to the tune’s composer. The text concluded, however, that both poem and melody became the common property of the German nation when the war broke out in 1870 (“Gemeingut der Nation”). Two months earlier, under the same section titled “Briefkasten,” Der Deutsche Correspondent had printed a shorter text, which only stated that “Die Wacht am Rhein” was written by “M. Schneckenburger” in 1840 and set to music by K. Wilhelm in 1854. A month later, in Chicago, the editor of the Abendblatt (Illinois Staats-Zeitung), Andreas Simon, was furious about a news report he had found in the English-language press (Evening Post), which claimed that the melody of the Dutch national anthem sounded like the “Wacht am Rhein.” The Evening Post argued that since the song was written specifically for the Franco-Prussian War, the melody must be the result of plagiarism. The editor of the Abendblatt emphasized that people should not write about things they knew nothing about and concluded that such an embarrassment could have been avoided by simply casting a glance into the “Conversationslexicon.” The popularity of a text also resulted in critical international debates about its origins leading to heated discussions about piracy and plagiarism.

The various transformations and recreations from poem to song, from news to translations (“The Guard of the Rhine”), established “Die Wacht am Rhein” as a traveling verse about German nationalistic ideas being reproduced over a relatively extended period in the United States with several peaks from 1870–1877, in 1886, and 1895–1897. In comparison to his lyrics, Max Schneckenburger received only little attention in the news and was mainly unknown to German migrants who were not from the southwestern states of Germany. The author’s name seldomly accompanied these texts, which highlights the prominent lack of authorship attribution in the papers, specifically in the context of poetry. These findings support Cordell’s concept of the “network author” displaying the many interferences of a text once it has left the confines of its writer. Schneckenburger could not witness the success of his verse and its mutation as he had already died in 1849. “Die Wacht am Rhein” played a dominant role for German Americans to express their sentiments about “long-distance nationalism,” revealing how migrant communities constructed and mobilized an ethics of consensus and remembering, but also to respond to political and social developments in the United States.

These different versions of the poem underscore the concept of virality, encompassing a more complex and fragmented approach to the “social text” (McKenzie), “aligning fragmentary texts and textual echoes not only through books but also through ephemeral and hybrid media,” such as nineteenth-century newspapers (Cordell, “Viral Textuality” 36). The concept of the “viral text” goes beyond official and unofficial reprintings of a specific poem. It also includes parodies, quotations, reviews, paraphrases, allusions, and other forms of reception items. Virality explores the interconnectedness of texts across various media forms, highlighting the dynamic and evolving nature of textual dissemination in society. Apart from quotations, comments, and translations, the popularity of “Die Wacht am Rhein” becomes most visible by the adaptations it inspired. “Intertextual analysis, thus, can work out the dynamics of the similarities and the differences between the literary representations and the commemorative contexts in which they are embedded, in order to examine those processes in particular” (Henke 974). Similar to contemporary internet memes, adaptations serve(d) as indicators of a text’s popularity and enable(d) ongoing discussions to express discontent. But the more the reused text is rewritten, the more difficult it becomes to “computationally” identify it because the sequence of characters is inherently different (Gruber 271). “Die Wacht beim Wein” (“Keeping watch next to the wine”) was one of the parodies that emerged during the time of war but despite the first two words in the title does not have any other overlapping sequence of characters. Henke offers a set of guiding questions for intertextual analysis, particularly suited to cases involving not reprinted texts, but rather syntactic and semantic mutations: “What functions do the patterns, contents, and plot structures, referred to in the manifest text, assume in their new context? Which of the features that are explicitly, implicitly, or metonymically adopted by the new text are reshaped, transfigured, erased, or left untouched?” (974). In contrast to the “original,” in which men are called to the battlefield to protect their home, in this humorous tale a woman is anxiously awaiting her husband’s late return at night. Filled with worry, she contemplates the various possibilities why he does not return even going as far to assume an act of infidelity. However, when her husband finally arrives, she realizes that her worries were not necessary. The man was merely out drinking, in other words, instead of guarding his home, he was somewhere else guarding the wine. According to the introduction in the Scranton Wochenblatt on March 16, 1871, printed between the parody and the title (including its subtitle in brackets informing that the melody was well-known), the text was created for a festivity (see figure 1, bottom of middle row).^[36] More than 120 miles away from Scranton, Pennsylvania, it was performed in New York City on Ash Wednesday, marking the Christian holiday of abstinence and fasting.

Figure 9: Poem “Die Wacht beim Wein” in Scranton Wochenblatt, March 16, 1871.

Carnival speeches historically possessed a significant political dimension (Bungert 127).^[37] Within the German-American context, these speeches responded to current events in both Europe and the United States. While the woman in the text remains unnamed, her husband is identified as Wilhelm, an allusion to Emperor Wilhelm, which anchors the parody in themes of German national identity and patriotism. The use of the name Wilhelm adds a satirical layer by portraying the emperor as excessively drunk, thereby undermining his dignified image.

Beyond the reference to the Franco-Prussian War, the parody also resonates with contemporary political developments. By depicting Wilhelm as excessively intoxicated, the parody criticizes excessive drinking. This satire aligns with the temperance movement, which advocated for moderation or abstinence from alcohol. Emerging as an international campaign primarily in Western countries, the movement advocated for abstinence from alcohol and emphasized its harmful effects on health and family life. By the 1860s, it had evolved into a widespread mass movement, strongly supported by American women of the working and middle class, and ultimately led to the introduction of prohibition in several countries, including the United States from 1920 to 1933 (Dannenbaum 236). The text’s ridicule of a prominent figure through the portrayal of drunkenness reflects the movement’s concerns about alcohol’s negative effects on individuals and society at large. It echoed women’s fears of losing their husbands to alcohol, as well as concerns about men’s drinking behavior and its potential impact on their ability to work especially given that men were, in most cases, still the primary breadwinners. “Vielleicht passierte ein Malheur / O Gott, wenn ich schon Wittwe wär!” (“Perhaps an accident happened / Oh God, what if I were already a widow!”). “Lieb Frauchen” (“Dear wifey”) is deeply scared and worried as she does not know whether her husband was murdered, thereby also critiquing urban life in New York, which was marked by a high crime rate and social unrest. Even though we do not know whether this particular parody was authored by a man, a woman, or both, it confronted the exclusion of German-American women in this private, but simultaneously, political matter. In contrast to temperance advocators, German-American men were generally opposed to prohibition, as beer and other alcoholic beverages were integral to their cultural and social life. Many German-American communities had thriving brewing industries, and prohibition threatened their economic livelihood. Additionally, the temperance movement was often seen as an attack on their cultural traditions and way of life. The resistance to prohibition among German Americans was significant, as they viewed it as an infringement on personal freedoms and a disregard for their cultural heritage. From a women’s point of view, we can also read the parody as a response to women’s dissatisfaction. Published anonymously, such a critique rebuked the constant absence of men in the evenings for club meetings, theatrical performances, or visits to saloons, while women remained at home tending to the children. According to Wittke, the German-language press often reflected the conservative attitudes of many German migrants, who were generally opposed to not only women’s suffrage but particularly prohibition (217). This opposition was rooted in traditional values and a desire to maintain established social norms within their communities. While German-American men did not support the temperance movement, Kamphoefner argues that women prioritized their ethnic identity over gender, aligning with their male counterparts in their aversion to the movement (Germans in America 179, 224–225). According to Johnson, German-American men did not want to grant women the right to vote as they feared that they would vote in favor of prohibition (149). As with any text, the parody invites multiple interpretations. While it can be read as mocking the temperance movement, it also serves as a historical source (albeit cloaked in humor) that engages critically with broader political conditions of the time, including war in Europe, migration, temperance, and the suffrage movement. In this sense, humor becomes a powerful vehicle for confronting social and political tensions without naming them directly. Analyses of women’s pages and magazines, however, reveal a different picture that show women’s opposition to alcohol, as letters to the editor reveal (Blaschke, “German Immigrant Women” 323). By offering “the female perspective,” the issues provide “key indicators of changing cultural values” (324).

The complexities of the editorial practice of content reuse and commentary reveal the relationship of ideology to material print networks and professional newspaper work. These findings suggest that, while editors routinely appropriated content from other publications, their revisions—as well as their decisions regarding layout and placement—constitute acts of reimagining and rewriting. Broader historical patterns begin to emerge when these editorial practices are situated within the wider exchange networks of the German-language press. Towards the end of the nineteenth century, the German immigrant press experienced an increase in capacity and volume. This resulted in a faster circulation of information, which becomes evident in the dataset as most clusters emerged during this time. However, there were some texts that seemed to be timeless such as the nationalistic poem “Die Wacht am Rhein.” Examining its transformation into a parody or its incorporation into the newspaper layout further reveals that the poem’s significance extended beyond battle and patriotism. It came to symbolize vigilance, duty, and a protective presence resonating with ideologies ranging from preserving one’s heritage to protecting one’s rights. The poem’s popularity can be attributed primarily to its portability and textual adaptability, allowing readers to connect it to various social, political, and economic issues of the nineteenth century. Additionally, its humorous take on serious matters resonated with the public, further cementing its widespread appeal and relevance.

Through intertextual references, earlier texts are remembered and reconfigured in subsequent ones—extended, challenged, or transformed—thus creating new frameworks for what Lachmann refers to as “premediation” (Memory and Literature 37). Intertextuality (including text reuse as a method), therefore, allows us, in the words of Daniela Henke, to “identify writing as doing memory rather than as staging memory” (971). In the dataset used for this study, “Die Wacht am Rhein” was published at least hundreds of times (167) between 1860 and 1900 from Ohio to Texas. Through its travel, the lyrics transformed into different linguistic varieties, were shortened and recontextualized in news and announcements, and rewritten into parodies. The phrase “Die Wacht” became so popular among German Americans that it lead to the founding of papers and magazines that adopted its name, such as, for instance, the weekly newspapers Die Wacht (1884–1888) and Die Wacht am Hudson (1871–1878) in New York City, the monthly paper Wächter am Winnebago (1858–1860) in Oshkosh (Wisconsin), the weekly Der Wächter am Wisconsin (1878–1880) in Wausau (Wisconsin), or the Wächter am Erie (1852–1893), the weekly edition of Wächter und Anzeiger (1852–1954) in Cleveland (Ohio), one of the longest running German-American newspapers. Still boosting a circulation of 34,000 in 1915, it was the only paper after World War I in Ohio that did not receive a permit by the government that proved its loyalty to the U.S., which meant that all issues had to be translated (Arndt and Olson 470). As these titles suggest the popularity of the poem was not only responsible for the reproduction of the text but in general turned “Die Wacht” into a famous phrase. A simple keyword search for the term “Wacht” in Chronicling America may lead to over-inclusion, yielding an overwhelming number of results. However, by combining text reuse detection with keyword searches, frequency analysis, and close reading, it becomes possible to trace the broader histories of mobile popular texts: their transformations over time, including adaptations such as translations and parodies. These adaptations serve as key indicators of textual virality and cultural resonance.

3.4. Pre-Produced Pages “Für unsere Frauen” (“For our Women”)¶

The concept of viral news is not a new one, but the internet and social media have made it possible for news stories to spread more quickly and widely than ever before. Virality refers to the degree to which a piece of content such as a news article, meme, or social media post spreads and is shared rapidly and widely on the internet or social media (Cordell, “Viral Textuality” 32–33). When a piece of content goes viral, it often attracts a large audience and generates a high level of engagement, such as likes, shares, comments, or views. The factors that contribute to virality can vary, but typically include things like emotional appeal, humor, novelty, controversy, or relevance to current events or popular culture (ibid.). While the concept of virality is often associated with modern social media, as Salmi points out, it can be argued that the essential condition for virality, that is the rapid dissemination of information and messages, was already present during the expansion of the press in the nineteenth century (“Long Term reuse” 71).

To provide readers across the country with reading material, German-American editors did not only exchange and reuse news reports or parodies. They even shared entire pages. On May 24, 1895, the Scranton Wochenblatt (1865–1918) in Scranton, Pennsylvania, which had a subscription size of approximately 1,000 at that time, printed a text titled “Jugendspiele und Volksgesundheit” (“Youth Games and Public Health”) on its second page (Arndt and Olson 597).^[38] The article, which discussed the importance of playing for children, argued that their freedom to play was being restricted in cities due to a lack of space. The author of the text, who was not mentioned, suggested that parents were also at fault for not allowing children to play freely and for dressing them in restrictive clothing. The article emphasized that playing is a natural human need and should be encouraged for physical and mental health, but it has been neglected in favor of more academic pursuits. In addition, it suggested increasing the amount of time for children to play and for physical activities to respond to the increase of melancholia, which was said to be “contagious” among the young generations. It was printed two days earlier in Der Deutsche Beobachter (1869–1911) in New Philadelphia, Ohio, which had a similar number of subscribers at that time (597).^[39] The page featured several other opinion pieces, poetry, and prose, such as a history about fashion (“Die Geschichte der Mode”), a poem titled “Flitterwochen” (“Honeymoon”) by German writer Richard Zoozmann, and many jokes (some of which were written in vernacular) such as “Abgelehnt” (“Rejected”). This humorous piece tells the story of Amalie and her husband. Amalie asks her nameless husband why he never pays attention to her or asks about the reason for her misery. In another joke titled “Überrumpelt” (“Taken by Surprise”), someone asks a shoemaker how he came to marry his wife, who responds that when he was measuring shoes for her, the mother came around and gave them her blessing. The page also featured reports about “Berühmte Todte” (“Famous Dead People”), which listed men, among them Karl Vogt and Gustav Freytag. The images on the page featured a range of subjects, including women, men, children, animals, and buildings. As these several examples of content, layout, and images indicate, this was not a regular page, but a “woman’s page.” These pages were invented in the 1890s to offer women a gendered space in the newspaper that discussed issues and concerns predominantly related to women. According to Blaschke, German migrant newspapers quickly adopted the concept of women’s pages by American papers to provide content for the ever-increasing numbers of female readers by the end of the nineteenth century (Die Entdeckung 93). When examining other reprinted texts, it becomes evident that the Scranton Wochenblatt and Der Deutsche Beobachter had a special relationship of exchange as they regularly reused pages throughout the decade. On August 17, 1894, for instance, the Scranton Wochenblatt reprinted a page that featured many texts for, by, and about women including an authorless report titled “Für unsere Frauen: Das Wissen der Hausfrau” (“For Our Women: A Housewife’s Knowledge”) and “Die Wahre Schönheit der Frau” (“The True Beauty of a Woman”), and jokes such as “Nutzlos” (“Useless”) or “Ungalant” (“Discourteous”).^[40] The page also featured an article that was accompanied by a woman writer named “Elise M.” (“Über das Heirathen” [“On Marriage”]). These reprint examples of women’s pages illustrate what Blaschke describes as the use of “vorproduzierte Ware” (preproduced goods) (28). Since the texts in the reuse clusters are not classified by genre and provide no contextual information about their placement within the newspaper, only a redirect to Chronicling America and a visual analysis of the layout reveals that the page in question is a woman’s page.

The evolution of women’s pages provide insight into the changing landscape of the migrant newspapers around the turn of the century. The women’s pages allowed women to have their own gendered section, but at the same time it resulted in a repositioning of information on child education that, in the press, became exclusively connected to women. It is therefore not surprising that these pages—and the texts featured on them—show a high frequency of terms referring to women. While women’s pages provided female writers with greater visibility and offered readers content on topics such as women’s health and fashion, they also reinforced the division of gendered spheres traditionally assigned to men and women. Advice for parents, jokes about naïve girls, or reports on important people were also published in the preceding decades; however, these were typically within non-gendered spaces in the newspaper. A comprehension of the history of the migrant press at the turn of the century, but also in the preceding decades, requires the inclusion of concerns, topics, and issues related to women. The following chapters, therefore, aim to demonstrate how these themes were textualized beyond the confines of women’s pages.

Recent technological innovations have enabled computational periodical research to study text reuse with various objectives. To name just a few, projects have examined temporalities of reprinted texts, the routes of transnational news flow, and the concept of authorship in nineteenth-century newspapers, among others. Text reuse as a phenomenon, however, seems more complex than the examples above can capture. This chapter has introduced text reuse detection as a computational method outlining its usefulness in addressing research objectives of popularity, authorship, and news circulation. Computational text reuse detection, when combined with digitized newspapers, offers a productive way to identify, cluster, and contextualize reprinted texts. However, as the few examples listed above have sought to illustrate, the output generated by text reuse detection is too vast (and unstructured) to be analyzed manually. Moreover, the output extends beyond literary works, but also includes entire pages, such as those with a multitude of stories about women. The current capabilities of computational text reuse detection only scratch the surface, and further research is needed to fully harness its potential on a larger scale. As demonstrated by the examples in this chapter, information about women is scattered throughout the digital archive. Yet, varying degrees of visibility pose a significant challenge in systematically uncovering and examining these stories comprehensively. “When every nineteenth-century newspaper brims with original and reprinted content of all kinds, it is difficult to know where even to begin studying that content” (Cordell, “Reprinting” 421–422). Another avenue worth exploring to continue “studying that content” is computational text classification. In the next chapter, I will provide an overview of machine learning methods in computational periodical research and explain how to employ computational classification to analyze newspaper texts. By doing so, the chapter demonstrates the potential of computational classification for studying different textual forms, regrouping text reuse clusters based on semantic properties, and locating women within the digital archive.