The Third City: Sydney’s original monuments and a possible new metropolis

How did Sydney’s original dwellers build? What has happened to their structures? And what should happen to the tracts of now-obsolete bungalows sprawling across the suburbs? Peter Myers provides historical evidence and the design strategy for a future metropolis.

This is an article from the Architecture Australia archives and may use outdated formatting

What if our present historical city was not the first urban structure to occupy the coastal region extending from Port Stephens to Kiama? What would this primordial city reveal, what lessons of history could we learn; just what of this first, prehistoric, Sydney was admitted, and what was denied, in the making of our second, historic, metropolis?

Well, proving a vast and ancient First Sydney is, like Newton’s falling apple, ridiculously easy. I cite Newton because it was the collapse, in 1991, of a section of lime plaster cornice in our circa 1853 Blacket villa, that led me to, maybe, the First Fleet’s best-kept secret.

Meaning in the Shells 

Previously, in 1974, while working in Arnhem Land, I assessed the suitability of a local cement aggregate brick for domestic construction. Unwittingly, the coarse material for this sample had been obtained from a previously undisturbed system of shell middens, some 100 metres wide and extending 20 kilometres along Manakadokajirripa’s dunes, east of the Blyth River estuary. I obtained two of these bricks and sent one to Professor Rhys Jones at the Australian National University for analysis. Jones subsequently confirmed that his sample, being made from midden materials up to 7000 years old, was indeed the world’s oldest brick containing a machine-pressed ‘frog’.

Meanwhile the plasterers ran a new section of the missing cornice and I, for some reason, kept the original fallen fragment. Then, years later, it hit me: this cornice section contained still unburnt shell fragments! These had to be shell fragments, they could not be partially slaked chips of quarried rock lime, because they had not subsequently hydrated in-situ.

So if Edmund Blacket had specified shell lime for the construction of our house, where did he get it from? The answer is that not only Blacket, but every builder in old Sydney town, obtained building lime from the region’s approximately 200 shell lime kilns. So we, or rather our predecessors, burnt the shell monuments of the prehistoric or First City, in order to construct the present historic or Second City.

So much for Terra Nullius. The First Fleet shipped no building lime; it was assumed that limestone deposits would be quickly found and that, in the interim, buildings could be constructed from the dense coastal forests noted by Cook. But the British did not anticipate white ants, so their hurriedly constructed timber buildings just as hurriedly collapsed and, bereft of building lime, the first attempts at brick and masonry construction could not withstand even a summer shower. Phillip, by all accounts an enlightened Governor, was now forced to exploit the fabulous shell monuments lining the inlets and estuaries of Port Jackson, Botany Bay and beyond.

I use the term ‘shell monuments’ deliberately; to describe them as ‘kitchen middens’ or ‘discarded refuse’ is a limited vision of their true intent. There are recorded sightings of shell monuments 12 metres high along the water’s edge (perhaps significantly, that is equivalent to the height of the southern podium of Jørn Utzon’s Sydney Opera House). Can you imagine how many thousands of years of gathering and accumulation went into their making? So Phillip, whatever may have been his aesthetic, was forced to destroy the urban framework of a by-now tragically depleted Aboriginal population.

It was really an impasse of refinement, where neither the citizens of the First City, nor the British insurgents could conceive of the other’s sensibilities. This fearful dilemma is most poignantly, if unintentionally, portrayed in the title vignette to Hunter’s journal.

I think that Sydney’s ancient locals just couldn’t take another minute of it! After all, the British were burning not only their past, as represented by their shell monuments, but also were – for their most significant buildings, such as Fort Macquarie – consuming the live shell beds, because these were known to make the strongest possible lime mortar. So, now deprived of their past and their future shellfish resources, the First City’s Aboriginal communities found themselves without a present. Indeed, it is prescient that the first and largest shell lime kiln was on the eastern shore of Bennelong Point: not so much Phillip’s Arcadian Farm Cove as remnants of the First City’s shellfish Farm Cove! If only La Perouse had prevailed; if only because the French had a long tradition of architectural treatises, going right back to Philibert de L’Orme (1500-1570), wherein the technology and merit of pisé construction are described. Only several years after 1788, the immortal François Cointeraux’s first earth building pamphlets were published. Although gatherings of Cointeraux’s pamphlets are known to have been in America before 1800, we must also acknowledge Georgian Britain’s grinding francophobia, which alone guaranteed that this amazing and rational technology would never sully their new colony at Port Jackson. But to return to our place, the entire house is sandstock brickwork laid up in shell lime mortar. I had no trouble in finding bricks with intact mortar beds containing whole cockle shells. The argument (first put by Phillip2 ) that the so-called ‘original’ population was small – around 1500 between Botany and Broken Bays – only confirms the antiquity of the first Sydney, because there is no way that this much lime could have been extracted between 1788 and 1888 from less than one million cubic metres of shells, dead or alive.

Second Sydney

Thus the construction of the second, historic Sydney was really a ‘reworking’ of the city fabric that Phillip discovered in 1788. Brick clays were discovered by clearing away managed forest cover, while quarried sandstone caves and outcrops would have been adorned with rock engravings at least equal to the fabulous surviving examples recorded by W.D. Campbell a century later3 . The beginnings of any colonial enterprise are proverbially insensitive to local cultures, so why should we be surprised that Phillip ignored, or chose to ignore, the very definite indications of an existing urban structure?

The fact that any substantial colonial pastoral holding had its homestead sited exactly on top of a former Aboriginal campsite should be proof enough that the same siting rules applied to historic Sydney. Thus, Bennelong Point was a most important Aboriginal site, upon which a number of very large shell monuments once stood, and which we can also be sure was overlooked by an exquisite range of rock engravings, just precisely where Edward Blore’s Government House was subsequently located.4

The same applies for all the significant buildings and places of historic Sydney. Do we seriously believe that we ‘discovered’ the rock swimming pools along Sydney’s beaches, or that the few remaining sandstone ledges along Woolloomoolloo Bay have only recently been frequented by local Aboriginal families, reminiscing as they scan the water surface for rushing small fry? Every important colonial building was placed upon a significant First City site, and that goes for both Sydney University and Gladesville Asylum, for example: if you consider their sightlines, and then look at the map in Campbell, it is obvious that these sites have been occupied since remote antiquity.

The vast suburban grids of the Second City also continue traditions of the First. The benign environmental conditions that favoured a decentralised urbanism for the First City also easily accommodated the English colonists. An effortless outward expansion of the Second City was thus foreshadowed, with surveyors carefully dividing Port Jackson’s hinterlands into regular quarter-acre subdivisions. The subsequent housing stock is, even today, endearingly temporary, while the contained landscape is appropriately massive – just like it was in the ancient past.

Sydney’s Second City is probably the largest urban system ever built from, and upon, an existing fabric. Certainly no historical city in the Americas is so directly constructed from the urban structure of a preceding civilisation, nor made with a more culpable silence. Our region, from Port Stephens to Kiama, has always been a relatively low-density conurbation, with Port Jackson as its locus: all we actually did was to replace one economic landscape with another, substituting one form of dispersed, temporary dwelling with another.

Certainly, the introduction of, for example, corrugated iron sheeting, by Henry Parkes in 1847, greatly aided the colonists in their adaptive settlements. However, I now believe that these frontier materials are important only insofar as they emulate those technologies, such as bark sheeting, that were then used by the Aboriginal communities to make their temporary, moveable shelters. Similarly, the Garden City movement is often cited as the catalyst of Australian suburbia, yet it is worth considering, via Norman Tindale’s magnificent overprinted maps,5 just how diligently we have supplanted Aboriginal Australia’s language densities with our own dispersed populations.

Recognising the Suburban Landscape 

Perhaps the original cultures of this continent are still the most pervasive ‘shapers’ of our rapidly decentralising urbanism? Perhaps we Sydneysiders are indeed guided by some intuitive recognition of the now subsumed First City, thus becoming more aware of our responsibility for what has occurred? Or, as the philosopher Ross Poole proposes: “We are only free to the extent that all others are also free”, and so we yearn for a regenerative urbanism; an urbanism that will again embrace Sydney’s ancient cultural landscapes.

Taking the train west from Sydney’s Central Station, one is soon speeding above what seems to be a continuous forest cover. From Ashfield to Blacktown, there is a dense canopy of advanced eucalypts, brushboxes and the occasional introduced evergreens of the doctor’s big house garden. Alighting at Blacktown and strolling up Flushcombe Road, I am astonished by the vistas north and west, beyond the K-Mart foreground, across the most beautifully tended forest canopy, for as far as one can see. What is this vast work? Is this the key to the suburban dream? Does this close-nurtured forest represent our highest urban ideal? Have we perhaps got the wrong end of the stick in our belated attempt, via the all-too expedient dual occupancy housing initiative, to defend our bungalow suburbia as the epitome of the ‘Australian Way of Life’?

I believe so. I propose that we forthwith recognise the landscape inventories of suburban Sydney for what they are: works of art of the most profound importance, and that we incrementally abandon the hundreds of thousands of no longer functional, worn-out, suburban cottages to their Second City obsolescence!

Such a view is entirely contrary to that held by the so-called heritage industry, intent on not only the conservation and/or conversion of these now redundant cottages, while also actively campaigning for their status as prototypes from which, they maintain, a hybrid of smaller subdivisions and multiple occupancies will somehow deliver us a new suburban paradigm.

Despite such sentimental reverie, dual-occupancy zoning and the Landcom-inspired craziness of ever-reducing block sizes is starting to bite. Magnificent suburbs, built by the Housing Commission in the immediate post-war years as manifestly utopian communities, complete with houses, factories, roads, shops, schools and railway stations are now to be jettisoned by the Labor government under the cynical rubric of a ‘cost-neutral’ public housing policy.

The intactness of Seven Hills, for example, was something most carefully conceived by committed civic planners and then nurtured by successive generations of Commission tenants. Street after street of immaculately kept roadways, with no front fences and no concrete footpaths and continuous canopies of native trees create a friendly, intimate scale. Consort to the gently radiused street curtilages are magnificent mature eucalypts, generally two per house and set behind in the dormant interior of each block.

These eucalypts are both the genesis of the Third City and a passionate link to the First. For each cottage, the original occupants were given two tube-stock eucalypts from the Housing Commission’s nursery. The choice of where to plant these trees was left to each tenancy, though invariably it was the interior court formed by the six foot paling boundary fences that was chosen. This court, which, with self-deprecating affection, we all know as ‘the back yard’, was defined on its fourth side by the rear elevation of a Commission cottage, offset to accommodate a motor car in a garage-cum-shed at some future date. This land-and-cottage ensemble was more or less standard and always set at twenty feet from the street boundary, in uninterrupted grids of 1000 by 300 foot blocks containing 40 cottages each, with street reservations of one chain, or 66 feet. It is as utopian as you could wish for; it is the First City recast for the returned service Everyman, and it is, in its incredible landscape inventory, a communal work of art of an hitherto undreamt of scale and perfection.

 What happened with these tube-stock eucalypts is a good lesson in close nurturing. The paling fences and single storey cottages provided both a wind break system and ground conditions free of complex evolutionary vectors; the seedlings could surge along in the deep reactive clay lenses of western Sydney, and so reach maturity in less than fifty years. So, from Flushcombe Road, one surveys a prospect of uninterrupted forest cover, made almost entirely by first generation Housing Commission tenants. It is a very different landscape from the long-established streets of the upper North Shore; the species range at Blacktown is much more like the First City’s original forest cover, and it is so defiantly civic, rather than private, in its purpose and accessibility. Now in their first maturity, we can only guess just how mighty these trees will become in the next millennium, if only we can persuade the State to abandon its current ‘cost-neutral’, that is ‘flog it off ’, housing policy!

However, in order to defend these now established west Sydney landscapes against further subdivision, it was necessary to study, in-situ, some examples of this urban forest public housing. Fortuitously, I won an architectural contract to rehabilitate a group of first generation (circa 1946) Commission cottages near Seven Hills station. To my amazement, I then found that no architect had ever evaluated the Department’s now ageing bungalow stock; any remedial works had always been done by maintenance contractors, without consideration for even the simplest of functional modifications.

Even so, it was surprising to find kitchen counters at 800mm height, when every appliance manufacturer and Australian Standard specifies 914mm benches. Similarly, kitchen benchtops were only provided at 450mm width, not 600mm as is now normal in new housing; so now nothing fits and it is impossible, moreover, to widen these kitchen benches without further reducing the adjacent and already cramped banquette dining alcove.

However, it is the mould-blackened ceilings which are the most critical design issue. None of the Commission’s 1940s cottages have opening windows above door-head height, 2100mm, though ceilings are a generous 2700mm throughout. As a result, all the ceilings and walls above window head grow a hideous black mould which requires painting out at least once a year; though several households I studied had given up and sealed off those rooms not in regular use. To rectify this functional deficiency is not at all complicated; the few cheap vents and plastic ducts required are readily available from any local hardware store. The real difficulty is that the now funding-strapped Department of Housing owns many thousands of these cottages, all with exactly the same deficiencies.

Until the early nineties, the Department always built ahead of the market, so the functional shortcomings of their stock are only furthered in the ‘open market’ speculative cottage tracts. It is not much better with the inter-war stock; the brushbox-canopied streets of, say, Punchbowl or Ashfield are encumbered with similarly dysfunctional dwelling types. In a ‘big picture’ sense, all that ever really mattered was the capacity of all these modest cottages to nurture the enormous, maturing urban forest resources of the Second City’s suburbia. So, keeping these whacky little houses as precinct archetypes, while also advocating dual-occupancy subdivision, is a policy madness that will certainly confirm our latent incapacity to learn from the lessons of history. Why destroy these urban forests, which are inherently good only to replace them with dinky dual occupancy streetscapes, which will only ever be merely interesting? As Ludwig Mies van der Rohe famously proclaimed: “I do not want to be interesting, I just want to be good!”

Consequently, for the Third City there is only one simple, irreducible rule: To retain, in the name of the people, the landscape inventories of metropolitan Sydney in their present entirety, and in perpetuity.

Tomorrow’s Suburbs: Another Strategy

OK, so what will it look like, this new city? Well, the Third City will be nothing like the present bungalow grids, or any of those comfortable old-world archetypes either. This will be a city predicated on temporary housing in a permanent landscape and so, hopefully, this new metropolis will regenerate the First City’s diaphanous matrix of landscape and shelter. Consequently, each new housing precinct would occupy only the ground plan zone of the former bungalow tracts. ’s new housing would be slender in cross section and from four to seven levels high, 21.5 metres overall, with a 6.1 metre setback, so as to match neighbouring Second City cottages. This slender housing would be required to step down at its junction with existing housing stock, to minimise overshadowing and preserve neighbourhood privacy. However, structural provision would allow for the full cross section to be completed and  continued later. This new building type would be constructed using screw-pile technology, thus enabling both Sydney’s extensive reactive foundation conditions to be better used, and for these new buildings to be entirely removed at any time, and the site completely rehabilitated. This structural system also allows for stormwater retention pipes to be placed beneath each block, and for the water so retained to be slowly released into these now vast internal landscapes and/or allotment gardens, vineyards, fish farms, hydroponic legume beds, etc.

This housing would be designed to give complete sightlines from each apartment into the street and landscaped courts, so providing the level of interest and viewing that is mandatory for sophisticated urbanism, and where no kitchen sink need face a blank wall.

To be sure, innovative public transport systems would soon be constructed along the now densely built street grids. On that basis, this new housing approach only requires an area of four by three miles, or 6.43 by 4.83 kilometres, to house one million people, complete with a one-by-one mile ‘Central Park’, and all this without cutting down even one existing tree. So light rail will certainly be there!

However, it is the catalyst of the existing urban forest which can actually give us this new, very radical Third City. Other than to define a possible cross section, I do not want to anticipate the architecture that could be engendered. In this diffidence, I can only cite Charles Percier and Pierre-François-Léonard Fontaine’s 1801 cross-section for the Rue de Rivoli; it is very much the generative idea for modern Paris – a deceptively simple building section, without which Baron Haussmann could not have even begun.

To journey across Sydney’s three cities, it is worth contemplating William Maddock’s panoramic photographs, taken from the spire of St James, Queen’s Square, that illustrate his 1872 Sydney guidebook. Each of these complex urban vistas would have required something like a quarter of a million cubic metres of shells to be burnt for the lime for their construction; that is a volume equivalent to the monolithic 1959 platform of Utzon’s Sydney Opera House. The complete panorama was held together with only the lime extracted from this immense quantity of shells, so patiently gathered into monumental assemblies by the First City’s women and children over all those thousands of years. This is a daunting reminder that we have really only just begun a long quest to understand Sydney’s devastatingly beautiful landscapes.

Peter Myers is a Sydney architect in solo practice since 1970, specialising in public housing, architectural conservation and projects for Aboriginal communities. This paper was originally delivered at the Republic of Ideas seminar held at Artspace, Sydney, on 30 October, 1999.


1. See God Is In The Details by Joanna Susan Capon, MA thesis in historical archaeology, University of Sydney 1988, for an excellent study of ‘The Development of Non-Metallic Cornice and Ceiling Enrichments in New South Wales 1788-1900’.
2. Mackaness, George, 1937. Admiral Arthur Phillip, p. 140.
3. Campbell, W.D, 1899. Aboriginal Carvings of Port Jackson and Botany Bay.
4. The sandstone outcrop where Blore’s Government House now stands can be seen in the oil painting catalogue No. 9, Sydney: Capital New South Wales, circa 1800, artist unknown, in the Picture Gallery at the State Library of NSW.
5. To accompany his 1974 book, Aboriginal Tribes of Australia, Professor Norman Tindale overprinted Tribal Boundaries in Aboriginal Australia onto the standard four-sheet map of the continent then produced by the Department of National Development in Canberra.



Published online: 1 Jan 2000
Words: Peter Myers


Architecture Australia, January 2000

More archive

See all
The November 2019 issue of Landscape Architecture Australia. November issue of LAA out now

A preview of the November 2019 issue of Landscape Architecture Australia

The February 2020 issue of Landscape Architecture Australia. February issue of LAA out now

A preview of the February 2020 issue of Landscape Architecture Australia.

The May 2020 issue of Landscape Architecture Australia May issue of LAA out now

A preview of the May 2020 issue of Landscape Architecture Australia.

August issue of LAA out now August issue of LAA out now

A preview of the August 2019 issue of Landscape Architecture Australia.

Most read

Latest on site