The Personal library and presumed reading of Algernon Charles Swinburne

Brief Project Description

Our data set is a TEI-encoded bibliography—with links to full-text resources—of the personal library and presumed reading of Victorian poet and critic Algernon Charles Swinburne. Swinburne is a central cultural figure of the Victorian period. He is an important poet who influenced 19th-century writers such as Oscar Wilde, Walter Pater, and Thomas Hardy and important modernists, including William Butler Yeats, Ezra Pound, and T. S. Eliot. His poetry was innovative, with metrical and rhythmic brilliance and virtuosity, and shattered conventions and barriers in its treatment of taboo sexual, religious, and political topics. Swinburne had a close association with the Pre-Raphaelite painters and poets and was a close friend of the Rossettis (Dante Gabriel, William Michael, and Christina), William Morris, and Edward Burne Jones. Beyond Pre-Raphaelite circles, he was a friend or acquaintance of other prominent literary figures of the period, including Tennyson, Browning, Arnold, and Ruskin, and visual artists such as James McNeil Whistler and Simeon Solomon. Swinburne was also a prolific critic of literature and art and an early champion of William Blake, Edgar Allan Poe, Charles Baudelaire, and Walt Whitman. He revived interest in many Elizabethan and Jacobean dramatists, including Ben Jonson, Thomas Dekker, John Webster, Beaumont and Fletcher, John Ford, and others.

Because of Swinburne’s fame and influence and his centrality and prominence in the network of Victorian literary and cultural people, events, and documents, the data presented here should be of interest to anyone studying Victorian literature and culture. In particular, the data could be used to:

• profile the reading habits of an important Victorian literary and cultural figure;
• compare Swinburne’s library with the libraries of other cultural figures;
• assemble a full-text corpus for computational text analysis of the content of a Victorian personal library;
• explore questions related to book history and readership.

For our own research, we compiled the data to analyze and profile the library and presumed reading of Swinburne. We want to know, for instance, the authors, languages, genres, and literary periods represented in Swinburne’s library. Building on our analysis of the bibliographic data, we assembled a full-text corpus of Swinburne’s personal library and presumed reading. We are now undertaking a comparative full-text analysis of that corpus and Swinburne’s own poetry, looking for similarities and differences, influences, and intertextual connections.

Creators’ Names, Institutions, and Contact Information

• John A. Walsh, Indiana University, jawalsh@iu.edu
• Alexandra Wingate, Indiana University, alewinga@iu.edu
• Caroline Nurkkala, independent scholar, caroline.nurkkala@gmail.com
• Alyssa Ollier, Kentucky Historical Society, alyssa.ollier@ky.gov
• Damien Thomas, Indiana University, thomdaro@iu.edu
• Jennifer Christie, Indiana University, jechri@iu.edu
• Evan Brandon, Indiana University, eobrando@iu.edu

Funder

We are grateful for generous funding from the Indiana University Presidential Arts and Humanities Program.

Date of Creation & Dates of updates

The project began in 2019 with the transcription of the Sotheby’s sales catalog. In the 2021-2022 academic year, the sales catalog data were augmented by additional “presumed reading” derived from citations in Swinburne’s poetry and other sources. Data checking, cleaning, and authority control continued throughout leading up to the 1.0 release on 17 March 2025.

Languages

Most of the items in the bibliography are English-language texts, although there are a number of items in French, Italian, Greek, Latin, and other western languages.

Collection & Creation Methodology

The foundation of the data set is a TEI-based bibliography derived from the sales catalog for Swinburne’s personal library, which was sold at auction by Sotheby’s over three days, 19-21 June 1916. The catalog is an 83-page document listing 938 auction lots. We transcribed from a facsimile of the catalog published in volume 6 of Sales Catalogues of Libraries of Eminent Persons (London: Mansell, with Sotheby Parke-Bernet Publications, 1972), edited by John Woolford, 247-363.

We identified 1540 items in the sales catalog. An “item” could include many individual works, since many of the items are multi-volume sets of an author’s complete works. For instance, lot 693 includes a 48-volume edition of Sir Walter Scott’s Waverley Novels (Boston: Estes and Lauriat, 1893).

It is important to note that while this sales catalog should represent the majority of Swinburne's library, it is not his complete library. The infamous bibliographer and forger T. J. Wise had first access to Swinburne’s library and claims to have acquired from it all of Swinburne’s unpublished manuscript material. But he also acquired printed volumes, including first editions of Elizabethan dramas (Woolford 249). To expand our reach beyond the sales catalog, we have searched the catalogs of forty-four national, public, and university libraries and located 16 additional books to add to our dataset. Twenty-seven total items have a note[@type="provenance"] indicating the current institution holding them. We will continue to search library catalogs for additional copies and update the data set accordingly. Eight additional items owned by Swinburne have also been identified in the sales catalog of the library of his close friend Theodore Watts-Dunton.

For our own research, we are interested not just in the items in Swinburne’s library but in other works he may reasonably be presumed to have read. So we supplemented the foundational bibliography of works in Swinburne’s library with additional works that Swinburne cites in his own poetry and Swinburne’s list of “The Best Hundred Books” published in the Pall Mall Gazette January 26th, 1886.

The titles, epigraphs, and authorial notes of many of Swinburne’s poems include references to specific authors and works. For instance, Swinburne’s “Birthday Ode” for Victor Hugo is a sort of bibliography in verse, and the poem is accompanied by extensive notes that identify the individual works by Hugo that are referenced by specific lines in the poem. We easily and systematically identified such items because the six volumes of Swinburne’s collected Poems (London: Chatto & Windus, 1904) have been transcribed and encoded for The Algernon Charles Swinburne Project (Walsh). The TEI encoding of the poems explicitly identifies titles, epigraphs, and notes, as well as any bibliographic references contained within those elements.

Due to our interest in performing text analysis with the dataset’s full texts, we have not added new bibliographic entries—nor their accompanying full text links—for works indicated by the Pall Mall article and Swinburne’s poetry if they are already represented by entries in Swinburne’s sales catalog. This is to avoid duplication of works in our corpus. We do, however, preserve instances where Swinburne owned multiple editions. From a distance reading perspective, the differences between editions are not as crucial as they would be in a close bibliographic analysis, though we have added specific editions when Swinburne has cited them, and they were not already represented in the sales catalog, such as the first edition of Fitzgerald’s translation of Khayyam’s Rubaiyat in no. 19 of the Pall Mall list. However, Pall Mall works and those referenced by Swinburne in his poetry can be extracted from across the entire dataset since we respectively have used #top100 and #referenced in the @corresp of <biblStruct> across the entire dataset to flag these items.

Data Structure

The data consists of a single TEI XML file. Using TEI div (division) element, the file is sub-divided into sections for the sales catalogs (div[@source = 'sales-catalogue-swinburne'] and div[@source = 'sales-catalogue-watts-dunton'), the “Best Hundred Books” list (div[@source = 'best-hundred-books']), the citations from the six-volume collected Poems (div[@source = 'poems-1904']), and items owned by Swinburne found in our search of library catalogs (div[@source = 'libraries']).

This separation of sources into separate div elements is a practical organizational decision but also supports a variety of research uses of the data. Swinburne is recognized as one of the most widely read of English poets, a poet with a tendency to break out in a recitation of entire long poems or passages that he had committed to memory. Given what we know of Swinburne’s extensive reading and uncanny memory it seems more likely than not that he read any individual volume included in the data set. Henry Adams (1918) described a dinner party at the home of Richard Monckton Milnes (later, Lord Houghton) where a young Swinburne discoursed on literature. The dinner guests “could not believe his incredible memory and knowledge of literature, classic, mediæval, and modern; his faculty of reciting a play of Sophocles or a play of Shakespeare, forward or backward, from end to beginning; or Dante, or Villon, or Victor Hugo” (p. 140). Of course, we cannot know that he read everything in his library just as we cannot know that he owned (or read) volumes that he quotes or cites in his poetry. Researchers are invited to come to their own conclusions or inferences about matters of ownership and reading with the data we provide and their own research. And the organization of the data allows researchers to include or exclude particular sources. For instance, if one wished to limit the data sets to volumes for which we have evidence of ownership, this is easily done by limiting the data to volumes within div[@source = 'sales-catalogue-swinburne'], div[@source = 'sales-catalogue-watts-dunton'), and div[@source = 'sales-catalogue-watts-dunton').

Below is an example record for a single work, William Michael Rossetti’s 1870 edition of Percy Bysshe Shelley’s poetry:

  <biblStruct corresp="#owned" xml:id="sb719.1">
    <monogr>
      <title>Poetical Works, edited, with Notes and Memoir, by W. M.
      Rossetti</title>
      <author corresp="#shelley_p">Shelley, P. B.</author>
      <note type="description">portrait</note>
      <imprint>
        <date when="1870"/>
      </imprint>
      <extent>
        <measure unit="volumes" quantity="2"/>
      </extent>
    </monogr>
    <relatedItem
      n="1"
      type="full-text"
      corresp="#ht"
      target="https://hdl.handle.net/2027/hvd.32044015680069"/>
    <relatedItem
      n="2"
      type="full-text"
      corresp="#ht"
      target="https://hdl.handle.net/2027/hvd.32044090294596"/>
  </biblStruct>

In some cases, the sales catalog has sparse bibliographic detail such that it will not accommodate the more complete <biblStruct> element. In such cases we use the more flexible <bibl> element.

Within the more complex XML structure of the biblStruct or bibl elements, the discrete fields include, when available:

Note that, whenever possible, we include links to full-text versions of the works in the data set. Of the 1731 items in the data, 1651 include links to full-text versions. There are a couple caveats to share about the full-text versions. Whenever possible, we tried to find a full-text version for the specific edition cited in our sources. However, the sources frequently lack the bibliographic detail necessary to identify a specific edition. If the data are ambiguous or we cannot find a full-text version of the edition specified in the data, we opted for another edition published before or during Swinburne’s lifetime, an edition that Swinburne might have owned. As a last resort we would include an edition published after Swinburne’s death. Most of the full-text versions are from the HathiTrust digital library, and the full text is generated through optical character recognition of page images. Thus, there are errors in the automated transcription—these are not carefully transcribed and edited full-text editions. However, they are suitable for many distant reading and text-data-mining analytical methods.

The data also contain structured lists of authority entries referencing established controlled vocabularies for the following fields identifying known people, places, and organizations:

The URL reference to the controlled vocabulary’s proper entry is placed in a @source. For places, we reference the Getty Thesaurus of Geographic Names (TGN). For people and organizations, in the first instance we use Library of Congress Name Authority Files (LCNAF), but if an LCNAF entry is unavailable, we link to the Virtual International Authority File (VIAF). If neither LCNAF nor VIAF is available, no @source is added to the entry. People also may have information encoded about their birth, death, or floruit dates if it is included in the LCNAF or VIAF preferred form of reference.

Each authority entry receives a unique identifier in @xml:id, which is referenced in @corresp of the <author>, <publisher>, and <name> of <respStmt> and @ref of <place>. Below is the authority entry for Swinburne.

<person xml:id="swinburne">
  <persName
    source="http://id.loc.gov/authorities/names/n79047585">
    Swinburne, Algernon Charles, 1837-1909
  </persName>
  <birth when="1837"/>
  <death when="1909"/>
</person>

Future work

The data set as described and presented is not a complete picture of Swinburne’s library and presumed reading. It does represent a complete picture of the personal library as represented by the Swinburne and Watts-Dunton sales catalogs. This representation of the personal library is augmented by presumed reading gleaned from references in the six volumes of Swinburne’s collected Poems (1904) and “The Best Hundred Books” list that Swinburne sent to the Pall Mall Gazette in 1886. However, we hope to augment the data set further with other sources, including references found in Swinburne’s letters, dramas, prose criticism, juvenilia, and any poetry published after the 1904 collected Poems. The six volumes of letters edited by Cecil Y. Lang and four volumes edited by Terry Meyers are likely to be the richest sources of new data, as well as the eleven-volume catalog of the Ashley Library constructed by T. J. Wise, which contains books belonging to Swinburne, presumably acquired by Wise along with Swinburne’s autograph manuscripts.

Ethics

We are not aware of any ethical issues related to the collection or use of these data.

Format

The data are formatted as XML (eXtensible Markup Language), an open file format documented at https://www.w3.org/TR/REC-xml/. More specifically, the data are encoded according to the Text Encoding Initiative’s Guidelines for P5: Guidelines for Electronic Text Encoding and Interchange and conform to the schemas of the most recent version of the Guidelines (at the time of writing, version 4.4.0. Last updated on 19th April 2022, revision ff9cc28b0).

Statement of Collaboration

The data were produced collaboratively, with John A. Walsh designing and supervising the project. The initial transcription of the sales catalog was performed by Caroline Nurkkala, with assistance from Jennifer Christie. Alyssa Ollier performed additional encoding, with a focus on data cleaning, correction, and implementing controlled vocabularies for author and publisher names. Alexandra Wingate and Damien Thomas continued with cleaning, correction, and authority control and also supplemented the library sales catalog data with items cited by Swinburne in his poetry.

Versioning

As we continue with the research projects supported by these data, we will periodically update the data. Significant updates will include the addition of items cited in Swinburne’s letters, his dramatic works, and his prose criticism and fiction. Updates and revisions are documented in the project’s git repository hosted at https://github.com/jawalsh/swinburne-library.

Licensing & Rights:

The Personal library and presumed reading of Algernon Charles Swinburne © 2023 by John A. Walsh, Alexandra Wingate, Caroline Nurkkala, Alyssa Ollier, Damien Thomas, and Jennifer Christie is licensed under CC BY 4.0.

Data Citation

Walsh, J. A., Wingate, A., Nurkkala, C., Ollier, A., Thomas, D., Christie, J., & Brandon, E. (2025). The Personal library and presumed reading of Algernon Charles Swinburne (Version 1.0) [Data set]. https://github.com/jawalsh/swinburne-library