Tags

, , , , , , ,

Introduction

Historical texts describe events and characters in the past, while literary texts describe those in fictional worlds. Understanding the places that these texts refer to, and changes that occurred there over time can be beneficial in many ways. Traditionally this is being done by historians and scholars without using any particular digital technique. However, it is seen that certain specific technologies can make their task easier, and also allow them to extract information that was unavailable before.

Distant Reading and other Techniques

Distant reading is a technique that allows one to extract useful information from a large body of texts, without closely reading any individual text. This is usually performed by computers, by analyzing word frequencies, grammatical structures etc. By applying distant reading, one can for example, analyze the characteristic features of Victorian literature, by feeding a large number of novels to a computer, instead of reading closely each novel from a small sample set of novels. In [1] the authors use a technique called Geographical Text Analysis (GTA) which uses distant reading, to probe the records of the Registrar General about information regarding places of cholera epidemics in Britain. The authors find that London was most commonly mentioned in the records, and the probable cause was indicated as poor quality of water supply.

Although distant reading is extremely useful, it has the disadvantage that it misses subtle details that may be important, that can be covered only by close reading the relevant portions. Digital texts offer the flexibility to read in a non-sequential manner, but it is difficult to organize information obtained in this way. Correlating this with places is a possible solution. Place names can be used to search through the data to obtain relevant information, which can then be organized under that place. In [1], the authors used names of places in rural England to search through newspaper databases to find the reasons of variation in rate of decline of infant mortality rate in neighboring places, and the efficiency of the local government was found to be an important factor. Similar technique was used to identify the context in which different places in the Lake District were mentioned in literature describing the region. The authors in [1] also performed network analysis, and cost surface analysis techniques to obtain additional information not directly mentioned in the texts, such as the likely route followed by a writer through the Lake District, by looking at the places and the order in which they were mentioned.

Fictional Maps

Understanding geographies in literary texts also helps to create maps for fictional places, which is explored in [2]. This can provide a better understanding of the text, and especially help in linking the author’s ideas with real world equivalents, as many fictional places are related to real places in some sense. For example the cultural aspects of the author’s time period would have an effect on the depiction of the fictional places. As digital tools for map making, GIS based tools, and modelling software, become easier to use, more readers are able to perform this task, and this allows them to create maps unique to their own interpretation of the text. Digital maps can also represent the evolution of a place over time using layers to represent time periods. Using such tools in the classroom while teaching literary texts would be of much benefit to students.

Problems in Visualization 

Maps and other visualizations are great tools, as they enable us to better understand relationships, as compared to texts. However visualizing changes in place and time as described in historical texts present several problems, as discussed in [3]. The root of the problems lies in the fact that there is vagueness and uncertainty in the way in which events are described in historical documents with respect to place and time. This can occur due to reasons such as reference to outside events about which little information is available, or transcription errors while transcribing from oral or written sources that are lost. Fuzzy logic provides a way to deal with this uncertainty, as it allows the representation of vague descriptions such as ‘slightly’, ‘very old’ etc. commonly found in historical documents.

Another difficulty is that traditional visualization methods require that the historical data should be classified into clear categories. But this can lead to the loss of information that is contained in the context in which the data appears. An example is given in [1] regarding an unnatural death and the statements of witnesses regarding it. Different witnesses may describe the same event as a murder, accident or act of self defense. Categorizing it simply as an ‘unnatural death’ loses the information in the context.

Conclusion

Comparing the three articles, it can be seen that while [1] explores various leading technologies used to obtain geographical information from historical and literary texts, [2] describes the the value of representing such information through maps, particularly for literary texts. Finally, [3] describes the obstacles remaining to be overcome in order to obtain such a visualization of changes in place and time, particularly in historical texts. Thus we see that developing digital techniques to better understand the places and times mentioned in historical and literary texts is a currently active trend in digital humanities, as exemplified by the three articles described.

References

[1]  Gregory,I., Donaldson,C., Murrieta-Flores, P., Rupp, C.J., Baron, A., Hardie, A., Rayson, P. Digital Approaches to Understanding the Geographies in Historical and Literary Texts Digital Humanities Lausanne-Switzerland 2014. http://dharchive.org/paper/DH2014/Paper-790.xml

[2] Lynch, J., Kurtz, W., Rocchio, M. A sense of Place: Mapping Fictional Landscapes in Literary Narratives.  Digital Humanities Lausanne-Switzerland 2014 http://dharchive.org/paper/DH2014/Paper-723.xml

[3] Ó Murchú, T., Lawless, S. The Problem of Time and Space: The Difficulties in Visualizing Spatiotemporal change in Historical Data. Digital Humanities Lausanne-Switzerland 2014 http://dharchive.org/paper/DH2014/Paper-644.xml

[4] Link about Distant Reading http://www.nytimes.com/2011/06/26/books/review/the-mechanic-muse-what-is-distant-reading.html?pagewanted=all&_r=0

[5] Corpus Linguistics – Wikipedia http://en.wikipedia.org/wiki/Corpus_linguistics

[6] Geographic Information System – Wikipedia  http://en.wikipedia.org/wiki/Geographic_information_system

 

 

 

 

Advertisements