This blog post was born out of serendipity. Which is only fitting, since it also covers the idea of serendipity. It all started with a typical day in my life as an academic: I did about ten different things at the same time and two of them suddenly created an unexpected connection in my mind; something I hadn’t thought of previously and got me really excited. The first event was a phone call with a journalist, who interviewed me about evidence-based design and the science behind workplace design. We talked for quite a bit about cellular offices and open-plan offices and I argued that neither form is superior or inferior per se. Rather the quality of an office space depends on how well it supports the main workflow, business requirements and organisational cultures of a company, but also how well it suits personal preferences and fosters productivity of the people working there. Both cellular and open-plan offices can be designed as enabling spaces. The second event was me preparing for my first lecture for my new cohort of MSc students at UCL. I had assigned one of my favourite texts as reading for the class (Robin Evans: ‘Figures, Doors and Passages’) and re-read it again for the umpteenth time to refresh my mind on the details of the argument. And then it clicked and I saw the text in a new light. So here I want to share my newly gained insights on the relevance of Evans argument for workplace design.

In his essay, Evans sets out to analyse human relationships as they are realised in an architectural plan. Drawing on a fascinating comparative analysis of paintings and floor plans of domestic spaces, Evans contrasts two types of layout: 1) the 16th century Italian Renaissance model of interconnected rooms and 2) the corridor-based arrangement becoming popular in England in the 19th century.

Villa Capra ‘La Rotunda’ (1567-1592) by Andrea Palladio in Vicenza

In the Renaissance period, it was common for houses to have a series of rooms, all connected to one another, such as in Palladian villas. This plan layout does not distinguish between the way through a house and the occupied parts of a house. The villa was permeable and offered lots of choice for movement. It also meant paths of inhabitants would intersect often.

Evans argued that this was reflected in Italian Renaissance paintings of the time (for instance Raphael’s Madonna), where bodies lean into each other, interact, touch each other.

The plans of interconnected rooms therefore represent an architecture and a society of habitual gregariousness, passion, carnality and sociality according to Evans.

Coleshill House in Berkshire, designed by Sir Roger Pratt

In contrast, a very different type of layout was common in 19th century England. The corridor was born. An additional passage was introduced and all rooms were connected to this passage, so that movement and occupation could be separated. Early examples such as Coleshill House still have interconnected rooms, yet the passage allowed for segregation of people and activities. Evans writes that “the occupants of a house (…) had become nothing but a source of irritation to each other” (p.49).

Again, Evans shows examples of artworks, where bodies no longer interact or touch each other – they rather coexist in some distance to one another (for instance Samuel Butler’s ‘Family prayers’ in 1864)

The corridor-based plans thus represent an architecture and a society aimed at avoiding human contact and segregating functions and activities.

So how does this idea of two different types of floor plans discussed by Evans relate to offices and workplace design?

The interesting question that Evans allows us to ask is this: how much is encounter and interaction planned into the system of an office and is therefore part of everyday life? Or, alternatively to which degree is the chance to bump into other human beings seen as nuisance and distraction? In other words: is your office designed to support sociality? Does it contain overlapping paths? Is movement brought into close proximity to occupation? Or in contrast, is it made to support solitude and privacy? Does it offer functional separations? Is movement kept in a distance to occupation?

I have only just begun to think about this, but it seems that office typology (open plan versus cellular offices) is not necessarily related to this question. If we combine both dimensions – open-plan versus cellular, connected versus compartmentalised – we end up with four different types of office layout:

• A) Cellular and compartmentalised
• B) Cellular and connected
• C) Open-plan and compartmentalised
• D) Open-plan and connected

In order to sharpen this argument, I’ve made sketches of each of the four layouts in their ideal form. While I realise that real life cases are often mixtures of forms, I should add that I’ve seen variations of all four types in reality to some degree.

Four types of idealised office layouts and likely patterns of movement

Case A is a very typical office layout and often considered quite old-fashioned these days. Wates House at UCL, which hosted the School of Architecture for decades (but is now being refurbished) was exactly that type of workplace environment. Doors were closed, corridors were deserted most of the time and you hardly ever bumped into anyone.

Case B follows the model of the Palladian villas and the Renaissance floor plan most closely and it could be argued that offices of this configuration are rare. However, I have visited an architect’s practice in Hamburg this year and their offices very much resembled this type of layout. They occupy a series of connected and refurbished former flats in an Art Noveau style building, which are very common for the building stock in many German cities such as Berlin, Hamburg, Dresden or Munich. Most remarkably in this type, there was no separation between movement and occupation – as you moved through the space, you were in close contact with people working at their desks.

Case C again is a very common sight – an open-plan office with a strictly rationalised configuration of desks. Not only do banks and trading rooms often follow this layout, it is also common in the media sector. I have actually studied an office as part of my research, which looked pretty much like the one sketched. Movement here follows the same rational order as the desk layout: a heavily used central circulation space, which is tree-like in its shape. Whilst visually everyone is connected to everyone else, I’ve classified this as compartmentalised, because it cuts movement off from occupation to some degree, for instance people deeper in the plan with desks near the windows will not be in touch with movement.

Case D as an open-plan layout resembles the concept of the Bürolandschaft, which became popular in Germany in the 1960’s as an organic form of desk arrangements, meant to enable human scale. Desks basically form islands, but most essentially, movement can flow freely through the space and there is lots of choice in the system of where and how people want to move. This means movement will spread more evenly and everyone will be more or less in touch with movement flows as they sit at their desks and work. Therefore I’ve categorised this as a connected model, since it follows some of the logic of the Renaissance structures of interconnected spaces.

Essentially the compartmentalised layouts differ from the connected layouts in two major points: firstly, compartmentalised configurations separate movement flow from occupation and therefore limit chances of random encounters and sociability. Secondly, due to a hierarchical structure, movement is concentrated in main circulation spaces in compartmentalised layouts with very little choice for people which path to take, whereas movement spreads more evenly in connected layouts, leaving people choice of where to go and which route to take. This multiplicity of possible routes in a connected layout enables processes of recruitment to happen (i.e. spontaneous short conversations between a seated person ‘recruiting’ a passer-by).

Maybe this typology of office structures and the two criteria to distinguish compartmentalised workplace configurations from interconnected ones, i.e. 1) separation or integration of movement and occupation, 2) concentration or spread of movement, will allow to understand existing office structures more easily and systematically.

So now, go and look around your workplace and ask yourself: are you working in a Renaissance style office of interconnected spaces? Or does your office resemble the stiff and rational 19th century English corridor plan of compartmentalised spaces?

And if we want more collaboration to happen (which is a theme that keeps popping up in wish lists of organisations), maybe we need to go back to the Renaissance ideas of passion, gregariousness and intersections of human paths to create the underlying configurations that foster sociality.