On December 18, 1889, the Detroit Free Press asked an intriguing question in the headline: ‘Will It Come Here?’Footnote ¹ The ‘It’ in the headline was the so-called ‘Russian Flu’, an outbreak of influenza that was first noticed on a global scale in St Petersburg, the capital of Russia.Footnote ² As this disease spread across Europe, American medical authorities as well as the popular press expressed increasing concern about whether the disease would cross the Atlantic and reach the United States.Footnote ³ To answer this question, newspaper reporters all over the country, as in this Detroit Free Press article, turned to local authorities, doctors and physicians from their cities, who were asked to share their opinion about influenza. In this same period, many medical journals published editorials commenting on the spread of the disease, the likely course of further development and the means for preventing or treating outbreaks. This paper examines comments by local doctors published in the Detroit Free Press and one editorial published in Medical Age, a semi-monthly journal also published in Detroit, as a way to introduce a broader research project engaged in three innovative methods that link the digital humanities and medical history: comparing medical reporting in daily newspapers and medical journals to explore connections across local, regional, national and international levels; combining traditional historical methods of close reading with new tools available for large-scale textual analysis; and developing methods that engage undergraduate students as research collaborators at every stage of the process.

A close reading of the Detroit Free Press article of December 18 cited above reveals the complex ways in which information about the Russian flu came to doctors, their processes for making sense of diverse information sources and their role in interpreting that information for public dissemination. The first doctor quoted in the article, Dr Donald Maclean, stated that the disease could be communicated from one person to another, and thus it might spread to Detroit, and that, while influenza was rarely fatal, it should not be disregarded. Dr T.A. McGraw was ‘somewhat skeptical’ about reports of influenza in New York City, because whenever many cases are reported in newspapers, people assume that all victims suffer from the same disease. McGraw challenged the theory that influenza microbes were blown from place to place on the grounds that ‘the general direction of the wind is from west to east, while the influenza comes here from the western world’. Instead of being carried by the wind, McGraw declared, ‘People bring it in their clothes.’ Regardless of how the disease was transmitted, according to McGraw, Detroiters should not ‘fear any epidemic of this sort’, but should ‘treat it carefully’, as even though it was ‘seldom fatal’, it was dangerous to ‘aged and sickly people’ and can even give ‘a strong man a pretty severe attack’. Finally, Detroit physician Dr B.P. Brodie had not heard of any cases and, although he conceded that the disease ‘might’ break out in Detroit, he too reassured the public: ‘influenza is, at the worst, but a very troublesome and disagreeable disease’. Repeating the claim that influenza ‘very rarely proves fatal’, the newspaper did affirm that the disease ‘requires constant attention and treatment’.

Just one week later, on December 26, 1889, the Medical Age published an editorial under the title, ‘The Prevailing Influenza’, which provided a general account of the disease, its previous historical manifestations and specific information about its origins, causes and symptoms.Footnote ⁴ The first sentence set out both the scope of the disease and the sources of information available to the journal editors: ‘That this affection is at present epidemic and very generally distributed over the East and West is apparent from the cases reported in the medical journal and the interest manifested by the daily papers.’ This opening paragraph indicates that ‘the prevailing influenza’ is a familiar disease yet is also manifesting special traits and conditions. The remainder of this article, which includes a substantial amount of text reprinted from an editorial in the Medical Record, published in New York City, provided a more detailed account of the disease’s spread across Europe, some historical background on previous influenza epidemics, the likely means of transmission, a list of symptoms, recommendations for treatment and reassurance that the disease should not cause serious consequences for Detroit or the United States as a whole.Footnote ⁵

In the course of one week, from December 18 to December 26, these two articles provided Detroiters with a sequence of authoritative statements about the prevailing influenza. A reader who looked at these two articles would probably come away with a general impression that included the following elements: that the current outbreak of influenza was more widespread than usual, although the reasons for the speed and scope of the outbreak were not known; that the disease was spread by microbes, although the path of transmission was not readily identified; and, that, although Detroit had only a few reported cases and more were expected, the disease was not likely to cause many deaths and thus was not seen by experts as a serious threat to public health. The fact that doctors and the Medical Age editorial attributed much of the attention to the disease to the influence of newspapers is suggestive of a self-consciousness about the complicated relationship between disease outbreaks and information transmission.

To explore the relationship between information and disease on a larger scale, historians can make use of online tools for text visualisation. The Medical Heritage Library and the Internet Archive make it possible to save the entire text of periodicals such as Medical Age. A ten-year run of this journal, including the five years before and five years after the Russian flu, amounts to more than five thousand pages and close to five million words, a quantity of text that would impose a great burden on an individual scholar or even a team attempting a close reading. One way to address this research challenge is through a method, such as the online Voyant tool, which quickly generates a variety of visualisations, tables and other outcomes.

Of these many options, the collocation tool is an easy way to reveal which terms appear in relative proximity to each other. By identifying terms which appear in the same phrases or sentences, this tool suggests more meaningful connections than just the term frequency revealed in a word cloud. Using two keywords related to the Russian flu, influenza and epidemic, indicates that these terms are collocated with each other approximately twenty times in the ten years of this journal. While these twenty instances make up a tiny fraction of the entire corpus (0.0008%), a comparison to other disease terms that collocate with epidemic is suggestive of the relative importance of influenza as an epidemic disease. For this purpose, a network mapping tool such as Palladio, available online from Stanford University, reveals terms that connect to both influenza and epidemic as well as those that are linked to only one term or other (Illustration 1). The terms that are linked to both words are suggestive of this historical event; these include diagnostic terms, such as fever, cough and disease, as well as terms indicating the nature of the disease, such as severe and prevalence. By contrast, epidemic is linked to many other diseases, including cholera, typhus and smallpox, while influenza is collocated with pneumonia, grippe and bacillus, all of which reflected efforts to define the medical significance of this disease. Perhaps most interestingly for this project, influenza is collocated with the word editorial seven times over this ten-year period, a pattern suggestive of the important relationship between the spread of the Russian flu and coverage of this epidemic in journals and newspapers.

Illustration 1: Collocation Map, Medical Age, 1886–95.

Collocation is an example of a digital humanities approach suggestive of links and connections that require further research and analysis. The close reading of texts, as modelled in this first part of this article, is an important step for understanding how physicians, newspapers and, indeed, the public understood this epidemic in its earliest stages. Yet text visualisation tools also have great potential to identify important themes, to suggest connections and to identify possible relationships. Researchers can pursue this research on their own using a combination of full text sources in the Medical Heritage Library, text visualisations in Voyant tools and network analysis in Palladio. For students working on medical research projects, these tools combine relative ease of access and applications with the possibility of increasingly sophisticated analytical strategies that yield new insights. This approach allows scholars as well as students to appreciate the value of a digital humanities approach as well as the importance of close-reading skills to explore more fully the historical significance of an event such as an influenza epidemic.