Conference Papers

Taking stock of 150 years of collecting: stocktaking at the NGV

Article that appeared in the Melbourne newspaper The Age on 25 January 2012

In late January 2012, an article appeared on the front page of the Melbourne newspaper, The Age. The article featured an image of an artwork by Richard Parkes Bonington, Low Tide at Boulogne, 1824 under the heading, ‘Have you seen this painting? Thieves have struck — but don’t panic.’ So what was the story behind the headline?

Low Tide at Boulogne was formally declared as lost, believed stolen, from the National Gallery of Victoria collection in early 2011. The public declaration of the loss followed a period of unprecedented collection movement and the introduction of a comprehensive stocktake program at the NGV.

In late 2002, the Bonington wasn’t sighted during routine inventories of the painting collection in storage. In an effort to locate the work, the Registration department undertook a number of complete inventories of the paintings storage areas, as well as researching the location history and movement records in the Collection Management System (CMS) and cross- checking paper-based records. When these efforts failed to locate the painting, it was reported as possibly missing to the Council of Trustees in early 2003. In their meeting of February that year, the Trustees stated that ‘…until a full audit of all works is able to be carried out on our return to St Kilda Road, the work be considered as missing only…’

As the Trustees decision suggests, despite the rigorous efforts to locate the Bonington and the extensive security, location tracking and documentation systems in place to prevent its loss, the Gallery could not be certain that it was not somewhere in storage, yet to be discovered.

This is a small framed work, measuring only 22 x 17.8 cm. The Gallery has a procedure of boxing small paintings for transport, so it was possible that the painting had been packed in a box, the box put into storage and the movement not correctly documented. While this isn’t in keeping with our procedures for tracking the movement of artworks, unfortunately human error does occur from time to time and is usually resolved through the sorts of inventories and research that were undertaken in this case. But the move of the Bonington wasn’t part of the usual day-to-day movements of the collection; instead it was one of the 63,000 artworks the Gallery relocated when it closed its doors for redevelopment in mid-1999. Between that date and December 2003, the Gallery’s collections experienced five major relocations.

When the building at 180 St Kilda Road closed for refurbishment in 1999, the majority of the collection was moved to another site and housed over two levels of custom-fitted storage space. In November 1999, the ‘NGV on Russell’ opened in the Gallery’s first home in Swanston Street, at the rear of the State Library building. At this time, artworks also travelled on loan to galleries across regional Victoria, as well as travelling to the United States and New Zealand as part of a travelling exhibition of masterpieces from the NGV collection. In November 2002, the Ian Potter Centre: NGV Australia opened at Federation Square. While Gallery staff were preparing to move and install about 800 artworks at NGV Australia, they were also busy readying the collection for the closure of the Russell Street annexe, the move back into St Kilda Road and the consolidation of storage at the off-site store from two floors, into one. By the end of 2003, in addition to the 50 gallery spaces at both NGV Australia and NGV International, the collection in storage was spread across two sites.

In light of this, the full stocktake of the collection foreshadowed by the Trustees in early 2003, which began in mid-2004, was developed with a number of key objectives:

  • To compile a full and accurate record of works of art in the care of the NGV;
  • To compile accurate location records for these works of art;
  • To determine if any works of art are missing following the conclusion of the redevelopment projects;
  • To recommend an on-going process for Stocktaking the NGV’s collection; and
  • To establish baseline data for future collection tracking and collection management programs.

While this stocktake project follows a period of massive upheaval, in addition to regular inventories and surveys of high-traffic areas, the NGV has undertaken stocktake projects in the past. The first collection stocktake started in 1986 and presented its final report to the Council of Trustees in 1992. During that time, two staff worked to sight and document the Gallery’s Collection. The first stage of that project involved the creation of the NGV’s CMS. Registration staff entered information from the Gallery’s original stockbooks and catalogue cards into a database system.

The stocktake staff then progressed through the collection, working department by department to verify the location of the accessioned collection. The stocktake inventoried works and established the first entries in the electronic location histories for a large portion of the accessioned collection. Not all works were able to be sighted and while a large number of ‘found item’ or tracking records were created in the database for unidentified works, the project wasn’t able to tag or label all of these. This first stocktake established the Gallery’s electronic collection documentation systems. The findings were also instrumental in identifying the need for the expansion of the Registration department’s responsibilities to include the receipting and documentation of all new acquisitions and the accessioning of the collection, ensuring consistency of documentation procedures.

For the current project, with the CMS and procedures for documentation and collection tracking well entrenched and regularly reviewed and in light of the major collection relocations, our approach was quite different. Our methodology is informed by a recognition that this is an uncommon opportunity not only to verify locations but also to create a CMS record and establish a location for all works held in the Gallery’s care. Accordingly, we have worked wall to wall, rather than collection by collection, with staff inventorying, documenting and tagging all works of art as and where they are sighted. Confirmed location information is entered into the CMS record for each object and new records created to track work previously not documented or where the identity of the work cannot be confirmed. As a result, CMS records with at least basic catalogue details and a unique ID number now exist for all works held in storage and on display at the Gallery and on long-term outward loan.

The stocktake project has been divided into 4 key phases:

  • Phase 1: Sight and inventory artworks across the NGV’s 3 sites and on long-term loan at external venues;
  • Phase 2: Reconcile the findings
  • Phase 3: Prepare a final report, including the formal reporting of any works unable to be accounted for;
  • Phase 4: Develop a future Collection Stocktaking Policy.

With the project due to be completed in December 2013, this paper will first consider the approach we have taken to the first two stages and will then look ahead to the final two stages.

Phase 1

Sight and inventory artworks across the NGV’s three sites and on long-term loan at external venues

After developing a project plan and comprehensive procedures manual, in May 2005 the first team of two stocktakers started sighting in the two-dimensional screen area at the Gallery’s off-site store. The vast majority of works in storage here are paintings. As can be seen in Fig 1 above, paintings comprise 4% of the NGV’s collection of over 69,000 artworks. Housing a collection of well-documented and regularly inventoried works, the screen storage area was seen as an ideal place to start implementing and refining the procedures that would see us through the next 7 years. The initial sighting of over 370 double-sided painting screens took a little over 2 months. Over the next almost 7 years, the stocktake sighting team, which initially comprised two staff and was increased to four in late 2009, have completed sighting 98% of the storage, display and long-term outward loan areas.

A snapshot of stocktaking

Stocktake staff prepare for each area of sighting by producing a worksheet from the CMS detailing all works documented as being stored or displayed in a specific location.

To ensure transparency and for safety of the artworks and staff, the Stocktake staff work in pairs at all times, moving progressively from shelf to shelf, box to box, screen to screen, sighting all works in the selected area, and making note of:

  • Works sighted in their correct location
  • Works that are not in the location, as it is recorded in the CMS
  • Works sighted in the area that are not recorded as being there on the CMS

The stocktake team also document basic catalogue, condition and packing information about each work. In particular, they note any queries or discrepancies with catalogue information recorded in the CMS for each work and any additional information that may be useful, as well as documenting significant condition or conservation concerns and details about packing or storage that may need to be reviewed or addressed. Any immediate concerns are reported to the appropriate Registration and Conservation staff. All data is then entered into the CMS.

The Gallery has been collecting for over 150 years and over this time procedures and systems for tagging, labelling, documenting and tracking artworks have developed considerably. In order to identify, and in turn avoid entrenching, historical errors or assumptions, the project has taken a precautionary approach: if a work isn’t tagged, labelled or clearly identified, stocktakers create a ‘Found Item’ record on the CMS. The work is tagged with the unique identifying number, ensuring it can be located and tracked into the future. Found records in the CMS include basic catalogue information, photographs and any other identifying data that might enable the identity or source of the work to be determined.

In some instances sighting can be relatively straightforward. Rembrandt’s Portrait of a white-haired man is a two dimensional framed canvas that is well known and thoroughly documented. The work has been regularly displayed by the NGV and loaned to borrowers both nationally and internationally since it was acquired in 1951. Consequently, sighting of this work was a straightforward process, requiring art handling staff to assist with removing the painting from the wall to enable the stocktake staff to view the label affixed to the backing board. The work was then returned to the wall, and its location confirmed for the Stocktake on the CMS.

In other cases, stocktake documentation can be far more complex and time consuming, requiring the un-crating, handling, identification, documentation and tagging of hundreds of parts or components, and necessitating the involvement of Art Services, documentation, cataloguing and curatorial staff.

Durer’s The Triumphal Arch of Emperor Maximilian I – The Art of Honour consists of fifty sheets, with forty-nine fixed to four panels. One sheet is stored separately and not displayed. Each panel is housed in a separate crate and all sheets are tracked individually in the CMS. The Triumphal Arch was sighted as it was un-crated for installation. Curatorial and Registration documentation staff worked with the stocktake team to confirm the part allocation for each sheet on each panel. Detailed part information was entered into the CMS to improve the storage and collection management documentation of this significant work.

The pair of candelabra, designed in 1903 and manufactured by Baccarat, France in 1922 once flanked the proscenium of the auditorium of Melbourne’s Capitol Theatre designed by Walter Burley Griffin. Acquired by the Gallery in 1982, the candelabra are comprised of over 1250 individual components that had not been accurately documented. The stocktake team, in concert with conservators working to clean the works, and Registration documentation staff, patiently documented the individual components.

In addition to the physical complexity of individual artworks, the stocktake is regularly confronted by other issues complicating the sighting process. While some artworks are thoroughly catalogued and documented and are clearly tagged or accession marked, in other instances there can be very little information on which to base the verification of the artwork, or the data that is available may be inconsistent or inaccurate. As noted earlier, where the identity cannot be definitively confirmed, temporary identification, or Found Item records are created in the CMS to enable the work to be tracked. Research is then undertaken to confirm if the work is an already documented work in the NGV’s holdings, or to attempt to identify the artwork’s identity and source.

In an example of the sort of confusion that can arise, a two part lacquer-ware table in the Asian Art Collection had been tagged and tracked using an accession number for a group of three textiles, also in the Asian Art Collection. The textiles were also correctly tagged and tracked using this number. With limited access to the CMS in most storage areas and the fact that the table has not been regularly accessed or displayed, this tracking error had not been previously identified and the location history in the CMS included tracking of both the textiles and the table.

The stocktake team quickly created a Found Item record and retagged the table, noting the historical error in the CMS record for both the new tracking record and the incorrect accession record. In the process of researching this matter, an additional issue was unearthed, when it was discovered that an image of this table had been linked to the CMS record of a different table in the Asian collection. With curatorial assistance, the Stocktake team were able to confirm the correct identity of the second table, amend the CMS record and link the correct image. The lacquer-ware table and cover continues to be tracked by its Found item and further research will be undertaken as part of Phase 2 to attempt to resolve its identity. This confusion highlights the importance of accurate labelling and documentation of works, and is one of the many instances in which the Stocktake has helped to improve knowledge of and access to the collections.

In another example, it was discovered that contact prints of accessioned works in the Photography Collection had been labelled with the accession number of the formally accessioned work. While the labelling of the contact prints had been logical at the time the work was acquired, as the Photography collection expanded to the point where it now consists of over 10,000 artworks, this practice has resulted in confusion. The location histories of these works reflect a zigzag pattern whereby both the accessioned work and contact print have been tracked by the accession number. Resolving this issue not only enabled the contact prints to be tracked under their own unique CMS records and identification numbers, but also provided us with an opportunity to improve collection management and documentation of this collection more broadly, providing further information and access to important secondary material in the Photography collection.

Sighting is now complete in over 98% of the NGV’s storage, display and long-term loan sites. Since May 2005, the stocktake has sighted almost 73,500 artworks. This translates into over 106,000 individual items, including object parts. Over 4,600 Found Item records have been created to track unknown or unconfirmed works located during the Stocktake, with almost 7,500 Found Item records currently in use in the CMS.

  • To compile a full and accurate record of works of art in the care of the NGV;
  • Percentage of initial sighting completed: 98%
  • Number of works of art sighted: 73,320
  • Total number of individual items (including parts) sighted: 106,167
  • Number of Found Item records created in the Stocktake: 4,670
  • Total number of current Found Items records: 7,492
  • Projected completion date: December 2013

Phase 2

Reconcile the findings

With initial sighting nearing completion, the stocktake team have also been undertaking research to resolve the approximately 10% of Found Items that we believe may actually be part of the documented collection that have been recorded as not yet sighted.

The stocktake team will complete the first check of these works to resolve straight- forward matches, before calling on the expertise of curatorial and cataloguing staff. Research sources include the extensive paper-based and electronic documentation that exists in the CMS, Registration records and the NGV’s archive, including:

  • Original collection stockbooks (maintained up to 1992)
  • Catalogue worksheets and cards
  • Trustees papers and acquisition submissions
  • Registration, Curatorial, Conservation and archive files
  • Incoming and outgoing receipts, loan documentation, incoming and outgoing works registers and other historical collections management records
  • Exhibition catalogues and reference books

Where the identity of the work can be established, the artwork is correctly sighted and documented for the stocktake project and the Found Item record removed from the CMS.

Much of the confusion about the identity of works and many of the errors in documentation can be traced back through the 150 year history of the NGV. Some of this confusion is a legacy of the shared early history of the institutions that are now the State Library of Victoria, Museum Victoria and the NGV, with all three institutions previously housed in the one site and governed by a single Council of Trustees. The three major Victorian collecting institutions were formally separated by an Act of Parliament in 1944, but these common beginnings are still evident in the collection management and documentation legacies that curators, registrars and collection managers across all three institutions know very well.

Over the last 150 years, professional procedures and standards for collection documentation have become far more sophisticated. At the NGV, collection documentation, including object receipting and accessioning was first centralised with the Registration department in 1994/1995 following the end of the first stocktake. Prior to this, curators in each collecting area were responsible for receipting works of art, allocating accession numbers and registering works in the collection stockbooks. Some of the inconsistencies that emerged from this were then further entrenched when the sometimes partial or inaccurate information was entered into the first collection database during the stocktake project in the 1980s. This has left the Gallery with a number of errors in the documentation of the early collection that have solidified through repetition, now requiring considerable time and effort to unravel.

The significant improvements in collection documentation that have been made by the Gallery’s documentation project have been an enormous benefit to the stocktake project. Not only have these collection documentation efforts contributed to a steady increase in the proportion of collection information accessible via the NGV’s website; they have also significantly enhanced CMS records, helping the stocktake to confirm the identity of a number of works previously tracked as Found Items. The documentation project has seen two registrars working consistently since late 2006 to update and verify catalogue details in the CMS. The project is addressing the backlog of collection data entry and is aimed at bringing records to the point where they are verified by curators and can be made available to the public via the NGV website (see ngv.vic.gov.au/explore/ngv-collection). To date, 48% of the accessioned collection of 69,155 artworks is digitally catalogued, with records for 27% of the collection available online.

At the end of the reconciliation phase of the stocktake which is due to be completed in late 2013, we intend to be able to identify a number of key areas of concern:

  • Any artworks that haven’t been sighted but that may be tracked as Found Items, the identity of which cannot be resolved because of insufficient documentation;
  • A comprehensive record of all Found items;

For both the first and second points, resolving the identity of these artworks will be a significant legacy project flowing from the stocktake. It is essential that any efforts to resolve the identity and sources of these artworks recognise and incorporate due diligence processes.

  • Any areas of storage that need to be resighted to resolve stocktake findings. This includes areas where there’s been an obvious pattern of location tracking errors that suggest we can’t rely on the validity of the findings.
  • Any works that can be confirmed as not being sighted and therefore should be declared as missing.

Legacies – the future of the stocktake project

With the project still in Phase 1 & 2, the following discussion pre-empts the final phases of the project with the aim of exploring where the approach to date has taken us, where we intend to head in the immediate future, and what legacies this project will leave for the Gallery’s collection documentation and management practices.

Phase 3

Prepare final report, including the formal reporting of any works unable to be accounted for

To contextualise this next phase, I’d like to return to the missing work by Richard Parkes Bonington. The Age notes that the Bonington went missing in 1999, describing the final formal reporting of the loss in early 2011 as ‘leisurely.’

While the Gallery’s Trustees and the Minister for the Arts were promptly notified that the work was not located in early 2003, in keeping with the Trustees’ decision to undertake a full and complete stocktake, the Gallery was only able to declare that the work was no longer in its possession after sighting was complete in the painting screens, as well as all storage areas where this small work could potentially have been stored. This includes all solander box and small boxed object storage. The collection physically stored in this way includes Photography, Prints and Drawings, and large sections of the Decorative Arts, Asian, Oceanic, African, Pre-Columbian and Antiquities collections which make up well in excess of 75% of the collection. The scale of this undertaking can perhaps be best reflected in images of some of the storage areas that had to be sighted before we could begin to confirm the work was no longer held by the NGV.

Once this sighting was complete, all relevant Found Item records were reviewed and cross-checked against the details of Low Tide at Boulogne to find if the small, unsigned and undated Bonington work was being tracked under another number. After exhausting these possibilities, we were able to conclude that unfortunately the Bonington was no longer in the Gallery’s possession. In accordance with the processes we are ethically and legally obliged to follow in such a circumstance, the loss was reported to the Council of Trustees, Victoria Police, as well as the Gallery’s insurers, the Ministers for Finance and the Arts and the State Auditor-General. It was also listed on the Art Loss Register.

By working wall to wall, the stocktake project will be able to identify, document and report artworks that have not been located and are not believed to be tracked as Found items, and therefore can be suspected as missing from the Gallery’s custody. Ensuring we do so this in a rigorous, considered, and transparent manner is essential; as is avoiding unsubstantiated reports of loss or possible theft.

Throughout the stocktake we have aimed to document and confirm the location of works of art and to enhance the standards of collection documentation, contributing to the Gallery’s records and understanding and knowledge of the collection. As this paper offers a chance to reflect, I would like to look at what will be some of the legacies of the current stocktake for the Gallery’s future:

  • Compile a comprehensive record of all works in the Gallery’s care
    As well as improving documentation and verifying locations for works known to be held by the NGV, as I mentioned, the stocktake will leave us a legacy of over 4500 works of art (Found Items) that have never been previously documented. Resolving the identity of these works will be a significant project in its own right, to be addressed following the end of the current stocktake.
  • Identify long-standing documentation and object identification errors
    While in some instances these will be resolved during the project in many others, the findings of the stocktake will point to the need for further work to be undertaken to resolve these issues.
  • Sight works with no previous location records
    A number of works exist in the CMS that have no recorded location. As the previous stocktake was not comprehensive, these works were not previously considered as lost or missing.With its systematic, location-by-location approach, the current Stocktake project has sighted over 100 works that had previously not been located, significantly improving our collection documentation and facilitating access to the collections. Of the works that still do not have a recorded location, after these are cross-referenced against Found Item records, we will be able to significantly refine and more accurately report any artworks that may be missing.

Improve basic catalogue documentation, including enhancing detailed part records in Collection records.

As the stocktake has physically sighted and documented all works in the NGV’s custody and care, this has enabled us to enhance historical catalogue information about the physical details of many artworks, including component parts. This information can then be reviewed and verified by curatorial staff as part of the documentation project, improving collection documentation and providing more detailed and accurate information about the collection to curatorial staff, researchers and the public.

  • Document collection packing and basic condition information
    The stocktake has been able to assess, document and in urgent cases address the packing of all boxed collections, as well as noting the basic condition of works, and immediately reporting any conservation, storage or pest management problems.
  • Identified and gathered data for future documentation and collection management projects
    Found Item and gathering information about documentation errors, we’ve compiled data that will help us to scope and plan future collection documentation projects. Briefly, these projects include:

    • Accession marking –To date the stocktake has identified over 24,000 works (35% of the collection) that are not physically labelled with an accession number or identifying information.
    • Review of existing location tracking and documentation procedures – This will involve reviewing the sighting data to identify areas where significant numbers of works were found in locations other than those recorded on the CMS, evaluating existing procedures and revising our approach to improve the accuracy and reliability of location tracking systems.
    • Collection packing and storage planning Future and on-going stocktaking strategies

Phase 4

Develop a future collection stocktaking policy

With a collection of over 69,000 works of art that has been growing at an average of over 400 works a year, it isn’t feasible, sustainable or even desirable to conduct a stocktake in this way on an ongoing basis. The collection stocktake policy is currently under review, and a new approach to stocktaking is being developed.

At this stage, it is intended that a risk management approach will be applied to stocktake sighting in the future. Individual works or storage and display areas will be sighted according to a sampling size and frequency determined by allocating risk ratings based on the specifics of the artwork and the variable risk factors the work may be exposed to due to its inherent characteristics and current location.

The risk matrix model being developed is informed by thinking around risk management and strategic use of physical and environmental resources. According to this framework, a rating will be applied to each artwork at the time of commencing the annual stocktake. The rating will reflect a range of factors, including the physical nature of the work (its size and media), location, insurance value and significance. This approach looks to factor a range of elements into the evaluation of risk and understands that the likelihood of loss or misplacement needs to be balanced against the inherent risks specific to each work in order to develop a prioritised approach that best uses available time and resources.

With the current stocktake having established a baseline, this approach will strive to ensure stocktaking and location verification is not seen as a ‘one-off’ project, but an integral and integrated part of our ongoing collection management and documentation processes.

More than simply confirming where the collection is, the process of inventorying and stocktaking offers a systematic opportunity to review procedures and strategies for collection location tracking and documentation. This serves to help us to meet standards for access and accountability for the collections in our care, and contributes to the greater understanding of an institution’s collections, and to their sound documentation and management.

By Julie Banks, Associate Registrar, Collection Operations at the National Gallery of Victoria