Skip to main content

Review: Stories in Stone

A review of Stories in Stone, an annotated guide to the collections and papers of Ernest Westlake, directed by Rebe Taylor with Michael Jones and Gavan McCarthy

Published onNov 15, 2021
Review: Stories in Stone

Stories in Stone

Project Leads
Rebe Taylor, University of Tasmania
Mike Jones, University of Melbourne
Gavan McCarthy, University of Melbourne

Project URL

Project Reviewer
Paul Turnbull, University of Tasmania

Project Overview

Gavan McCarthy

Stories in Stone: An Annotated History and Guide to the Collections and Papers of Ernest Westlake (1855-1922) is a web resource by Rebe Taylor with Michael Jones and Gavan McCarthy that was first published in 2013 by the University of Melbourne, Pitt Rivers Museum, and the Oxford University Museum of Natural History. The resource annotates and reproduces four, large paper-based archives held in the Pitt Rivers Museum and the Oxford University Museum of Natural History in their entirety, along with materials from one other archive, and all their associated metadata. These papers relate to three large stone collections created by English amateur scientist Ernest Westlake from about 1870 to 1920—13,033 Tasmanian Aboriginal stone implements; an estimated 10,000 English palaeoliths, eoliths, and fossils; and more than 4,000 French eoliths—and to Westlake’s research into spiritual and psychical phenomena from the mid-1880s.

In 2006, Taylor approached Gavan McCarthy, director of the Australian Science and Technology Heritage Centre (later, from 2007 to 2020, the eScholarship Research Centre, or ESRC) at the University of Melbourne about creating a digital archive guide of the Westlake Papers. McCarthy prepared an initial inventory listing in the Heritage Documentation Management System (HDMS), a standards-based system developed by the ESRC to manage archival and artifactual documentation projects and to generate HTML and Encoded Archival Description (EAD) finding aids for publication on the web. McCarthy divided the archive into series or sets of records based on the evident logical divisions of the records themselves but retained all previous identifiers. In 2008, Taylor and McCarthy travelled to the Pitt Rivers Museum and, in one week, captured the entire Westlake Papers and the documents related to Westlake’s stone implement collections and related archives and metadata: a total of 3,678 images. McCarthy then processed the images using software related to the HDMS, linking them to the relevant inventory item. A draft guide could then be produced, with the relevant images accessible via a link beside each inventory item's description. In 2010, Jones, also of the ESRC, went to Oxford and captured the Westlake Papers in the Oxford University Museum of Natural History, including his papers relating to his French eolith collection. These were simultaneously listed as inventory items in a further three series added to the Westlake Guide, and the additional 4,586 resulting images were processed and attached as before. We then had a comprehensive inventory for the records held by both institutions, with complete sets of digital images for all those records attached to every item at inventory level.

Taylor worked during these intervening years to describe the series and the inventory items, also writing their contextual history and drawing connections between inventory items and sources, peoples, and events outside of the resource. Taylor’s goals stretched the design of HDMS; where necessary, the heading “detailed notes and cross-references” followed and separated the physical description of the records from Taylor’s historical explanation in the inventory descriptions. We also added pages to the standard HDMS guide template to accommodate the historical metadata: a bibliography with links from inventory items, timelines showing the chronological history of work completed on the Westlake collections, and introductory pages with historical background designed to ensure archival terminology did not create a barrier to potential users. To comply with the citation requirements of the Pitt Rivers Museum, the digitized records in Stories in Stone are presented in a tailored image viewer with the capacity to create customized citations of each image from EAD XML, including their location as paper records and their reproduction within the scholarly resource.

Some pages had to be added or manually adjusted, at times by hand-coding HTML, but these changes made it possible to tell a more complex story. In this way, Stories in Stone went beyond an archive guide and became a history. The 8000-plus images are informed by over 40,000 words of scholarly history writing. While Stories in Stone includes historical writing, it aims to reproduce and describe the papers without prioritizing any part of the collection or guiding viewers in any prescribed direction or interpretation.

We did not design Stories in Stone for a specified target audience. The resource was created with a recognition that Westlake’s papers, and those of researchers who studied his collections, were never published, but were of potential interest to a range of people. We expected those people would include the Tasmanian Aboriginal community because Westlake’s diaries from his journeys in Tasmania include interviews with their ancestors. Other anticipated users included scholars of the histories of archaeology, geology, collecting, Victorian spiritualism, languages and evolutionary anthropology; histories of Tasmania, France and England, of travel and museums; and genealogists and biographers.

Stories in Stone arguably set a new benchmark for web-based archive guides and historical web resources. Digital archive guides are often confined to a single collection and do not always capture that archive in its entirety. Even when item-level archival description is included, historical contextual material is usually only present at collection and provenance level and is often sparse. Stories in Stone includes the richest source Tasmanian Aboriginal culture, history, and language dating from the early 20th century, the result of Westlake’s interviews with more than 90 Tasmanians, many Aboriginal, from 1908-1910. Digital media made it possible to reproduce Westlake’s notebooks without loss to linguistic symbols, sketches, and addenda. Stories in Stone was produced as part of the Australian Research ARC Discovery Fellowship “From Race to the Genome: The Tasmanian Aboriginal People within the Scientific Imagination, 2006–2013” and received the 2013 Mander Jones Award for best finding aid to an archival collection (Australian). 

Project Review

Paul Turnbull

Stories in Stone is an annotated digital history project and guide to the collections and papers of Ernest Westlake (1855-1922). An English geologist and paleontologist with wide-ranging interests, Westlake was especially drawn to the study of prehistoric stone tools. Stories in Stone draws on research by Rebe Taylor, an award-winning Australian historian, and the expertise of Mike Jones and Gavan McCarthy, two internationally-known researchers in cultural informatics and digital archiving.  

This remarkable resource, built and hosted through late 2020 by the now dissolved eScholarship Research Centre of the University of Melbourne, offers readers the chance to examine the five boxes of Westlake papers in the possession of University of Oxford’s Pitt Rivers Museum. The papers include Westlake’s notebooks and related writings from his years in Tasmania (1908-1910), where he collected over 13,000 stone tools made by the Palawa (first people) of the island over many thousand years before their violent dispossession and loss of their traditional lifeways in the wake of European colonial invasion. 

Westlake believed—as was the consensus in European scientific circles of the time—that prior to European invasion of their ancestral lands, Palawa people were socially and culturally similar to the inhabitants of Europe through the Paleolithic era. Indeed, Palawa people were erroneously imagined to be an evolutionarily “primitive” race that had become extinct by the 1870s, leaving individuals of mixed Palawa and European ancestry with supposedly little or no connection with their ancestral lifeways and culture. Hence Westlake’s motivation in collecting Palawa stone tools—European worked stones could be arranged to show the occurrence of a gradual, linear evolution from ancient “primitive” types to more “sophisticated” artifacts. The stories that the stone tools, collected in Tasmania, were made to tell by Westlake and fellow anthropologists was one of sociocultural evolution that implicitly and perniciously privileged Eurocentric perceptions of human diversity, confirmed the legitimacy of Australian settler colonialism, and proclaimed the extinction of Tasmania’s first people. 

Even so, Westlake believed that the people of Palawa and European ancestry he encountered in his Tasmanian travels—most of whom were subsisting on the fringes of white society by sea fishing and seasonal agricultural work—might nonetheless be able to recall Palawa life prior to European invasion. This would in turn potentially provide insights into the lives of the “stone age” peoples whose fossilized remains had been discovered across Europe.   However, as Rebe Taylor points out, Stories in Stone allows us to see that Westlake’s notes on his interviews with individuals of Palawa ancestry are a rich trove of information about people still richly connected to a complex culture and rich history extending back many thousands of years into our deep past. In this respect, Stories in Stone is an important contribution to the decolonization of Australian history. 

In terms of information architecture and usability, Stories in Stone is an exemplary research-based online resource. It was created using a content creation and publication system called the Online Heritage Resource Manager (OHRM). OHRM has been in use since the late 1990s and was developed by Gavan McCarthy, Joanne Evans, and Tim Sherratt of the University of Melbourne’s Science and Technology Heritage Centre. It has since been refined through numerous research projects involving the creation of digital resources. In recent years, further development of OHRM informatics has occurred under Gavan McCarthy’s direction of the eScholarship Research Centre at the University of Melbourne.1

Technically, OHRM uses Microsoft Access and bespoke Visual Basic code to create data that is described using archival standards developed by the International Council on Archives (ICA). It exploits the print function of Microsoft Access to output data in HTML web pages, and as XML records in standard international schemas, such as the Encoded Archival Context Description (EAC-CPF) maintained by the Berlin State Library (Staatsbibliothek zu Berlin). In this way, OHRM takes advantage of the power of a relational database like Access to create relations between entities such as people, persons, places, and events, and to reproduce these relations in OHRM’s HTML output. Data within OHRM is also exportable in open structured formats, allowing for its long-term curation and re-use in future platforms. To date, for example, there has been experimentation with the RO-Crate approach to packaging up OHRM content and associated metadata.2

OHRM separates the creation and curation of data from its output in HTML (or XML, should this be required). This has the advantage that the HTML output sits as static files within a public_html directory. It differs from most knowledge bases currently being created by scholars in the digital humanities in that information is not called from a Structured Query Language (SQL) type database using scripts in written Python or a comparable programming language. OHRM-created files are thus not only speedily served but also can be confidently cited by other web resources. Links to and between the HTML files in question will not break. Even if they are physically moved elsewhere, users can access them by a simple redirect file. Also, the files can have rich metadata records in their headers that ensure the information they provide has a high visibility in searches by commercial and scholarly search engines.   

In these respects, the informatics of OHRM could be said to have anticipated the JAM Stack approach to web development that has emerged over the past 3-4 years. JAM Stack web resources similarly entail the creation of pre-rendered pages which can be served directly to users without using dynamic servers. Where they differ from earlier static websites, which proved harder to maintain than dynamic content publishing solutions like WordPress, is in the combined use of JavaScript, application programming interfaces (APIs), and static content files. In its most effective form, JAM Stack web resources are made up of files with simple markup (e.g. Markdown) that is transformed into HTML using a static site generator such as Hugo, Jekyll, or Gatsby, which are hosted on GitHub or GitLab or accessed by users via Netify or other companies offering hosting services that enable the serving of content to users across geographically distributed networks. It is also possible to serve content files created by one of these static site generators locally. What is lost in using this approach, rather than a content management system like WordPress, is the ease and speed with which new marked-up files can be uploaded and an entire site can be re-rendered in a matter of seconds to incorporate new content and links to information in other files it might contain. It also requires rebuilding site indexes and files associated with the site’s search engine.   

There are no immediate plans to move from using OHRM to a fully JAM Stack framework, despite the attraction of ending reliance on Microsoft Access and Visual Basic code in favor of open source software. Access and Visual Basic will be around for some time yet, but as with all proprietary software, there is a significant risk that its future development may adversely affect OHRM’s functionality. Secondly, employing a JAM Stack framework would allow OHRM to overcome its most significant long-term shortcoming, which is the inability for online addition and editing of content, as is possible in a database back-end web resource built on an open source content management system such as Drupal, Joomla, or WordPress.  

Regrettably, future development plans for OHRM are on hold, due to the University of Melbourne’s 2020 disestablishment of the eScholarship Research Centre, which has seen McCarthy and the center’s talented staff made redundant. Researchers collaborating with the center on OHRM-based projects—the quality of which is reflected in resources such as Stories in Stone— have found themselves forced to move their projects and related assets off University of Melbourne servers at short notice and with meagre assistance. Why the university would destroy such an important, pioneering facilitator of Australian humanities and social science e-research as the eScholarship Research Centre has yet to be explained and justified.   

No comments here
Why not start the discussion?