Initial public offerings (IPOs) in the dot-com world do not always turn out to be the darlings they are expected to be. Ask Mark Zuckerberg about Facebook's IPO in May 2012. But even successful new ventures often defy their founders' expectations. As we hope to suggest in the following report, German History in Documents and Images (GHDI)—a project that has put thousands of primary source texts, drawings, photographs, and maps on the internet, along with hundreds of pages of accompanying commentary—has drawn critical appreciation from specialists and nonspecialists alike, but it has also raised thorny questions about authorship, authority, and audience.Footnote ¹ Those questions concern the writing of history in general and the newer, more specific discipline of “history on the web.” Like the project itself, this report is the result of a collaboration among the GHDI project staff, which is based at the German Historical Institute (GHI), Washington, D.C., and the GHDI volume editors, all of whom teach (or taught) German history at colleges and universities in North America. In the following pages, we will discuss the origins and early goals of the project, describe the challenges associated with the realization of a large, collaborative history project of this nature—whether in book or digital form—and reflect upon what we perceive as the promise and perils of digital history anthologies.Footnote ²

*    *   *

The origins of the project date back to 2002, a time not terribly far removed from us—especially when the subject is history and the writing of it. But when viewed in “internet years,” 2002 already seems like another lifetime, a moment when the web, as we know it, was relatively new. The first web pages appeared in the early 1990s, but it was only after the introduction of Netscape, in 1994, that “the History Web came into its own.”Footnote ³ According to Daniel J. Cohen and Roy Rosenzweig, the mid-1990s marked the start of a “fever to bring the primary sources of the past online.”Footnote ⁴ It was a fever shared by scholars and teachers, but also by amateur enthusiasts with specific interests—sometimes vested interests—in particular historical topics.

German History in Documents and Images was both part of this fever and a response to it. By 2002, the German Historical Institute and a group of scholars associated with it had already seen the potential of primary source websites to enliven the teaching of German history, but they had glimpsed some attendant dangers as well.Footnote ⁵ They were inspired by early teaching archives, such as Paul Halsall's “Internet History Sourcebooks Project,” which gave students greater access to quality historical texts in translation, filling a void felt keenly by educators; yet they were concerned by the proliferation of other, less academic websites, including ones devoted to military history.Footnote ⁶ Such websites sometimes reflected an undue fascination with the Nazi war machine. Moreover, the initiators of GHDI were particularly dismayed to see that student term papers were making increasing—and wholly uncritical—use of homegrown websites that advocated particular agendas, well-intentioned or not.

In spring 2002, Christof Mauch, then director of the GHI, consulted with two board members of the Friends of the GHI, Roger Chickering and Konrad Jarausch. Together, they forged plans to develop a project that would make a large selection of German historical documents available in both German and English on the website of the GHI in Washington, D.C. Their goal was twofold. First, and most generally, they wanted to take the lead, as one of them recalled, in putting “responsible historical content” on the web. Second, as they explained in a broadly sketched grant application, they aimed to use primary source documents “to revitalize the study of German history and culture in the United States and beyond.”

By the end of 2002, funding had been secured from the Max Kade Foundation and the ZEIT-Stiftung Ebelin und Gerd Bucerius, and the project's steering committee was officially in place. In addition to Mauch, Chickering, and Jarausch, the committee included Dirk Schumann, who had recently joined the GHI as deputy director, and Gerald Feldman, who, like Chickering and Jarausch, was a board member of the Friends of the GHI. In 2003, work began in earnest, and the committee started to define the parameters of the project. One of the first questions they addressed was the project's chronological scope. The members of the steering committee were, without exception, historians of the late-nineteenth and twentieth centuries. They appreciated that however dire the shortage of primary source materials in their own field, it paled in comparison to what historians of early-modern and eighteenth-century Germany faced in their teaching.Footnote ⁷ Realizing that the project needed to be far-reaching in scope, they set the time frame as 1500 to “the present.”Footnote ⁸

They then divided this time span of more than 500 years into ten chronological periods, which form the bases of the project's ten volumes. The committee decided that each volume would have a separate editor (or team of editors), that each would include documents in both English and German and a selection of images, and that each editor would write a brief but comprehensive introduction to the history of the period in question. To prevent an overemphasis on political history and to ensure that more recent topics of historical inquiry, such as gender and popular culture, were well represented in the project, the committee devised a list of subject headings for the editors to consider in selecting their materials.Footnote ⁹ This decision was made with an eye toward enhancing the pedagogical value of the site. “After all,” as one committee member observed, the goal of the website was “to provide sources for classroom purposes as well as historical research.”Footnote ¹⁰

With these guidelines in place, the GHDI steering committee began the first phase of its work and set about finding editors for the volumes. To start, three committee members (Chickering, Feldman, and Jarausch) agreed to sign up as editors themselves: Chickering as coeditor (together with Steven Chase Gummer) of volume 5 on Wilhelmine Germany (1890–1918); Feldman as editor of volume 6 on Weimar Germany (after Feldman's untimely death in 2007, Eric D. Weitz and Eric S. Roubinek replaced him in this role); and Jarausch as coeditor (together with Helga A. Welsh) of volumes 9 and 10, which cover the years 1961–1989 and 1989–2009, respectively.Footnote ¹¹

As the steering committee assembled the editorial team, the GHDI project staff started thinking about the website that would display all of the materials. From the start, there were two core requirements. First and foremost, the site would endorse the principle of open access. Second, the site would be self-contained. In other words, it would not link to content—primary source texts, images, commentary—on other websites. The staff then began meeting with web designers, who posed questions such as, “Which three adjectives would best describe your ideal website?” Words such as “transparent,” “straightforward,” and “trustworthy” came up again and again. But one word, whether explicitly mentioned or not, informed the design of the site like no other: “book-like.” Although the project was “born digital,” it was never a web project per se so much as a primary-source reader in digital form—a point that becomes clearer as the site gets older. But reports of “the death of the book” have been greatly exaggerated. Many users of German History in Documents and Images appreciate its straightforward division into volumes, which are then divided into chapters, and so on. In short, they like its familiar feel.

One related question that preoccupied the GHDI staff early on was the way in which primary-source texts would appear on the site. Would readers have to scroll through one long text, or would they navigate by “turning” pages? This question sparked larger discussions about readers' expectations and the new medium of “writing for the web,” whose postulates—brevity, clarity, and “scanability”—seemed incompatible with the challenging and recondite texts that GHDI planned to offer. Therefore, to make long and difficult texts more appealing and less overwhelming to readers (although there are limits to what can be done in any medium), the staff opted to break texts down into manageable pages and, ultimately, to add a print function.Footnote ¹²

Images occupied an important place in the project from the outset. The emphasis on images was a reflection of the steering committee's appreciation of the internet's visual nature. But it was also an acknowledgment of the increasingly important role of visual studies in general and an attempt to reach a younger generation of students. As one committee member put it in 2003, the project had to be “more now.” Lastly, the visual emphasis was also an effort to expand the corpus of free, high-resolution digital images available to teachers of German history. The committee believed that the selection of historical images on the web was too limited and too homogeneous—too many reproductions of the same famous scenes that could already be found in countless textbooks.

That the images were to be understood not merely as “illustrations” or accompaniments to the documents, but rather as historical sources on a par with them, was made clear from the start by the name of the project and the structure of the website. The committee's goal of making a vast selection of visual imagery available—legally—to a worldwide audience started to become a reality when the German Historical Institute entered into a crucial partnership with the Bildarchiv Preussischer Kulturbesitz, which has supplied the project with more than 1,000 images from its collection.Footnote ¹³

Whereas German History in Documents and Images was principally envisioned as a primary-source website, the steering committee recognized the need to include explanatory material as well. That each editor would write an introduction to his or her volume was clear from the beginning. But over time other forms of editorial guidance began to assume greater importance, and, for better or worse, greater length. This material included abstracts that introduced each document, captions for the images, and, less frequently, footnotes that explained obscure people, places, or events in the texts. The decision to include all of this editorial material in German as well as English represented a major turning point for the project. At that moment, GHDI went from being an English-language website that included primary-source documents in both the original German and English translation to a completely bilingual website where every page appeared in parallel form in two languages. This decision—we have no doubt it was a good one—had far-reaching implications, not least with respect to the project's target audience.

By November 2003, the entire GHDI editorial team was in place, and the volume editors were invited to Washington, D.C., to meet with the GHDI steering committee and project staff. The agenda included a range of topics: the basic parameters of the project and its component volumes, practical matters such as deadlines and document delivery, and administrative concerns (such as copyright procurement and publicity). Tellingly, it was not these issues, but rather larger ones regarding periodization, categorization, interpretation, and uniformity, that dominated the conversation that Saturday afternoon. The chronological division of the volumes, especially those devoted to Germany after 1945, was a point of particular interest. Some participants in the meeting argued in favor of two postwar volumes (1945–1968 and 1968 to “the present”) rather than the proposed three (1945–1961, 1961–1989, and 1989 to “the present”). This suggestion responded to various concerns. There was a general sense, for example, that 1968 might have represented a greater break than 1961. But some participants expressed worries that volume 10, which covered 1989 to “the present” (which was then 2003), was too short, meaning that the events of these fourteen years would be given too much weight in the project.Footnote ¹⁴ In the end, other voices cautioned against reorganizing the project's chronology. Konrad Jarausch and Helga Welsh, coeditors of volume 9 (1961–1989), made clear that 1968 would be given its due as a watershed year. Others pointed out that, over time, volume 10 might be extended—which has indeed been the case.

The liveliest discussion was sparked by the question of categorization. Did the editors want to adhere to the steering committee's proposed categories? Did they prefer to come up with their own? And did any uniform categorization suggest that the editors were overstepping their epistemological and pedagogical mandate? Some editors argued that the very act of dividing the materials into categories put too much weight on the editorial commentaries, insofar as editors would be compelled to explain why a document or image was assigned to one category and not another. Others emphasized the project's role as a teaching aid: they argued that grouping the documents and images into categories was best left to classroom instructors. In the end, it seemed clear that there was a practical need to organize the materials in some way, even if doing so imposed a level of interpretation unwanted by some. As a concession, the editors were granted ample leeway in the organization of their volumes. By the end of the meeting, volume-to-volume uniformity had emerged as a less achievable (and less desirable) goal than it had been at the start. Although that Saturday discussion raised more questions than it answered, notes from the meeting sounded a hopeful tone: “Some of these questions will surely be solved as work proceeds and as themes emerge.”Footnote ¹⁵ At the end of the meeting, the volume editors, the GHDI steering committee, and the project staff promised to meet again as a group—but ultimately never did.

The conversation did continue on a smaller scale, however, as the work progressed. In late 2009 and early 2010, two conference panels were dedicated to German History in Documents and Images.Footnote ¹⁶ After these two panels, we, the authors of this article, put together a short survey and sent it to each GHDI editor. While our survey was no substitute for that second meeting that never took place, the editors' thoughtful responses did represent a continuation of that conversation. The questions we posed were: How and why did you select certain documents and images over others? Did you try to “tell a story” as you shaped your volume? Have you used this resource in your own teaching? Why did you choose to organize your volume as you did? Did you try to strike a balance between social, economic, political, cultural, and international history? The editors' replies offered a host of wide-ranging reflections on the novelty and scope of the enterprise and on the broader questions it raised about “doing” history in the twenty-first century. How do historians imagine they are using these documents and images to communicate with other scholarly researchers, with students in classrooms, and with the lay public in far-flung corners of the globe? And how are they actually communicating with these groups?

*   *   *

The replies we received to our questions cannot be brought unter einen Hut. No surprise there: the editors' teaching institutions, career paths, and research agendas provided no common denominator. One note of consistency was nonetheless apparent. In selecting the texts and images in his or her volume, each editor underscored the intention to mix well-known and unfamiliar materials to fire the imagination of readers and viewers. In this regard, the editors sought to find new images even more than they tried to present noncanonical texts. Helga Welsh, coeditor of volumes 9 and 10, put it this way: “The images should make the reader curious to learn more about particular topics and to help visualize broader themes; familiar images were complemented by less accessible ones.” Welsh's statement is borne out by the images that she and Konrad Jarausch chose for volume 9. These include iconic images, such as the photograph of Peter Fechter dying at the Berlin Wall, along with less familiar but provocative scenes of generational rebellion.

William W. Hagen, the editor of volume 2 (1648–1815), articulated a methodological note sounded by others, namely that unfamiliar images should, like unfamiliar texts, work against the grain of German historiography as presented in older textbooks or on the History Channel. Hagen wanted to present the “immediate” and “inartful” experiences of a range of historical actors. His volume, like all the others, had to provide “coverage” of the high (and low) points of German history, lest the selection become willfully idiosyncratic. At the same time, Hagen was aware that, for the early-modern period in particular, readers and viewers would be rewarded if the editor shone a light on less-well-researched areas of German life. The editors whose volumes covered more contemporary periods faced a similar challenge. Eric Weitz observed that the “highly contested nature of everything during the Weimar period” made it important to “get many voices” into the collection. Insiders and outsiders in the Weimar Republic had to be found in unfamiliar places: in foreign relations, the world of learning, and many areas of politics and economics that have been neglected in prior treatments of Germany's “failed” experiment in democracy.

This quest for diversity reflected scholarly, pedagogical, and pragmatic motives among the team of editors. Richard Breitman, the editor of volume 7 on Nazi Germany (1933–1945), was intent to appeal to many different types of audiences, some of which came to his volume with the images and slogans invented or perpetrated by the Nazis themselves. A different challenge faced Jarausch and Welsh in their two volumes. A general familiarity with the canonical texts of German history since 1961 could be assumed among older readers, whereas images were especially important for two reasons: the thirty-year rule for releasing German archival documents and the ubiquity of newspaper, television, and film images from these years. Jarausch and Welsh thus sought out nonofficial documents and made a virtue of easy access to memoirs, newspapers, and magazines. Unfamiliar sources were also chosen, wrote Jarausch, because they are “more experiential and communicate a feeling for a certain situation” in a way that treaties and speeches cannot.

The same approach characterized volume 8 (1945–1961), which was coedited by Volker Berghahn and Uta Poiger. They paid tribute to recent scholarship that had led them to important documents about under-researched topics not yet anthologized in any language—for example, returning German prisoners of war, the “Americanization” of German society, and “mixed-race” children in postwar Germany. For Berghahn and Poiger, collaboration was crucial to the success of their volume. The two settled upon a division of labor that allowed them to draw on their expertise in particular subfields: Berghahn in politics and economics; Poiger in culture, society, and gender. Jarausch and Welsh also made use of the collaborative approach—to similarly good effect—in volumes 9 and 10. That some of the editors, namely Berghahn and Welsh, have either a strong interest or an actual background in political science, helps to explain why the social-science approach was well represented in the project.

*    *   *

Whereas none of the editors presumed to offer a master narrative even for their particular volume, they provided a number of reasons for trying to tell a coherent story. More than one editor was working on a general account of his or her time period while serving as editor of a GHDI volume. This was true of Thomas A. Brady, Jr., coeditor (together with Ellen Yutzy Glebe) of volume 1 (1500–1649). His German Histories in the Age of Reformations, 1400–1650 appeared in 2009, at the same time that he was assembling his selection of documents and images. In fact, Brady, over a long career, had already taught with many of the texts in his volume, and he knew which ones “worked” with students. Jonathan Sperber, James Retallack, and Roger Chickering were also working on related books while editing volumes 3, 4, and 5 (which cover the Vormärz, Bismarckian, and Wilhelmine periods, respectively). By the time the GHDI editors' meeting took place in November 2003, Sperber had already spent years grappling with revolutionary and post-Napoleonic Europe and with Vormärz Germany: first as author of the textbook Revolutionary Europe, 1780–1850 (2000), second as editor of Germany 1800–1870: The Short Oxford History of Germany (2004). Moreover, he had also gotten an early start on his GHDI volume and had already chosen the documents that would appear in it. So he was able to contribute crucial practical experience to the project at an early stage. Retallack's “take” on Bismarckian Germany in his GHDI volume reflected the themes he emphasized as editor of Imperial Germany 1871–1918 (also in the Short Oxford History of Germany series, 2008). And Chickering remarked that his organizational schema for volume 5 reflected the same conclusions he had reached in writing a “total history” of World War I in The Great War and Urban Life in Germany: Freiburg 1914–1918 (2007). The “whole story” of Wilhelmine Germany, including the origins of the war, was to be presented in “the most effective and elegant way” possible. Thus, in both works, Chickering stressed the “reactions to the great social dislocations and dynamism of the high-industrial era” and the politics of mobilization. The balance between these aspects of Wilhelmine history complicated Chickering's task because he “had two stories to tell.” Therefore, in his GHDI volume, he changed course between chapters one through six, which culminated in the outbreak of war, and chapters seven through ten, in which military events and the materials used in his monograph necessarily figured more prominently. Lastly, Eric Weitz structured volume 6 along the lines of his survey, Weimar Germany: Promise and Tragedy (2007). Weitz wanted to emphasize that although every move to realize the promise of Weimar was, “needless to say, deeply contested,” he refused to read the ultimate triumph of the National Socialists back into the fourteen years of the republic.

Even more important than books the editors were writing at the time was the influence of their teaching on the shape of their volumes. Like Brady and Glebe, most editors were conscious of how their volume might be used in curricula offered in North American universities versus those in Germany, Great Britain, and elsewhere. Nowhere is teaching with documents particularly easy, but Konrad Jarausch noted that his American students had to be weaned off “predigested pablum.” They had to be pointed toward key phrases and subtexts in the primary sources. He and others stressed the importance of providing historical context for these sources but not feeding students the answers too quickly. Instead, they should be encouraged to research more deeply the historical contexts in which these documents and images are embedded.

Like all historians, the editors wanted their narratives to show both continuity and change. Most also emphasized, and proved in practice, that students should be exposed to the varieties of behavior in the past. In this regard, Hagen's reflections were programmatic. In dealing with the period “From Absolutism to Napoleon,” one had to offset “the accustomed and by now not-so-interesting representations of bewigged and bejeweled baroque elites and the worlds they moved in.” Equally important, he continued, was the need to avoid “clichéd representations of the common people, in the patronizing style of contemporary bourgeois art.” In other words, the “positively valorized liberal Whig account” of Germany's (and other nations') transition to “a modern bourgeois-liberal/democratic-rationalist-capitalist-industrial society” needed redress. Reflecting conclusions he had reached in Ordinary Prussians: Brandenburg Junkers and Villagers, 1500–1840 (2002), Hagen sought out documents and images that emphasized the “heightened articulation of self among women and ordinary people,” changes in rural society, and the many paths of capitalist development.

Once the sources were selected, translations from German into English presented another challenge. Hagen, for example, wanted translations “that avoided replication of eighteenth- or nineteenth-century upper-class English, or other forms of archaicism.” In this respect, the “stories” told in these ten volumes benefited in crucial ways from the talented team of translators that was mustered for this project. This was certainly true of volume 1 (1500–1648), the only volume whose editors did much of the translating themselves.Footnote ¹⁷ Like Hagen, Brady and Glebe sought to include the voices of women and ordinary people. Their first section, “Eyewitnesses,” includes accounts by a lady-in-waiting, an Augsburg burgher, a goatherd turned teacher of Greek, and a Swabian cobbler-farmer—ordinary people who emerge as anything but.

Reflecting on the organizational choices he made, Jonathan Sperber reported that he was eager to have his readers do what he asks of his own students: “investigate several different sides of a question, then compare and evaluate them.” Sperber's section on the emancipation of the Jews, for example, offers Jewish, Gentile, and state perspectives on particular topics. Likewise, his section on economics offers alternative views, pro and con, on the impact of free markets in land, labor, and capital. This technique was replicated in Retallack's volume 4 (1866–1890), which presents contrasting views of “provincial” and “national” Germany and asks readers to consider how those views were accepted or resisted by certain groups of Germans before and after unification in 1871. In a later period it was also possible to emphasize to students the multifaceted nature of the available material, even while acknowledging that contemporaries often faced binary choices. Volker Berghahn and Uta Poiger noted that such binaries included the two Germanies in the aftermath of National Socialism and war, and the impact of Cold-War divisions on the place of Germany in the world.

To a certain degree, Sperber's approach suggested that any narrative that might emerge from a GHDI volume, or a discussion based upon it, would be inherently Socratic. But the editors did not want to not stop there. Helga Welsh observed that in volume 9 (1961–1989), she and Konrad Jarausch had to write a history of two countries that would not marginalize East German history. Therefore, rather than emphasizing the “success story” of West Germany, they asked students to consider shared experiences during the period of transition and consolidation of the two states, as well as the challenges they shared, such as adjusting to a changing international order, the energy crisis, and postindustrial relations. Richard Breitman was conscious of the limited ability of undergraduate students to absorb a great deal of information at one time. Like others, he published the most compelling excerpts from documents that were sometimes dozens or even hundreds of pages long. Breitman conceded that the rubrics laid down for the editors “precluded any single or simple story.” Within each section of his volume, however, he did try to construct a narrative (a “story within a story,” as Welsh put it), albeit with due attention to continuity, change, and variations of behavior and experience. As he observed, the balance adopted by any given editor is always a matter of judgment. Conformity across ten volumes, however, was never imagined as attainable, or even desirable, either at the outset of the project or thereafter.

*    *   *

The preceding comments have chronicled some of the lessons learned in the course of this project, just as they have identified open questions. At the start of the project, few participants anticipated the breadth or diversity of the global audiences who now use these documents and images on a daily basis. Web logs monitored by the GHDI project staff report the number of “unique visitors” to the website each day. But they tell us much more, too: the visitors' countries of origin, for example, and the websites, search engines, and keywords that referred them to the site. Much of this data confirms the original expectations of the GHDI steering committee and staff. For instance, the fact that the site registers approximately twenty percent more traffic during the North American academic year than during the summer months seems to suggest that GHDI is reaching its original target audience of undergraduate students and faculty. More surprising is the extent to which K-12 teachers are also making use of the website.Footnote ¹⁸ This information comes not only from web logs, which record visitor IP addresses that are registered to public school systems, but also from the many emails written to the project staff by secondary school teachers in North America and Europe.

At the editors' meeting in 2003, there was a general assumption that the site “would be used in English mostly.” The record has been more complicated. Germany and the United States always occupy the top two spots on the list of referring countries, but the first place position alternates between them. On average, Google.de seems to drive more traffic to the website than Google.com, and the list of referring URLs is almost always topped by http://www.google.de/images. The popularity of the website among Germans validates the steering committee's decision to create a totally bilingual website. But while this move was beneficial in expanding the project's audience, it also complicated matters—and not just from the perspective of cost or labor. Putting all of the content in German meant that the site was, in effect, pitching itself to a German (or at least a German-speaking) audience. This represents something different from offering content in English that can also be read by Germans who know the language. For the most part, however, the site was not geared toward German readers: most of the didactic texts, image captions, and other editorial aids had been written in English, with North American students in mind, and then simply translated into German.Footnote ¹⁹ In some instances, this approach proved problematic. For example, compared to a North American reader, a German brings a different background and perspective to an image such as Willy Brandt's Warschauer Kniefall. A “one-size-fits-all” caption is impossible to write.

Questions involving language, culture, and target audience do not end here. Web logs and user email suggest that GHDI is also being used by people whose native language is neither English nor German. University instructors in the Netherlands, Sweden, and Denmark, for example, are using the website in courses in German history and German literature, making it harder to gauge our audience. This issue is further complicated by the fact that excerpts from GHDI volumes are being reprinted, in English, in bilingual history textbooks for students in German Gymnasien.

One thing is clear, however. The popularity of the images is overwhelming, particularly among instructors. University faculty write enthusiastically about “popping” images into PowerPoint presentations “twenty minutes before lectures.” A cursory survey of online syllabi from North American and British universities confirms the ease with which images are captured from the GHDI website and incorporated into other teaching tools. Less clear is the extent to which the editors' introductory essays, document abstracts, and image captions are serviceable to frontline teachers. One line of argument is that the editors put so much effort into writing detailed abstracts that they provided a disincentive to teachers and students to read the documents or interpret the images on their own terms. In other words, they provided too much context. Some historians in the field have suggested that the abstracts and captions are good but perhaps unnecessary, since the “documents and images speak for themselves.” Others suggest that such an assumption is naïve—perhaps dangerously so. Here, it is also important to note that the website is used not only by historians of Germany, but also by nonspecialists who have to cover German history within the context of large European history and Western Civilization survey courses. Many of these nonspecialists seem grateful for a point of orientation.Footnote ²⁰ One hopes that individual instructors will choose carefully how much of this editorial material to serve up to their students. In the end, the project aims to present new materials without constraining anyone in their use.

On a different front, GHDI has given graduate students an important research tool. Those with the requisite language skills can read the texts in the original German, while others can rely on the translations. Some editors indicated that they were initially more comfortable using the primary sources in graduate seminars, less so in undergraduate tutorials. Others have taken the opposite stance, revising undergraduate lecture courses (where tutorials are not offered) to incorporate brief discussions of relevant documents to break up the lecture. It remains to be seen how quickly or deeply these new primary sources will be embedded into cutting-edge scholarship by advanced researchers. URLs and active hyperlinks to these sources are appearing in the footnotes and bibliographies of standard works, though less frequently than some editors anticipated. In the future, it may prove easier than it is today to incorporate active hyperlinks into the increasingly prominent electronic versions of scholarly monographs.

Roger Chickering has offered the most provocative reflections (so far) about the way these sources are actually being consumed. He has noted that some of the editors—most conspicuously Berghahn and Poiger—were willing, perhaps even eager, to liberate their interpretive, introductory essay from the documents, meaning that their essay makes no explicit reference to the documents included in their volume. Others took the opposite approach, either mentioning individual documents and images by name or providing links to them. Jarausch and Welsh opted for a third path: like Berghahn and Poiger, they authored a stand-alone general introduction that makes no reference to the documents, but they also wrote shorter introductions (or “overviews”) to each of the subsections included in their two volumes. These overviews make explicit reference to the chosen documents, and readers thus have the option of viewing the primary-source materials through a particular interpretive lens.

Given these different editorial choices, Chickering's question is germane: how much does this interpretive guidance matter? On the one hand, he suggests, Berghahn and Poiger were simply being realistic in assuming that users are happy to sever the editors' interpretive essay from the body of documents. This approach is consistent with the nature of the internet and the way it is used by older and younger generations alike. “The hypertext format frees documents from their narrative moorings and whatever interpretive claims the editors may seek to make. Readers are free to establish their own narratives with the click of a mouse, as the hyperlink subverts editorial privilege . . . Readers are now free to pick and choose in a vast sea of documents. The stories are theirs to construct.” Here, one could add that the pilots (teachers) who navigate this “vast sea” are also free to embrace nonlinear modes of instruction. The technical equivalent would be the internet “node” from which an instructor could set off in any direction, choosing hyperlinks that correspond to the particular needs and dynamic of a unique classroom situation. The by-now-“traditional” PowerPoint presentation, in other words, might become the dodo bird of twenty-first-century instruction thanks to the GHDI and other digital history anthologies.

Chickering's observations, on the other hand, did not dismiss the possible survival of the narrative. He noted that established interpretive claims, which persist in the GHDI collection, may prove more robust than we imagine. The editors have been far from powerless. “The documentary sea may be vast, but the editors have created it.” Moreover, the hyperlink itself possesses narrative powers that cannot be ignored. Mindful of these powers, the steering committee originally envisioned the site as “a network of links” that would allow users to “pull together information that is closely linked thematically but appears in different volumes.”Footnote ²¹ And today there are a number of links that do exactly that. One good example is the caption to the image “Jutta Limbach Opens the Exhibition ‘Fifty Years of the Basic Law’” (April 29, 1999), which appears in volume 10 (1989–2009). The caption includes several links to materials in preceding volumes, including excerpts from the Basic Law (1949/1956), a photograph of the Parliamentary Council (1948), a newspaper article on the Spiegel Affair, and a photograph of students protesting against the Emergency Laws (1968).

While the hyperlink has the power to “pull” information together, it can also create narratives, intended or not. Consider the caption to the photograph “Krupp Smokestacks in Essen” (1867), the first of many references to Krupp in the project. The photograph appears in Retallack's volume on Bismarckian Germany, where it is grouped with other industrial scenes. The caption includes two links to materials in the same volume—to a letter from Alfred Krupp in 1867 “on the charm of belching smokestacks” and a speech to his employees in 1877. The speech, an example of Herr-im-Haus paternalism, is then linked to a woodcut of the Krupp steel factory in Essen in 1890. The hyperlink narrative ends here, but it did not have to. Retallack's caption to the 1890 scene mentions Krupp's role as a producer of Prussian artillery, offering an easy invitation to jump forward to a 1910 Krupp factory scene in the Wilhelmine volume, where the theme of militarism and armaments production emerges more clearly. And from there, it would have been just as easy to insert a link to a photograph of a grenade exploding in Verdun, then to a photograph of members of the Nazi military-industrial complex, and finally to several pictures of the demolished Krupp factory in Essen in 1947.

This story of industrialism and militarism, which follows a narrative arc, is just one of the hyperlink narratives that could, by design or not, have found its origins in “Krupp Smokestacks in Essen.”Footnote ²² A different and shorter narrative could have emerged as well. Retallack's caption, which interprets the scene through the lens of the environment, provides ready-made opportunities for links to any number of materials in Jarausch and Welsh's section on “The Green Movement” (volume 9, 1961–1989). By this route, the reader would have bypassed two world wars and landed squarely in the ecological movement of the 1970s and 1980s.

As these examples suggest, questions of narrative and authority in historical writing, which are difficult in any medium, are further complicated by the internet. Is it a good thing that the caption to a portrait of Heinrich Heine (volume 3, 1815–1866) mentions that his books were banned and burned by the Nazis in 1933? Would it be dangerous—would we implicitly endorse the dubious proposition that Germans embraced an “eliminationist” antisemitic mentality even before World War I—if we chose to link Heinrich von Treitschke's infamous pronouncement from 1879, “Die Juden sind unser Unglück!” (volume 4, 1866–1890), to the NSDAP mass rally in Berlin's Sportpalast on August 15, 1935 (volume 7, 1933–1945), where the same phrase was emblazoned on a huge banner in the hall?Footnote ²³

Rather than attempting to answer these questions, the editors and the project staff continue to interrogate their own intentions and explore the epistemological ramifications of this resource. If this report encourages further reflection about the organizational, interpretive, and pedagogical issues raised in a project of this scope, and if more scholars prove willing to navigate the same uncharted seas we have, the teaching of German history and its relevance to global audiences will become more interesting and important objects of research in their own right.