The other day our project got a panel accepted at the upcoming conference, The New and the Novel in the 19th Century/New Directions in 19th-Century Studies in Lincoln, Nebraska next year. What we aim to talk about is the following:
The ongoing digitization of historical newspapers makes it possible to explore nineteenth century novelties in new ways. However, digitized 19th century newspaper files should be seen as novelties as well. The process of digitization, OCR, article segmentation, modes of presentation (depending on GUI) etcetera are, thus, all infrastructural settings that transforms newspapers into new objects with a specificity different from the original paper prints. By using ideas from the emerging field of digital artifact studies, the aim of this panel is to examine and methodologically analyze the potential (im)possibilities of digitized newspapers, as well as the characteristics of these digital objects. If DH-inspired textual research has regularly focused on graphically clean fiction and ’white media’ (resulting in a strong signal), applying digital methods to 19th century newspapers is a messy business—mostly resulting in noise. Departing from word occurrences of ”the electric telegraph”, and the way it was presented to Swedish newspaper readers from the 1830s to the 1860s, we have extracted every text containing the words ‘electric’ and ‘telegraph’ from one digitized Swedish newspaper, Aftonbladet. To discover patterns and trends we have analysed word co-occurrences and word frequencies. Such an analysis reveals that many, many words are ‘neologisms’ invented by the computer and OCR software—resulting, in for example, numerous (mis)spellings of ”the electric telegraph”—and hence that most ‘articles’ have been re-edited by the segmentation software. Noise reduction of signals from the past hence literally becomes part of the historical investigation process.
Digitizing 19th Century Newspapers at the National Library of Sweden—On Media Specificity
During the 19th century journalists and newspaper editors frequently copied and pasted news items. These were inserted into and tailored to fit the specificity of the medium of the press—form, in short, affected content. The ongoing digitization of historical newspapers at for example the National Library of Sweden makes it possible to explore nineteenth century novelties in new ways. Yet, in a similar manner as during the 19th century digitization transforms content. The growing reliance on digital reproductions also raises questions regarding the function of such documents, especially the relation between newspaper source documents and consequent digital reproductions. In short, how reproductive is a digital reproduction? My presentation will depart from the ways in which 19th century newspapers have been digitized at the National Library of Sweden, with a focus on the actual transmission from the analogue to the digital domain. On a more theoretical level, the informative capacity of digital reproductions will hence be addressed. In fact, digitized 19th century newspaper files should be seen as novelties in a similar manner as the 19th century copied newspapers. The process of digitization, OCR, article segmentation, modes of presentation—depending on graphical user interface—are all infrastructural settings that transforms old newspapers into new media objects with a specificity quite different from the original paper prints.
Avoiding the lure of the novel by using digitized material
That the newspaper press underwent profound transformations in the mid-1800s is an established narrative in Swedish press history: circulations grew, reaching new strata of society; new voices were heard; the notion of the public opinion, first and foremost articulated by and through the press, was established; with inventions like the high-speed press and the electrical telegraph time and space were compressed; etcetera. Much research has hitherto focused on what is understood as new and modern: new genres, new technologies, a modern layout, innovators and pioneers. These analyzes often depend on a limited number of canonized texts, written by leading newspapermen and published in presumably dominant papers. Although not false the picture needs to be completed – not everything transformed. Firstly, this paper argues that one reason for the dominating focus on the new and the novel in the history of the press is academic disciplinary traditions favoring certain questions and methods, combined with the quality of the source material. Hermeneutical interpretations and close readings of ‘key texts’, representing only a fraction of the vast total output, have led to a rather idealistic and teleological narrative of modernization and progress. Secondly, the paper discusses what new possibilities digitized press material and broad search methods opens for a better understanding of continuity, inertia and slowness. New methods make it possible to rediscover residual patterns and the mundane character of the newspaper press, beyond canonized texts and dominant papers.
The Electric Telegraph in Digitized Newspapers
The ongoing digitization of historical newspapers makes it possible to explore nineteenth-century novelties in new ways. The aim of this paper is to examine the possibilities of the digitized newspapers, as well as the characteristics of these digital objects. My starting point is the electric telegraph and the way it was presented to Swedish newspaper readers from the 1830s to the 1860s. How was this invention described? In which contexts was the electric telegraph put in the articles? What are these articles really about? To answer these questions I have extracted every text containing the words ‘electric’ and ‘telegraph’ from one digitized Swedish newspaper, Aftonbladet. With immediate access to the text files, without being dependent on the interface provided by the National Library, it is possible to analyse the texts using digital methods. To discover patterns and trends I have analysed word co-occurrences and word frequencies. To get a better understanding of the data which makes this kind of research possible I have combined the methods of distant reading with a close reading of the digital format. Such an analysis reveals that several words are actually ‘neologisms’ invented by the OCR, and that most ‘articles’ are re-edited by the segmentation software. Distant reading is still possible, but researchers must pay close attention to the noise in the channel – just as the users of the early telegraph.