American writers of the 20th century

Corpus description

This is a corpus of the second half of the 19th and the beginning of the 20th century British and American literature. It includes 53 books, further divided into 23 books of British writers (Joseph Conrad, Thomas Hardy, James Joyce, Joseph Rudyard Kipling, George Mac Donald, George Orwell and Virginia Woolf) and 24 books of American writers (George Washington Cable, William Faulkner, F. Scott Fitzgerald, Ernest Hemingway, William Dean Howells, Louisa May Alcott and Mark Twain). The remaining 6 books belong to Henry James and are an interesting element in my analysis because of the fact that James is considered to be Anglo-American writer. The reason why I chose such corpus is that I wanted to compare British and American literature from the aforementioned period and see if there are any differences or similarities. An explanation of the choice of the texts is the fact that the books were easily found at “Project Gutenberg” website and are commonly known as their authors, which is a helpful factor when it comes to describe the results.

The authors selected for the project are well-known and they lived and wrote at a similar time. Therefore, I suppose to see some strong similarities especially between the writers from the turn of the century and also between those who wrote only in the 20th century. Moreover, I assume that the style of modernist writers (Virginia Woolf, Joseph Conrad, Ernest Hemingway and William Faulkner) as well as the realist ones (George Washington Cable and William Dean Howells) will be displayed in some network of stronger similarities. I expect to see clear differences between American and British writers in one of my analyses. However, I could not predict the characteristic feature (either more American or more British tendencies) of James’ texts. Additionally, I expect to notice some similarities between Rudyard Kipling and George Mac Donald as they are both considered with children’s books. I am curious about the tendencies of James Joyce’s texts because he is associated with stream of consciousness method of writing and modernist avant-garde.

The list of all of the text that were analysed in this project can be seen in Table 1.

Table 1: British and American corpus

Author	Title	Publication date
George Washington Cable	Dr Sevier	1897
	Grandissimes	1899
	Kincaid’s Battery	1908
Joseph Conrad	Lord Jim	1900
	Nostromo	1917
	Under Western Eyes	1911
William Faulkner	As I Lay Dying	1930
	Collected Stories	1950
	The Unvanquished	1938
F. Scott Fitzgerald	Flappers and Philosophers	1920
	Tales of the Jazz Age	1922
	The beautiful and Damned
	This Side of Paradise
Thomas Hardy	A Pair of Blue Eyes
	Far from the Madding Crowd
	The Hand of Ethelberta	1895
	The Return of the Native	1912
Ernest Hemingway	A Farewell to Arms	1929
	For Whom the Bell Tolls	1940
	The First Forty-Nine Stories
	The Sun Also Rises	1926
William Dean Howells	A Hazard of New Fortunes	1909
	Indian Summer	1886
	Venetian Life	1867
Henry James	The Europeans	1878
	The Portrait of a Lady	1881
	The Turn of the Screw	1898
	Washington square	1907
	What Maisie Knew
James Joyce	A Portrait of an Artist as a Young Man	1916
	Dubliners	1914
	Ulysses
Rudyard Kipling	Indian Tales	1937
	Kim	1901
	Plain tales from the Hills	1888
George MacDonald	At the back of the North Wind	1871
	Sir Gibbie	1879
	The Portent : A Story of the Inner Vision of the Highlanders	1864
Louisa May Alcott	An Old-Fashioned Girl	1869
	Jack and Jill	1880
	Rose in Bloom	1876
George Orwell	Animal Farm	1945
	Down and Out in Paris and London	1933
	Nighteen Eighty-Four	1949
Mark Twain	A Tramp Abroad
	The Adventures of Huckleberry Finn	1884 UK and Canada 1885 USA
	Innocents abroad
	What is a Man? and Other Essays	1906
Virginia Woolf	Jacob’s Room
	Monday or Tuesday	1921
	Night and Day	1919
	The Voyage Out	1915 UK 1920 US

Method

The analyses were conducted with the help of R programme, which is a software commonly used for statistical computing and data analysis. The program enabled graphical representation of the data gathered in the corpus. In R program I was able to carry out cluster analysis, bootstrap consensus tree analysis and oppose analysis.

The first analysis- cluster analysis resulted in vertical dendogram in which I was able to investigate the relationship between the texts relying on the most frequent words used in the books. The bootstrap consensus tree analysis presented a network of the texts and marked their relationship by showing a “branches” of texts more or less similar on the basis of the most frequent words appearing in the books. However, the next analysis- oppose analysis provided a contrastive analysis between British and American writers testing James’ membership at the same time.

The last analysis was carried out in Gephi program, which constructs dynamic and hierarchical graphs and an interactive visualisation of various networks and complex systems. The data file was created in R and transferred to Gephi. The aim of this analysis was to display not only the nearest neighbours (as in the previous analyses) but also to show weaker connections with the use of colours and lines of different thickness.