Aims and objectives
The main objective of the Eighteenth-Century Poetry Archive — ECPA — is to develop into a lively, collaborative workspace in support of the teaching, study, and research of eighteenth-century poetry. To this end, ECPA is committed to the creation of an open-access digital archive of high-quality primary and secondary sources. ECPA comprises two central components: the digital collection of richly encoded full-texts and a research project, which focuses on the computationally-assisted analysis of these texts.
A digital collection
The ECPA full-text collection utilizes well-established open standards for creating and preserving richly-encoded digital texts. Based on the principle of peer participation, the corpus is edited and annotated collaboratively, and will grow and evolve with the requirements and interests of the scholarly community. ECPA thus participates in and benefits from a growing network of editors, scholars, and students who are collaborating and sharing texts, notes, and interpretations in a global network.
Wherever possible, we have supplemented the full-texts with digital images for scholars who are primarily interested in the appearance of the source document or who want to check the transcription against the original. We have also included an XML-editor, where corrections and improvements (descriptive and analytical) can be made directly in the source files and can be submitted to the editor for inclusion in future updates.
All digital surrogates can be downloaded in a variety of formats and re-used under the terms of ECPA's contents license. If you re-use and do exciting things with any of the digital surrogates, please do let us know about it.
A research project
The Eighteenth-Century Poetry Archive aims to make a scholarly contribution by producing a corpus of annotated texts, analyses, tools, and a host of secondary materials. The focus of the ECPA research project is the integration of texts and (digital) scholarship into a living research collection of eighteenth-century poetry. The main purpose of this integration is to facilitate interactivity and discovery, and to support a multi-layered reader engagement. For this purpose, we have introduced a variety of views, which reflect different modes of engagement with the texts.
We have represented each of the different modes of engagement in the side-by-side layout of text and view in the form of tabs (currently, Reading view, Analysis view, and Visualization view) in the hope that the information provided in each will best support the level of engagement the reader has chosen.
The reading view offers a set of bibliographic metadata, information about the source edition, a brief summary of poetic form, an editorial statement, and other contextualizing information, as well as a reading aid function. The five analytical layers that support the analysis view are intended to support a close reading of the texts, and help to reflect this engagement through enabling interactive selection, highlighting, and manipulation of the observed phenomena. Our visualization view comprises a growing number of visualizations intended to support the analysis/interpretation of the poems. By shedding light on the texts from different visualization perspectives (presentational/disseminative, operational/observational, analytical/interactive, creative/inventive) and with varying foci on one or more analytical layers, the visualizations serve as a toolbox for researchers interested in visual interpretation. We hope that through the combination of knowledge modelling and machine learning we can arrive at an "augmented criticism" (Bradley/Ullyot) based on a larger pool of supporting materials and more comprehensive analysis.
Over time, the ECPA research project will facilitate new modes of engagement (views), which will join the current views, and allow for increasingly sophisticated ways of interacting and thus re-creating the texts through each reading process.
Work on the small project that eventually evolved into the more ambitious Eighteenth-Century Poetry Archive began in 2006. Originally intended as a focused contextualisation to the Thomas Gray Archive, covering the lives and works of contemporaries of Gray's such as William Collins, Mark Akenside, Joseph and Thomas Warton, Christopher Smart, James Thomson, the materials quickly grew substantial in quantity and wide-ranging in subject matter and coverage.
In 2008, the idea for a much more comprehensive and ambitious project that would contribute to and benefit from the many exciting developments in digital literary studies and humanities computing at the time was born. With full-text production of ECCO-TCP in full swing in Oxford, early experiments in computational literary criticism proved promising and encouraged us to embark on a dedicated resource on eighteenth-century poetry. After several delays for various reasons, work on ECPA began in earnest in Winter 2012.
An alpha version of the Website was completed in Winter 2015. The beta phase of ECPA began on 1 May 2016 and ended on 1 May 2019. We continue to seek post-publication peer review, please do not hesitate to contact us with any issues, requests, or general feedback about the contents, functionalities, or design of the Website.
The Eighteenth-Century Poetry Archive is an unfunded research project. As a collaborative effort of scholars, teachers, and students of eighteenth-century poetry, ECPA is made possible primarily through the generous donation of its contributors' time and expertise. We anticipate funding opportunities, particularly through (interdisciplinary) collaboration on research projects or participation in national or international initiatives, will become available over time.
We are collaborating with the ERC-funded POSTDATA - Poetry Standardization and Linked Open Data project (2016-2022) to construct a digital platform of semantically linked poetry. POSTDATA is a research project at the Laboratorio de Innovación en Humanidades Digitales (LiNHD), Universidad Nacional de Educación a Distancia (UNED), Madrid.
ECPA builds on the digital texts created by the Text Creation Partnership from Gale’s Eighteenth Century Collections Online (ECCO). We use the TEI/XML P5 versions of the texts as they are made available by the Oxford Text Archive (OTA) as our starting point (all changes to the documents have been recorded as revisions). The following is a high-level summary of the current steps taken to create the digital surrogates in ECPA. Any tools used for a particular task are also listed.
Transcription and descriptive markup
- single-pass correction of transcription and markup (addresses gaps, structural errors, omissions) against the ECCO microfilm images
- illegible text was supplied either from the copies originally used for the microfilms or other available copies of the same edition
- enhanced 'semi-diplomatic' transcription of a small sample of texts (adding typographical features, drop capitals, indentations etc.), otherwise 'semi-normalized'
- enhancing of bibliographic information and references (Foxon, DMI, etc.)
- adding author information on item level
- assigning of IDs to poems and lower-level structural units
- recording verse, stanza, and poem form
- assigning of themes and genres (vocabulary adapted from DMI)
- assigning rhyme scheme (where applicable)
- assigning metrical pattern (where applicable)
- recording of syllable pattern
- addition of textual and explanatory notes
- tokenization/sentencing (white-space tokenization, Morphadorner)
- PoS-tagging (dictionary-based, Morphadorner)
- lemmatization/regularization (dictionary-based, Morphadorner)
- syllable count (Morphadorner)
- spot-checking and correction of the above steps
- phonemic transcription (mostly dictionary-based, self-developed tool)
- metrical and rhyme analysis (self-developed tool)
- rhetorical figure detection (self-developed tool)
- named entity recognition (mostly gazetteer-based, generated from several NER engines, including the Stanford Named Entity Recognizer, MITIE: MIT Information Extraction, and Apache OpenNLP )
- syntactic dependency parse (MaltParser [pre-trained, optimized])
- frame-semantic parse (SEMAFOR [pre-trained]) ~ experimental
- analysis of phonemic distribution and clusters
- automatic metrical scansion (based on expert and statistical rules)
- analysis of word-class specific phonological properties
- inclusion of (un)certainty measurements
- original and well-known adapted visualizations
- customized links to a range of externally hosted visualizations
- formal ontological modelling interface
- backend of RDF triplestore and SPARQL engine (rdflib.js)
- based on CIDOC-CRM family of core ontologies for the digital heritage domain
- implementation of CIDOC-CRM Fundamental Categories and Relationships for modelling purposes
- supplemented by domain (literary/poetic) ontologies (e.g. POSTDATA)
- adoption of Cytoscape.js for visualization and analysis
- based on the concepts of peer participation and peer content creation
- opportunities for user contributions (subject to peer review)
- review-based annotation system to any part of a text (notes/queries)
- direct editing of XML sources
- digital images (captured with an Atiz BookDrive scanning station and images taken in reading rooms)
- all content files available for download via the "Downloads"-tab, all data supporting this research are also available from our GitHub page
Current and future plans
The Eighteenth-Century Poetry Archive is currently being updated twice a year, in the Summer and Winter. Our current priorities include:
ECPA currently comprises works by authors. We will continue to work towards increasing the number of authors, particularly women authors, and works represented.
Please feel free to send us suggestions for additions you would like us to prioritise, or, if you have your own collection of texts, we would be more than happy to collaborate with you on their inclusion in ECPA .
We are working on providing better interconnection between all analytical layers as these inform all current and future modes of reader engagement. As each layer interacts with the others at any given point as well as over time, we are looking for better ways to make these points of connectedness visible and highlight the ways in which these are maintained, dissolved, and taken up again (in different forms) over time.
As ECPA evolves and matures, we will expand our research to go beyond the current modes (views) of reader engagement (reading, analysis, visualization, and modelling).
Output and dissemination
The Eighteenth-Century Poetry Archive's main output is a comprehensive set of tools and high-quality resources aimed at supporting scholars, teachers, and students of eighteenth-century poetry throughout the research process, from the early stages of discovery, collection, and annotation through to the later stages of comparison, analysis, and composition.
All of ECPA's digital collections and research outcomes are freely available online for review and secondary analysis, future digital scholarship, opportunities for (interdisciplinary) collaboration, re-use, extension, or adaptation, and long-term preservation. All data can be downloaded from the Website or our GitHub page.
Conceived of as collaborative from inception, ECPA is also seeking to increase its outreach through collaboration with libraries, scholarly societies, research projects, and other digital repositories and initiatives.
This is the complete post-release update history of ECPA. You can find the update history of the beta phase on a separate page.
- v1.7 (Summer 2022):
- Contents: added poems by Sir Richard Blackmore, Colley Cibber, Thomas Purney, and Sir Richard Steele (v1.7: 3217 works, 332 authors).
- Research: a prototype ECPA topic model browser is now available. It employs Andrew Goldstone's dfr-browser, which seeks to reflect the multifaceted nature of a topic model by offering multiple views of the model: topics, documents, words, and document metadata. Please send any feedback you may have to the editor.
- v1.6 (Winter 2021/22):
- Contents: added poems by Henry Carey, George Alexander Stevens, Thomas Holcroft, and John O'Keeffe (v1.6: 3212 works, 328 authors).
- Website: excited to announce an all new state-of-the-art search engine, offering many new avenues into the corpus, including full-text search and for the first time exposing the encoding of poetic form through a number of search facets.
- v1.5 (Summer 2021):
- Contents: 29 facsimiles have been added to the source editions page. All facsimiles can now be viewed in their entirety directly from the editions' pages. One poem by Susanna Blamire added (v1.5: 3200 works, 324 authors).
- Website: ECPA is delighted to announce adoption of state-of-the-art image viewer OpenSeadragon to underpin its digital library. We look forward to the greater flexibility and enhanced user experience offered by the new viewer moving forward.
- v1.4 (Winter 2020/21):
- v1.3 (Summer 2020):
- Contents: added works by Mark Akenside (ten poems), Elizabeth Hands (author page and 44 poems), and Christopher Anstey (one poem) (v1.3: 3105 works, 320 authors).
- Outreach: ECPA is very pleased to have become the first English-language poetry corpus in POSTDATA's corpora reader and downloader Averell. Averell facilitates cross-corpora poetry analysis at stanza, line, or word level granularity. ECPA has been a project partner in the POSTDATA project since 2016.
- v1.2 (Winter 2019/20):
- Contents: poems by three women writers, Elizabeth Craven (author page and five poems), ‘Christian Carstairs’ (author page and 35 poems), and ‘A Young Lady’ (author page and nine poems), added, taking the total number of women poets in ECPA to 60 (v1.2: 3051 works, 319 authors).
- Outreach: ECPA is delighted to have joined 18thConnect, making all of its primary and secondary resources discoverable alongside a host of high-quality digital scholarly resources in eighteenth-century studies.
- v1.1 (Summer 2019):
- Research: a prototype of the knowledge modelling view is now live, feel free to explore a sample model. Read all about the new modelling view, and, please let us know your feedback!
- Contents: added two late eighteenth-century authors, Mary Tighe (author page and 41 poems) and Matthew Gregory Lewis (author page and two poems) (v1.1: 3002 works, 316 authors).
We are interested in working with scholars and digital humanists on any aspect of the poetry of the long eighteenth-century. Some of the areas we think might be fruitfully explored include:
- interlinking of texts that form a clear dialogue with each other;
- exploring advanced digital editing of a text or author and edition visualizations;
- expanding an author or improving a single text or group of texts;
- collaborating on digital tools development, such as improving, training, or adding to the set of tools;
- (insert your idea here... and get in touch!)
If you would like to collaborate with us on a project, please do not hesitate to contact the editor.
Copyright information and citation guide
All digital surrogates of works not held by the Eighteenth-Century Poetry Archive are reproduced under license from the libraries, archives, museums, and other institutions which hold the originals and must not be reproduced without prior written permission. Unfortunately, ECPA is unable to provide that permission, nor can it provide images for re-use as the copyright holders may reserve the right to charge a fee. Please contact the relevant institution directly using the copyright information provided with every work in the Eighteenth-Century Poetry Archive.
Copyright in digital surrogates of works owned by ECPA, in the electronic texts, bibliographic information and markup, and the Website is with the Eighteenth-Century Poetry Archive. These works, including their XML source files, are licensed under a Creative Commons Attribution-ShareAlike License. This license is intended to encourage and promote re-use and re-purposing of the collection as well as ultimately to ensure its long-term preservation.
All contributions to ECPA are submitted to the editor in the first instance for review. When accepted for publication, all contributions become licensed under a Creative Commons Attribution-NonCommercial-ShareAlike License. This license protects contributions, but at the same time lets others re-use and build upon them.
If you want to cite material from the Eighteenth-Century Poetry Archive in online or print publications, please feel free to copy and paste the information provided in the the MLA-style citation in the footer of each page.
The Eighteenth-Century Poetry Archive is open access. All materials are made available under the license terms indicated above.
We use two external services which will store some small pieces of information on your computer in the form of cookies. We use AddThis to connect our Website to a number of Social Networking services and also to provide access to sharing, printing, and e-mail options. And we use Google Analytics to provide us with a measure of the usage and impact of our site, and to inform any outreach activities. We only submit anonymized IP addresses to Google Analytics. We consider these cookies essential to the provision of the service requested by the user.
We would like to thank the Text Creation Partnership for their efforts to create and make available searchable digital editions of the books in the Eighteenth Century Collections Online (ECCO) corpus.
We would like to thank the staff of Imaging Services at the Bodleian Libraries, University of Oxford for their expertise, advice and practical assistance in creating the digital images.
We would like to thank Zoomify, Inc. for the gift of the enterprise edition of the Zoomify software to ECPA.
The Eighteenth-Century Poetry Archive was founded by Alexander Huber (Bodleian Libraries, University of Oxford) who is also its editor. The Website was designed by .