Who’s responsible for this?

Extracts from a keynote address to the CSAAR Conference Responsibilities and Opportunities in Architectural Conservation: Theory, Education, and Practice, November 2008

Peter Phillips BArch MBdgSc FRAIA RIBA AIAMA MICOMOS
President, Australia ICOMOS

Responsible is a word with two slightly different meanings: one referring to the past (whom can we blame?) and one that refers to the future (what is our duty? – and to whom or what?). This talk describes a few of the things for which the author has been responsible (in both senses of the word) during his 25 years as a conservation architect, including some of the opportunities that have arisen (both those taken and those missed), and some tips for conservation practice.

Being a conservation architect in Australia means dealing with buildings that are at most a little over 200 years old. The Aboriginal people, who were there for 50,000 years or so before European settlement, saw no need to construct monuments because nature had already done so for them, although they certainly decorated their surroundings.

The first substantial structure in Australia, therefore, was Government House at Sydney Cove, the foundation stone of which was laid by the first Governor, Arthur Phillip, in May 1788. The building was altered and enlarged a number of times and eventually demolished in 1845, and it is now only an archaeological site marked by some paving in the forecourt of a modern government building (Figure 1). The oldest surviving building in Australia is Elizabeth Farm cottage, built in 1793 near Parramatta by John Macarthur, pioneer of the wool trade and future rebel against Governor William Bligh of the Bounty.

FIGURE 1
First Government House, Sydney, in 1807 (left) and the site in 2008 (right)
Source: State Library of New South Wales (left) and Orwell & Peter Phillips (right)

In the settlement at Sydney Cove, development proceeded rapidly in the early years of the 19th century. Millers Point to the west of Sydney Cove became Australia’s first residential suburb, developing from the 1820s. Many of those who lived here had maritime businesses which they could oversee from their houses. When these owners moved out to more salubrious suburbs to the east and west, the maritime workers moved in, and many of the houses became lodgings. In 1900 the whole area was taken over by the government’s Sydney Harbour Trust and the shoreline was reshaped and developed for finger wharves. Most of the houses in Millers Point have remained in government ownership ever since, which is why the area has been conserved.

FIGURE 2
Houses in Lower Fort Street, Millers Point, in about 1847 (left) and from the rear in about 1916 (right)
Source: State Library of New South Wales

Orwell & Peter Phillips became involved with Millers Point about fifteen years ago, as one of a number of conservation practices engaged to repair some of these houses and convert them from boarding accommodation (with shared facilities) to individual units (flats). It became apparent that, although the buildings were still going to be used for public housing, dividing them into smaller units was not the best conservation approach, because of the damage one had to do internally to insert new bathrooms and kitchens, and to provide adequate fire separation between units.

Although Department of Housing officers were aware that Millers Point houses were important heritage buildings, they were not at that time giving due weight to cultural significance when deciding how to adapt the buildings for public housing. As each dwelling became vacant, the Department looked at its list of people needing accommodation and decided what to do with the building based on what type of public housing was in greatest demand at the time.

Conservation architects engaged on these projects managed to persuade the Department to adopt a conservation approach – in other words, to make decisions about adapting the buildings on the basis of the type of public housing that would be most compatible with the retention of heritage significance. The Department also began to consider groups of similar buildings (for example, entire terraces) as a whole instead of working house by house, and arranged for a conservation management strategy to be prepared for all of its assets in Millers Point.

What has been described above is one example of a very common situation in a conservation practice, and illustrates that in conservation work the customer is not always right, at least to begin with. The client is, after all, only the temporary custodian of a building that (with appropriate conservation) will outlive both client and architect. In fact, in architectural conservation work the real client is often the building itself.

Events in Millers Point have recently taken another turn, with the Department deciding that it cannot spend so much of its scarce funds on conserving these dilapidated vacant buildings when the same resources could be more effectively used for public housing elsewhere in Sydney. It has therefore commenced a program of leasing some of these buildings to the private sector as single family houses for 99 years, under strict conservation guidelines. The buildings will therefore be returning to their original use as private family houses.

The conservation approach advocated in Millers Point is a good example of working in accordance with the Burra Charter, the Australia ICOMOS charter for good practice in conservation. The Burra Charter is based on its European predecessors, the Athens and Venice Charters, but is a broader document covering the range of issues that can arise in conservation practice in Australia. Although it was not written with any intention of being applied outside Australia, in fact it has been widely adopted and used in many other countries, and translated into other languages including Arabic. The essence of the Burra Charter process is to understand the cultural significance of a place before, and independently of, any other matters that may affect what happens to it (such as the client’s requirements or the condition of the building). The Burra Charter has now been adopted by the Australian national government and all Australian State governments as the accepted standard of best conservation practice.

Returning to Sydney, around the next bay to the west from Millers Point is the suburb of Pyrmont, the northern end of which was acquired in 1875 by the Colonial Sugar Refining Company. The CSR refinery site grew and developed until the early 1990s, and in 1996 it was sold to the developer Lend Lease for redevelopment as housing, and renamed Jacksons Landing. Today, most of the former industrial buildings have disappeared, and have been replaced by a number of large new ones (Figure 3).

FIGURE 3
The CSR Pyrmont Refinery site in about 1994 (left) and the same site (now Jacksons Landing) in 2008 (right)
Source: Adrian Hall in Fitzgerald and Golder, Pyrmont and Ultimo Under Siege (left) and Orwell & Peter Phillips (right)

Some time before the site was sold, a number of the buildings were assessed by the government and listed, and those were the only heritage buildings that were subsequently earmarked for preservation in the approved site Master Plan for residential redevelopment; they included the original Engineers Store, the façade of the Boiler House, the Colonial Rum Store, the Cooperage (where CSR made barrels for their rum), the Gatehouse and original Main Office, and the Tablet House (where they made sugar lumps). One other building that was listed, but later demolished, was the second laboratory building, Laboratory B. A number of other buildings were recommended as worthy of retention in a subsequent conservation plan for the site, including the original Refinery (the first permanent building on the site), Laboratory A next to the Tablet House, and the Char House at the western end of the site.

In its early years, CSR had been a very remarkable company for its time. From the start it retained a percentage of its company profits for use in research and development, and it was one of the first companies in Australia to employ scientists to improve its products and manufacturing methods. Moreover, it was famous for continually increasing the yield from every ton of sugar cane. There were elaborate return pipes throughout the factory to prevent sugar and syrup being lost during production, and CSR made rum from the unusable molasses left after sugar refining, and used the cane fibre left after the sugar had been extracted to feed the boilers, and later made building boards out of it. They also designed and built all their own buildings, plant and equipment, including some of the first automated derricks to load and unload their ships, so that they would not be affected by waterfront strikes.

The work of Orwell & Peter Phillips at the Jacksons Landing site included writing conservation management plans for the heritage buildings to be retained, such as the Tablet House (Figure 4). This building was designed by CSR engineers as a steel and cast iron frame with brick external walls providing both the external envelope and the bracing. This worked well at first, but the steel was not galvanised, and could not be properly protected where it was surrounded by brickwork. The consequent rusting of the frame had gradually deformed the brickwork and bent the cast iron windows out of shape.

FIGURE 4
Heritage buildings on the CSR Pyrmont site: the Tablet House (left) and the Char House (right)
Source: Lend Lease Development (left) and Orwell & Peter Phillips (right)

Other tasks included investigating reuse options for some of the buildings scheduled for demolition, such as the Char House (Figure 4), which was a most unusual and evocative industrial building with much of its equipment still in place. Bone char is what was used until the 1950s as the filtration medium for turning brown sugar into white sugar, and it was made by controlled burning of animal bones. These were obtained from the abattoirs at Glebe Island and later Homebush Bay (where the Olympic Games were held many years afterwards). The timber hoppers, brick ovens and steel grinders were grim evidence of this grisly process. The Char House was located next to an area proposed in the Jacksons Landing Master Plan for public recreation, and would have been ideally suited for adaptation as a site interpretation centre. However, because the building was not heritage listed, Lend Lease had prepared their Master Plan on the assumption that it could be demolished, and had therefore proposed a number of residential units on the land. They were quite keen to reuse the old building but not at all keen to lose the units, and the Planning Minister at the time would not agree to these units being added to the other residential towers. So the Bone Char factory was demolished – another example of how important it is in conservation work to have the heritage issues properly examined early in the project, instead of after all the big decisions have been made.

One other significant item that was only saved by great good fortune was the entire collection of CSR’s drawings. All the drawings that CSR had created since the late 19th century were originally kept at Pyrmont. They documented all of CSR’s buildings and equipment throughout Australia and the South Pacific, including light rail locomotives and track to service sugar cane fields in Queensland and Fiji. The drawings were relocated to another CSR site in Sydney when the building they were stored in was to be demolished by Lend Lease. Most of these drawings had been microfilmed, but several were old paper prints or negatives. The drawing collection was an important primary source for the conservation plans for buildings at Pyrmont, as were the correspondence records that CSR had previously deposited with the Australian National University archives in Canberra.

Towards the end of the project, a routine request for access to the drawings was met with the response that they had been thrown out a couple of weeks earlier because CSR had no further use for them. After a couple of days of anxious inquiry it was discovered that, although instructions had been given for the drawings to be thrown out, they had found their way to various members of the Light Rail Historical Society, who were going through the drawings with a view to pulling out the light rail documents and discarding the rest. The Society was persuaded not to throw anything out, and subsequently (with the assistance of a substantial donation from Lend Lease) to allow the whole collection to be sent to join the CSR correspondence records at the ANU archives after the light rail drawings had been copied.

The Pyrmont refinery site was next door to the quarry where some of Sydney’s best building stone was extracted (Figure 5). Three quarries, known as Hellhole, Purgatory and Paradise, operated at Pyrmont from the 1850s to the 1920s, and many of Sydney’s finest buildings, including the Eastern Range of the Main Quadrangle at Sydney University (completed in 1856) and most 19th century government buildings, were constructed of what became known as Pyrmont yellowblock, a fine-grained white sandstone that turns golden brown after exposure to the air.

Although the stone is of very good quality, like all sedimentary rocks it weathers over time, especially when the air contains salt spray or pollution from fossil fuels. The Government Architect who built the next stage of the Main Quadrangle in the 1890s, Walter Vernon, realised that within a century or so all the sandstone buildings would need substantial repairs, and advised the government of the day to prepare for it. However, as often happens, not enough was done, and when the time came to repair these buildings later in the 20th century, the Pyrmont quarries and most other sources of yellowblock had long been built over, and it was very difficult to find suitable stone.

One of the sites due for development at Jacksons Landing was known as McCaffery’s Hill, close to the location of the old quarries, and the NSW Department of Public Works had done some core tests there and found a seam of first quality yellowblock under the site. Lend Lease agreed to let them take over the site for a few weeks so that they could extract the stone. Because of the need to avoid delaying the development program, traditional quarrying methods would not do. Public Works and their contractor decided to use two giant stone saws mounted on hydraulic arms to make vertical cuts in the stone, and a kind of modified fork-lift to lever the quarry blocks up along bedding planes (Figure 5). This rough handling left the some of the blocks with internal cracking that was discovered when they were cut up later, but it allowed a large quantity of stone to be extracted in a very short time.

FIGURE 5
Stone quarrying at Pyrmont: in 1893 (left) and on the McCaffery’s Hill site in the 1990s (right)
Source: State Library of New South Wales (left) and Orwell & Peter Phillips (right)

Another stone much used for buildings in Sydney from the late 19th century was known as Bowral trachyte, actually a microsyenite found in the highlands about 100 km south-west of Sydney. Many of the kerbstones in the city of Sydney were also replaced in Bowral trachyte in the late 19th century, and most of those are still there.

One of many buildings to use Bowral trachyte was Railway House, the former headquarters of NSW Railways. It was built in 1936 just after Sydney’s underground railway was completed, with a façade that had trachyte at the base and green glazed terracotta above. When the building was extended in the 1970s, the original entrance and adjoining window were removed to create part of a colonnade. In the 1990s Railway House was listed on the State Heritage Register and subsequently sold by the government. The new owner engaged Orwell & Peter Phillips to provide conservation advice to help obtain development approval. He and his architects were intending only to repair the surviving original building fabric, but were challenged to go beyond that and enhance the presentation of the conserved building by reconstructing the original entrance and window, for which there was good documentary evidence. However, it was discovered that, although the trachyte column facings could be recycled to rebuild the window opening, there was not enough left over for the entrance, and more stone would need to be found – but there were no sources of stone other than the original trachyte quarry at Mount Gibraltar near Bowral, which had closed many years before and was now a nature reserve.

Because trachyte was such a tough stone, the traditional quarrying process was to drill down behind the quarry face, place explosives in the holes and blast the whole of the face off. Once the dust had settled, the quarrymaster would select pieces of sufficient size and quality to be used for building stone, leaving the rest on the quarry floor. A visit to the old quarry was therefore arranged, in the hope that there might be some discarded pieces there that could be recovered. Fortunately, it was found that some of the smaller blocks rejected by the quarrymaster were still big enough for the purpose, and these were able to be retrieved so that the reconstruction could be completed (Figure 6).

FIGURE 6
Trachyte: the quarry face at Mt Gibraltar (left), a remnant quarry block being measured by the mason (centre) and the reconstructed entry to Railway House (right)
Source: Orwell & Peter Phillips

The practice of disposing of public buildings to the private sector has been going on in Sydney and elsewhere for several years. The General Post Office in Sydney was built by the Colonial government in stages from the 1860s, taken over by the Commonwealth at Federation in 1901 and extended twice more in the first half of the 20th century before the Commonwealth decided in the 1980s that it was no longer viable. The GPO is now a hotel and offices, with a small post office in one corner.

Like most 19th century buildings, the original GPO had a narrow floor plate surrounding an internal courtyard to provide natural light. In the centre was a fine cast iron Imperial stair in a tiled hallway that projected into the courtyard at the back. In the 1920s, a new building was erected inside the courtyard with a new postal hall at ground level, and the back of the old main stair was cut off. The developer who had bought the lease of the GPO wanted to restore the courtyard and the main stair, and he did, incorporating the reconstruction date in the decorative plaster. Because the old postal hall was an important space, the outline of it (incorporating the original glazed roof) was also reconstructed, leaving the stair projecting into it. The cut off rear part was reinstated with a modern handrail in approved Burra Charter fashion, but apart from that it looks much the way it did (Figure 7).

When the NSW Heritage Council approved the proposal for the conversion of the GPO into a hotel and office tower, it appointed a sub-committee to implement the consent and approve the details of work packages as they were completed. They also persuaded the developer to pay for a conservation architect to advise the sub-committee, and appointed the author to that position. This arrangement both improved the conservation works and saved the developer time and money. When the 1920s building was demolished to create the new courtyard behind the original stone building, it was revealed that the stone facades facing the courtyard had been quite badly damaged in the past, with projecting courses hacked off and one level cement rendered. The developer had intended to reface most of the internal walls in new stone so as to improve the appearance of the courtyard, at least partly because he thought the heritage authorities would insist on it, and had allowed a considerable amount in his budget to do so. However, it was agreed (without much need for persuasion) that the façade should be only partly repaired because this would better allow people to read the history of the building from its fabric, and would at the same time reduce the cost by about AU$1 million (Figure 8). This was a good example of the Burra Charter rule that in conservation you should do as much as necessary, and as little as possible.

FIGURE 7
The central stair at the Sydney GPO as originally built (left), and as reconstructed, in 2008 (centre and right)
Source: National Archives of Australia (left) and Orwell & Peter Phillips (centre and right)
FIGURE 8
The central courtyard at the GPO as originally built (left) and after reconstruction, in 2008 (right)
Source: National Archives of Australia (left) and Orwell & Peter Phillips (right)

The project engineers presented more of a challenge. Engineers are accustomed to working with standard codes, which are mostly derived from mathematical models based on assumptions about how materials and structures behave. These assumptions may not be altogether accurate for older materials and methods of construction. (With earthquake codes, for example, it is sometimes possible to prove by calculations based on modern codes that a historic building should have fallen down long ago, even though the building may have been standing happily for hundreds of years.) The uncritical use of modern codes may therefore lead to over-design and unnecessary intervention in historic building fabric.

The original GPO was built in accordance with the best engineering practice of its time. Except for the ground floor which was mass concrete, and the top storey which was a late 19th century addition, the original building had a floor-ceiling construction with three distinct layers. The main steel structure supported the ceiling timber framing below, an intermediate layer of shallow lightweight concrete vaults for fire separation, and the floor timbers above. During redevelopment, the project engineers concluded that the old building had insufficient fire separation between floors, and was also structurally inadequate to resist earthquake loads. The services engineers also wanted to pull up the existing floors to install new services, particularly plumbing for new hotel suites. The engineering team therefore proposed to replace all of the timber floors with concrete floors, to solve all their problems at once. The concrete vaults themselves were cracked in places, so the engineers decided that they needed to come out because they might collapse in the event of an earthquake or fire. From a conservation point of view this would have meant a serious loss of the original building fabric.

The second floor had originally been an office for lower grade staff, and the window sills had been placed well above the floor presumably to prevent staff gazing out instead of working. Because this floor was now to be used for hotel suites, the architects proposed raising the floor level to give a more normal sill height. This was the inspiration for the conservation solution, which was worked out at a meeting of all of the engineers with the conservation team. Approaching the problem from first principles rather than by reference to codes, the structural and fire safety engineers realised that for their purposes only every second floor needed to be concrete – this meant that the new raised second floor could go in above the old floor without disturbing anything, and only the later top floor needed to be replaced. The services engineers then agreed that the multi-layer floor system on other floors would allow adequate room to install services by taking up only selected areas of flooring. The structural engineers were then encouraged to undertake a finite element analysis of the concrete vaults, and having done so they agreed that only a small number of very shallow vaults needed to come out, and the rest could remain and have steel mesh catch nets installed under them in case the worst happened.

This first principles approach to engineering had been adopted several years before to deal with other problems of fire safety in historic buildings. A series of fires in old buildings had led to a government requirement for all such buildings to be inspected, especially those in which people were sleeping, and if necessary the owners were issued with orders to upgrade provisions for fire safety. For heritage buildings such an order often meant the loss of significant interiors or even the demolition of the building if the owners could not afford to upgrade. The Institute of Architects’ Historic Buildings Committee convened a seminar on this topic, one outcome of which was the establishment by the Heritage Council of NSW of a technical panel, called the Fire Advisory Panel. At the time, the building regulations in New South Wales were extremely complex, controlled by several different authorities with different and in some cases conflicting prescriptive requirements. The Heritage Council therefore assembled a panel with representation from all of these authorities, with other representatives from building research centres and the fire protection industry, and a few conservation architects to comment on the heritage issues.

As well as advising on individual buildings, the Panel was keen to develop guiding principles and generic solutions to common fire safety problems. One of these was the typical replacement of timber panelled doors in old hotels and apartment buildings with solid core doors that were supposed to have improved fire resistance. The Panel commissioned research into possible ways to upgrade the old doors rather than replace them, and developed a method that treated the room side of the door while leaving its appearance from the hallway unchanged. The furnace testing of this door and a typical solid core door revealed that the upgraded door could achieve a 30 minute fire resistance, which was equal to the standard rating of the solid core door (Figure 9). It was also discovered that, because of changes in manufacturing methods, the typical solid core door being specified to replace a panelled door now had a fire resistance of less than 15 minutes. Fortunately, Australia now has a performance-based building code, which the approach of the Fire Advisory Panel may have helped to bring about.

FIGURE 9
Upgrading of a timber panelled door (left) and fire-testing the upgraded door and a solid core door (right)
Source: Heritage Council of New South Wales

The scope of the Panel’s work has now been expanded to deal with access and services issues in historic buildings, and it will probably need in future to turn its attention to other matters such as occupational health and safety, and environmental performance. This last is perhaps the single biggest responsibility now facing the architectural profession, but also one of the greatest opportunities for changing people’s minds about why they should preserve old buildings. It should be remembered that buildings began as environmental modifiers, and that there is a great deal of embodied wisdom about environmental performance (as well as embodied energy) in historic buildings. The conservation of historic buildings is truly an essential component of a sustainable future.

Lastly, Figure 10 shows a small recent example of the author’s work – the conservation of a little timber building on a suburban railway station, which had been so badly damaged by termites that in places it seemed that only the paintwork was holding the building up. It is a good illustration of the principle that however much work one has to do to a place, in the end the building should look almost as if one had not been there.

FIGURE 10
Homebush Station before (left), during (centre) and after (right) conservation works in 2008
Source: Orwell & Peter Phillips
  • Peter Phillips
    LFRAIA RIBA
    MICOMOS
  • Suite 1504, 109 Pitt Street
    Sydney, NSW 2000
    Australia
  • Tel 02 9221 8882
    Fax 02 9221 8887
    Email peter@opp.net.au