San Francisco-based author Stewart Brand has written two of the most influential books of the second half of the 20th century. The first, the late 60s Whole Earth Catalog is deservedly famous as it trailblazed ‘access to tools’ for social and environmental alternatives. The second, How Buildings Learn - What Happens After They're Built, written in 1994, is less well known, but more relevant for today's architects.

Brand's thesis is that all buildings change after they’re built: they grow, are altered, amended, added to, maintained (or not), and understanding these changes over time is critical in designing a good building in the first place. He argues, in the book, and even more convincingly in a 1997 six-part BBC TV series, that non-architectural buildings, which he calls the low road, offer the greatest adaptability and that architects should pay much more attention to these successes in designing new buildings.

He sees changes in buildings at different rates in six ways (with acknowledgement to the work of English architect Frank Duffy of DEGW). They are site (slow to change), structure, skin (increasingly changeable), services, space plan, (ever more changeable) and finally stuff, (FF&E as we call it), the most likely to change.

Brand’s book outlines a litany of failures by architects in not heeding these changes: arguing they spend more time on the exterior than the interior (because the brief for uses is downplayed or ignored); they see buildings as fixed and final objects, inflexible and unadaptable; and they never go back to learn from what really happens in the life of the building. No post occupancy surveys are ever read.

He accuses architects of being obsessed with facades, (in which he identifies the etymology of both ‘face’ and failure’), and he excoriates the hero worship of the hero image in architectural magazines. Rather, he lauds the vernacular, such as found in Bernard Rudofsky's seminal Architecture Without Architects (featured in the Whole Earth Catalog) and his friend Lloyd Khan’s book Shelter, which closely followed the Whole Earth Catalog’s concerns and format.

Essentially Brand argues that if we look more closely at vernacular buildings, and what changes occupants make to buildings when they are allowed to, we can better understand how to design for current uses as well as future flexibility. His evidence came from analysing hundreds of buildings in historical and contemporary photos, mostly in the USA in the book, and in the UK and more widely in the TV series.

Stewart Brand’s ideas challenge some of the very fundamentals of architecture, and if adopted can make you think very differently about the design of buildings. I’ve long been fascinated by this inversion of the conventional wisdom; to expound on some of the principles I would like to examine some terrace houses, a staple house form of inner-city Sydney and Melbourne.

Terrace Houses

 

Essentially Brand argues that if we look more closely at vernacular buildings, and what changes occupants make to buildings when they are allowed to, we can better understand how to design for current uses as well as future flexibility. His evidence came from analysing hundreds of buildings in historical and contemporary photos, mostly in the USA in the book, and in the UK and more widely in the TV series.

 

This is an aerial view of six ‘terraces’ in Hargrave St in Paddington in Sydney, typical of the typology: two-storey conjoined buildings on long narrow sites with rear lane access. Built by developers in the late 19thC, they were rented out to workers or middle-class families; later sold in the early 20thC as individual mortgages became more common; then changed as successive owners converted them for bigger families or as boarding houses; then declined as cheap rentals to artists and students; and more recently came back to life as houses for smaller families.

The Fronts

The original front was well-dressed in rendered details over brick, cast-iron ‘lace’, finials and palisade fences; whilst the service rooms at the rear (kitchen and outdoor ‘dunny’) were functionally plain. The Australian idiom in the vernacular of the ‘showroom and the shed’.

In many terraces the front verandas were often enclosed, fibro and aluminium replacing lace and balustrades upstairs, and in some cases, as here, downstairs in brick as well. This change is a vexatious issue is for today’s architectural historians. The front veranda has been a stalwart architectural feature of what makes our houses ‘Australian’, not least since their florid celebration in J.M. Freeland’s Architecture in Australia. They protected the soft rendered brickwork behind and provided much needed shade from the harsh sun, particularly when facing north (or NE as here).

 

Essentially Brand argues that if we look more closely at vernacular buildings, and what changes occupants make to buildings when they are allowed to, we can better understand how to design for current uses as well as future flexibility. His evidence came from analysing hundreds of buildings in historical and contemporary photos, mostly in the USA in the book, and in the UK and more widely in the TV series.

 

But the conversion of thousands of them to internal spaces (sometimes bathrooms) showed that extra space was preferred over the delights of climate control, particularly as occupancy increased during and after the depression of the 1930s (when many were converted into boarding houses). Most were modified illegally by the owners without any approval from Council but are now tolerated as ‘extant works’.

When Paddington began to be gentrified in the late 60’s the Council’s guardians of taste preferred that the rough DIY infill was converted back to the ‘original’. And so did the real estate agents seeking a strong ‘first impression’.  But not all owners saw value in losing the space, and so the original uniform run of fronts sometimes resemble jagged teeth as some additions were removed and others stayed (as in our example).

Most inner-city Councils’ have an increasing obsession with controlling appearances and would prefer that every house revert back to the original form, in uniform ‘heritage colours’. You’d expect that the authorities would be encouraging owners to remove the fibro enclosures to remake the original façade and return the terrace houses to their former ‘glory’. But not so.

The City of Sydney (which presides over vast swathes of terraces in south Sydney, as the most over-regulating council in Australia) now require a detailed development application for façade restoration, with detailed drawings, a planning statement and a heritage report. The costs of reports are almost commensurate with the cost of construction. No wonder some owners just give up. Or do it without permission.

When finally approved and built the result is usually an ersatz version of the original: aluminium not cast iron, laminated beams, not hardwood, reconstituted timber not real. What happens next, out of the gaze of the certifier, is the illegal installation of sliding glass doors and windows behind the replica lace and balustrades so the owners can have the ‘best of both worlds’ and can ameliorate the the sudden rise in noise they discover when the enclosure disappears.

Following Brand’s contrarian analysis, I would recommend that owners of a terrace with an enclosed space, keep it enclosed if they wish to, and resist the gentrification. The space is much more valuable in many cases than a little used veranda, and it acts as an acoustic barrier to increased vehicle noise. You don’t need a Council approval for maintenance and repairs of an enclosed balcony, so owners can remove the fibro (mostly asbestos - with precautions) and replace it with similar but better performed materials. And paint it in heritage colours if you must.

I know this is heresy for the exterior-obsessed archi-history industry, (which is half the fun of it) but often it makes for much better amenity and internal space for the occupants., which is what the house was built for.

The Backs

If challenging changes were made to the fronts, even more radical ones were made to the rears. Here are two photos taken almost 50 years apart.

 

Essentially Brand argues that if we look more closely at vernacular buildings, and what changes occupants make to buildings when they are allowed to, we can better understand how to design for current uses as well as future flexibility. His evidence came from analysing hundreds of buildings in historical and contemporary photos, mostly in the USA in the book, and in the UK and more widely in the TV series.

 

I took the one on the left for a class with the redoubtable Ross King, then head of the Ian Buchan Fell Housing Research Unit at the University of Sydney. King advocated that the best solution to the housing crisis is to have everyone own their own home. Everyone. It’s an idea that never left me and seems a prescient solution 50 years on. And Stewart Brand would approve; he advocates for everyone to make their home as their home. Adaptation and maintenance are vastly improved with home ownership.

The 1974 photo shows just that idea, as if every owner had to add a lean-to. Do it yourself add-ons to the original have created lean-tos against lean-tos against lean-tos, almost to the back fence. Certainly, there is a loss of light and ventilation, but it’s also an expression of the demand for space in these tiny houses. And I dare say none of it had planning or building approval.

There is a marked difference in the changes to the formal public front and private back. At the rear the DIY additions made in the first half of the 20thC are not up, not out, but along the ground, and then it changes even further over the next 50 years, a perfect example of Brand’s idea of how buildings learn, evolve and change over time.

In the interregnum between the first and second photo, local councils (here it is Woollahra) have sought ever more stringent controls over the design of alterations and additions, but not all the changes are the result of council approvals.

The roofs of the houses when built were presumably ‘corrugated iron’ (being too late for shingles and too early for tiles). At some stage the roofs rusted out, or blew off, and so by 1974 most of the set of six had replacement roofs of the Marseille pattern terracotta tiles, which had become a preferred material of prestige over the reviled ‘corry iron’.

Following the April 1999 hailstorm in Sydney, many of these tile roofs were quite literally shattered, and were replaced with corrugated steel, coming full circle. In areas of high likelihood of hailstorms, we now know it is madness to put up brittle terracotta or concrete tiles. But that is exactly what was required by heritage mandates in some jurisdictions, demanding that tiles go back ‘like for like’ and being more concerned about heritage ‘correctness’ and the external appearance than learning a lesson about materiality.

We can see that a dormer and skylight have been built into the rear roofs. Dormers are not usually permitted in the front, where they have better light (to the NE here) and views. Rather, Councils require them to be at the rear, ‘not seen from the street’, but ironically, they intrude on the privacy of the rear gardens either side. Another case of premiating the ‘man in the street’ over the amenity of the occupants.

One last comment on the roofs: it appears that at some stage, probably with the addition of the tiled roofs before 1974, the chimneys of these houses were removed. Which is a blessing for these owners for maintenance and possible future development of the roof space. Councils now require the chimneys to be kept: no matter that the CO2 producing fireplaces have not worked in 50 years, that the flues are drafty and the chimneys dangerous. They are a ‘heritage visual feature’ and must be kept. In one of our applications, it prevented any additions at all. Gawd help us.

Some changes demanded by Council are for the better: by codifying light and ventilation (largely ignored in the DIY additions) the extensions are larger rooms, with taller ceilings, in order to gain better head height at the rear with more outdoor space than in the 70’s.  In the corner of the 2021 photograph you can see one of several two-storey studios that have been built against the back lane. 

This is tacit acknowledgement that the original houses are too small for today's family home and that building up is better than covering the site with lean-to additions. But why not add a storey at the front? Again, a sensible form of development takes a back seat to ‘streetscape’. And privacy is lost into the neighbours’ rear gardens that would not be the case with an attic with dormers facing the street.

 

Essentially Brand argues that if we look more closely at vernacular buildings, and what changes occupants make to buildings when they are allowed to, we can better understand how to design for current uses as well as future flexibility. His evidence came from analysing hundreds of buildings in historical and contemporary photos, mostly in the USA in the book, and in the UK and more widely in the TV series.

 

Speaking of amenity and privacy in the back gardens, here are two images taken in the same series. The one on the left is testament to 2 great inventions of the 50s: a Hills Hoist and a small lawn needing a Victa mower. So small are the rear gardens it appears the Hills Hoist needs an easement over the neighbour’s land in order to rotate.

50 years later and we see a change in indoor-outdoor living:  the addition has a skylight and doors to the garden (but security is still uppermost in both schemes). The hard-to-maintain grass is replaced with paving for outdoor dining and a service area with a barbecue, and a more compact clothes drying line. The garden is re-oriented to provide some privacy between houses. The backyard is transformed into a back garden, with far more use. Pity about the loss of privacy from the two-storey studios adjacent.

In summary, we can see how these terraces have evolved over 130 years, adding enclosed spaces as needed, filling in and adapting the fabric to suit changes and improving materials for maintenance. But the ham-fisted requirements of council ‘heritage’ codes mean that the improvements made by owners cannot continue: the future goes backward to an imaginary better past. Soon they will want Uber to use horse-drawn carts for deliveries in the back lanes.        

This is a story of how buildings learn, but architects and planners don’t.

Tone Wheeler is principal architect at Environa Studio, Adjunct Professor at UNSW and is President of the Australian Architecture Association. The views expressed here are solely those of the author and are not held or endorsed by A+D, the AAA or UNSW. Tone does not read Instagram, Facebook, Twitter or Linked In. Sanity is preserved by reading and replying only to comments addressed to [email protected]