The core is a relatively recent invention. John Wellborn Root’s 1891 sixteen-storey Monadnock Building in Chicago never had one. What building services there were, were all in the middle of the building but hadn’t yet coalesced into a core. Structural rigidity was afforded by the load-bearing masonry construction.
Steel frame construction and the invention of elevators are always on the list of things that contributed to making tall buildings possible but the notion of the core is never mentioned despite it being a major building element integrating essential functions and services with a structural element positioned for maximum benefit. It allowed for quicker frame construction that was less resource and labour intensive. The modern cores is being made to work even harder with post-tensioned cores and tube structure two such developments. The former has construction advantages and allows a lightweight facade to complement a rigid core, while the latter has structural advantages that arise from the core and rigid perimeter walls creating an integrated structural system.
It’s still a happy accident that elevators, fire-escape stairs and service shafts don’t need windows which means there can be the maximum amount of rentable space on the perimeter. Cantilevering floor slabs means that the shape of the outer skin can change, whether there are balconies or not.
A typical office building of the International Style glass box type would have maximised floor plate area to produce elevations with little or no facade depth, but late 20th century high-rise buildings often had facade “incident” created by varying depths of floor slab cantilevers. Variously wavy, curved or angled facades were for a time an indicator of design effort, if not exactly design, just as shuffly windows once were.
With this next building, the corner columns begin to incline to shrink and grow the floor plates and shapify the building as if in response to some external force.
Architecture never takes good ideas in the direction of greater utility but the idea of a core that contains “core” services and functions in order to allow the periphery of the building to respond to local site conditions is a good one. This is a house designed by Skene Catling de la Peña architects. I mentioned it in The Catalogue House.
The architects’ solution to a catalog house and its conflicting demands of prefabrication and customisation “was to pull all of the complicated bits of the house into a central core, and then have the skin adapt to fit the awkward geometries of the given site.” I still think it would have been more convincing if the core configuration included such “complicated bits” as plumbing and services. Fireplaces are neither a complicated nor a necessary part of a house, and the houses of Yo Shimada show that getting from A to B doesn’t require a massive and symbolic feature. [c.f. Career Case Study #10: Yo Shimada]
This next house we know well. It has all the complicated bits grouped together but not as a structural element and only nominally as a spatial divider. This service core doesn’t free up the periphery of the house to respond to local site conditions (although the kitchen does face the less picturesque view). The perimeter walls have little to do other than provide a protective enclosure as the core function of a house.
This 1972 furniture unit by Joe Colombo isn’t much different. It supposedly enables all the activities of living and is a self-contained and 100% designed functional unit for some arbitrary shell to house..
The next apartment combines both approaches. [c.f. Cold Logic vs. Warm Logic] Services and utilities are grouped into a single unit that functions as a spatial divider. Moreover, the apartments are stacked and the shaft linking them is visible in the gap between the roof of this unit and the underside of the slab above and makes the apartment seem more spacious than it is.
In the Farnsworth House the effect of the continuous ceiling “floating” above the service core is somewhat killed by that gap being partially filled by boxing concealing a chimney, at least one soil vent pipe, and, more recently, some A/C ducting. You can see the A/C units in the image below right.
The word core turns out to be a slippery thing. The conventional definition has service and circulation shafts grouped into a single element that also has a major structural function. We also have service cores that group services but may offer some spatial dividend in lieu of a structural one. And there is also the notion of core functions that may be essential but not central – as in in the middle. We therefore have core structure, core services and core function. To this we could also add core meaning in the sense of FLW’s hearth as the heart of the home, and as with the fireplace/staircase house earlier.
Despite the problem of deciding what is core and what is not, and on what level, it remains a good idea to standardise the design and production of the unchanging and crucial parts of a building and to direct design effort to those parts for which site factors necessitate some degree of customisation.
Doing this means that at least part of a building can be optimised even if it is impossible or unlikely that the whole will be. True, some super-optimised solution could emerge from designing everything with regard for everything else but, by the time that solution is arrived at (or the millions of iterations pondered), there may well have been a change in the supply chain or some other environmental factor. It might be better, at the outset, to limit the amount of new design work to location-specific factors only, and to roll out a standard design for THE CORE, whatever it will turn out to be. I don’t mean standard in the sense of some run-of-the-mill, lowest common denominator design but something that has been optimised as best it can be. It is standard in the sense of the new standard and only changed when it can be improved.
It’s still a bit early to call Core Design a thing but many buildings have a standard floor plate with a series of spaces either opening off a corridor or having the potential to, and those standard floor plates are then stacked and vertically linked by service and circulation shafts.There’s a lot here that doesn’t need designing from scratch each time. The core spaces of hotels, prisons, dormitories, apartment buildings, asylums and hospitals are all interchangeable. [c.f. Machines for Living Longer] If you remember, Colditz Castle has been a residential castle, a prison, a nursing home, a hospital and a psychiatric clinic, and is currently a museum and youth hostel.It makes me think it might be more beneficial to think of the core in terms of functions and to work towards optimising those.
I see signs this is happening already. This is the ROVE Hotel Dubai Marina.
This is the ROVE Hotel Downtown Dubai.
This is ROVE Hotel Dubai Trade Centre.
This will be ROVE Hotel Citywalk Dubai.
You get the idea. The Downtown Dubai iteration has surface car parking filling the relatively large site, while all the others have none. Obviously, the bit singled out for standardisation is the typical floor because the typical floor is the core functionality of a hotel. It has double-loaded corridors with fire escape stairs at each end, and elevators, a linen room and a cleaners’ room somewhere in-between. The lobby, shops and restaurants form a podium that may or may not have a pool deck and bar, and may or may not occupy all of the remaining site. (It’s not like the size and shape of Dubai sites vary all that much.) These hotels are not designed to be landmarks but they stand out all the same by being the same. [Is branding efficiency a thing?] Through architecture or more accurately, through the absence of gratuitous design display, this chain of hotels manages to project an image of value-for-money. This is a good thing for a hotel to have, and even better if it extends as far as reasonable rates and real value for money.
• • •