In conversation: Sheila O’Donnell and Luke Caspar Pearson

This is an edited transcript of a conversation between Sheila O’Donnell and Luke Caspar Pearson, one of a series of such discussions held to reflect on past and future changes in architectural practice and culture as Architecture Today turns 30

SO I started studying architecture in 1971. I had been very interested in art at school, and didn’t exactly know what architects did, so drawing probably was one factor that brought me in. Shane de Blacam taught first year and was just back from working with Louis Kahn, so we had a slightly American culture of pencil drawings on paper. We were almost obsessive about the quality of the drawing, the nature of the line, the sharpness of the lead. Ink drawing on tracing continued the same concerns, worrying about whether you had a 0.13 or 0.18 pen, and the exact difference between them. Line weights – and how light you could go – were were very much on our minds.

Despite the American influence, we didn’t use models much; drawings were the main method and it was plans, sections, elevations, axonometrics in pencil and/or ink, and a bit of colour. You did not cross your lines because technicians and commercial offices in Dublin crossed the lines. Our lines met at the corners.

Amazingly, that was a subject of discussion because we were trying to express an idea that had a precision and thought that it needed a certain kind of drawing in which you wouldn’t be distracted by blobs of ink. Looking back, we were trying to make drawings that looked like they might have come out of a computer.

LP I started studying in the early 2000s at the Bartlett, and my experience was very similar. There was an interest in, for instance, the drawings of Peter Salter and seeing how he drew as a way of creating rhythm in the plan. But at the same time, we had similar concerns about precision and dexterity which – perhaps naively – was interpreted as a kind of emerging expertise. There was a competitiveness about achieving lines that looked like a computer had produced them.

Because we now have the ability to make such precise drawings all the time, what I detect in younger architects is that they have a slightly different relationship to that idea of precision within the drawing. It doesn’t matter so much to them. There’s a celebration of the particular qualities and idiosyncrasies of a hand drawing. When I was an undergraduate, I wanted to eliminate all of that from my work.

SO There’s a kind of editing that you would do, say in axonometric, which a computer still wouldn’t do. Like leaving out a certain line that brings out the three-dimensionality, or making one line in a heavier pen because it sits against space and you want to emphasise an edge – you use judgement to make something slightly different in order for all the lines to look the same.

You mentioned that first-year students at the Bartlett might still be quite interested in being able to draw. Are they drawing mostly by hand or by computer?


Top left: Interior view of the Liverpool University School of Architecture made in watercolour by Sheila O’Donnell.
Top right: ‘Tokyo Back-Up’ by You+Pea draws on Japanese pinball parlours to imagine a pop-up alternative capital that might be required in the event of an earthquake, and was modelled in three-dimensions with hand-painted textures.
Above: Axonometric drawing of the foyer at the Irish Film Centre, Dublin (1992), by O’Donnell & Tuomey.

LP They still draw in pencil on tracing paper, graduating onto film. Talking about how moving from pencil to pen translated into even greater potential precision, I love working in pen now because there’s something quite idiosyncratic about using it for hard lines, which we can do on a computer so easily – a pencil line is hard for a computer to replicate, but a pen line is very easy. But there’s something about the process of connecting all those lines up; the feeling when you start to see the three-dimensional object emerge from the lines is one that you can’t replicate when you’re 3D-modelling something.

SO Because the order in which information appears on the page is so different.

I don’t draw on computer, but of course my team in the office does. One of the things that I found odd about working in 3D is the way the drawing appears, which is sometimes quite strange and counterintuitive.

At UCD the first years draw in pencil. There’s a belief that you have to know how to set up drawings in order to understand how things are made, how three-dimensional things go together. Then you move onto computers.

For some reason I can’t fathom, the students draw on cartridge paper. But to me, working on transparent paper seems really important when you’re developing a design, because you can lay one thing over another, and see the evolution. On opaque paper, you’re making a single drawing that is in itself a finished piece, whereas if you draw on tracing paper there’s always the sense that if you’ve got another sheet of tracing you can keep developing or adjusting the thinking.

Drawing and building

SO As a new graduate, living in London in the 1980s, there was incredible excitement and activity going on; everyone was drawing but no-one was building. So I thought, and still think, that the drawing is architecture. People are probably still working on ideas developed through drawing at that time.

Sometimes drawings take on a life of their own, but it’s hard to draw the line where it moves from being architecture to being something in its own right. The drawings that we make are in the service of architecture – they’re towards describing or understanding architecture, or trying to give birth to an idea. It’s a contribution to the discussion about architecture, so it is architecture.

LP I would echo that, coming from an academic position where we’re researching things and the outcome is not necessarily a building – nor intended to be. We still look back to points in the past where architects were exploring complex ideas about how society could be changed through drawing. Some of those plans were outlandish, but the drawings allowed them to be scrutinised in some way. It would be a shame if that type of agency was to disappear, and we were to associate architectural drawing only with information provided to build something. It’s important for the wider project of architecture to have the possibilities that a drawing affords you to experiment, to make conjectures as to what architecture could be.

SO You never really thought that Superstudio was going to flood whole cities, but the drawing existed as a provocation to people to think about the way things are and might be.

When you are designing buildings that you know will be built, it’s also important to keep examining them through drawings which are about speculation or making your own assessment of the clarity of your proposal, rather than just describing it for planning, or clients, or for construction.

In some ways that is more difficult now, as we use Revit and everything is designed through a 3D model. It took us a while to realise that what is produced by that is not a drawing, really. You can take a shot of a model but that is never the kind of drawing that you would want to have. Somebody has to take something from the model, then use other programmes or media to turn it into something with the clarity of a drawing, which involves editing and emphasising things.

Media and method

LP One of the interesting things that I see with students is a back-and-forth between hand drawing and the software. Students are rediscovering a fascination with the aesthetics of handmade drawings. So, one way they might work is to draw hatching by hand, and composite that on top of a view from a 3D model in the computer.

I work a lot with video game technologies and we did a project with some students in a Kosovan city to make a little virtual world composed of sketches. The city had undergone a lot of hardships in the war, so they were very attached to particular spaces which they drew as a way of remembering certain things. We then placed them into a three-dimensional world – an inhabitable memory that they share with other people. With the ability to go back and forth between media, you can get a sense of inhabiting a drawing in quite a different way, and students are incredibly adept at navigating those lines between things.

SO I’ve only had a few experiences with virtual reality, but I don’t think it feels like drawing – as a way of representing architecture it is somewhere between drawing and something else.

LP I would agree that it fits under the banner of representation – which is a helpful catch-all for all the different ways that architects might describe their work – but is different to drawing, which generally has a certain detachment to the viewpoint; in a plan or a section, it’s not implied that you’re inhabiting the architecture as such – you’re seeing it from another perspective, often one that is impossible in reality.

There are still grey areas because virtual reality is a way of inhabiting a 3D model that places you in there as a person at a human scale, whereas when you’re spinning round your Revit model you’re not inhabiting it in that way at all. But there is something in the process of creating a drawing which I think is different. Drawing might be implicit in the construction of a virtual version of a building, but it’s not the end-point.

In game engine software, the one-point perspective is still ubiquitous, and many titles use isometric or cross-sectional views that relate to architectural representation. But rather than new technologies changing the nature of the drawing, I think the media of architectural representation has expanded.

SO Because I don’t draw on a computer I am not making the final hard-line plans of our buildings anymore, but I still work a lot on sketch overlays. More recently, I’ve also been working in watercolours. It started by making watercolour drawings of mountains, stones and objects, and then making drawings of the sites of buildings, and thinking that it was a good way to move between the process of observing and the process of making and proposing: moving from drawing sketches of a rock on a site to thinking, “The house might have this kind of form in relation to the way the rocks sit on the site.” Because of the nature of watercolour as a liquid and solid at the same time, you can convey volumes very easily with a single stroke of the brush; it has a weight which other forms of drawing that I’ve used don’t have.

For most competitions, I make perspective sketch images of both interiors and exteriors, which are done in consultation with someone working on computer. I use the computer images to set them up, or adjust to find views.

An artist friend gave me a watercolour box a few years ago, which I was a bit intimidated by and then started using and thought, “Okay, I quite like this”. So techniques aren’t always adopted in a very clear and strategic way, but for a recent competition we did in Liverpool for the School of Architecture, we were very conscious of Jim Stirling’s association with the school, and we made a lot ‘up-view full-frontal’ drawings because we wanted to put ourselves into his way of thinking, and how he would have drawn the thing.


‘Up-view full-frontal’ perspective produced by O’Donnell & Tuomey for its successful entry to the competition to extend the Liverpool University School of Architecture, and exterior view made in watercolour by Sheila O’Donnell.

LP The materials that you use to draw can really change the way you approach drawing. When you had the Rotring Isographs or Rapidographs they were extremely delicate, which required precision. But I saw an artist I admired using a pen by Pilot, called the Hi-Tec, which you have to order from Asia. It’s a thin-nib black gel ink pen, so it dries really fast. It gives the quality of an Isograph line, but you can draw as quickly as you could with a fineliner. As soon as I found that pen, the way that I draw started to become quite different.

That implicit connection with the actual thing, the stylus you’re using, is also a factor in digital practices. One of the first things I notice in drawing on a tablet is the difference in feeling. Some people overcome that, and some people always find it difficult because you don’t have the tactility of the pen.

SO It’s funny how we all become slightly obsessive about the tools that we use. It may well be there is another pen that could do that, but it seems special and important. We use those Caran d’Ache metal clutch pencils. You tell yourself it feels somehow different in your hand and you can draw differently with it.

Information and ambiguity

LP One part of my research is to use things like hand-drawing to reveal or uncover the way that virtual worlds are structured – taking things that are super synthetic, that are very much designed around the viewer’s eye. I think it’s also really interesting, the agency of drawing as a recording tool and how it’s different to photography.

We had a conference on drawing, called ‘Drawing Futures’, at the Bartlett three years ago, and a piece was shown by Bernadette Devilat, who was doing 3D-scanning of earthquake zones. She made a really interesting point about recording buildings that had been knocked down through drawing, rather than through scanners, and whether people were happier and more comfortable with a drawn version of something rather than a scan. Because a scanner has this warts-and-all precision, but the idiosyncrasies or subjectivity of a drawing can also potentially be closer to someone’s memory than this super-scientific record which just treats it as pure geometry.

SO Do you think it’s so real that it’s almost unreal, or unrecognisable to people because everything has the same level of emphasis? What’s interesting about drawing is that you’re editing all the time, maybe almost without knowing.

The thing about watercolours is they appear to communicate a lot of information. Often, you don’t have any more information than the simple thing you’ve drawn, and that’s why in competitions it’s really useful. Yet, it’s a risk because you’re setting down something which you know you will, somehow, have to work out how to make it like that but you don’t yet know how it’s done.

Increasingly clients and others want an absolutely photographically accurate picture of what every part of a building looks like. It’s usually too early when they ask for it, and it sometimes pushes decisions to be made out of sequence, which is a pity. It overcommits everybody to an exact relationship between things which may not be possible, because things are discovered as you develop.

Drawing and authorship

SO Even hand drawings on paper, which you might think of as being the work of one person, were often the work of a number of people. Depending on what you were drawing you might have the sheet of tracing paper on the board for weeks. You’re working over and over, and other people might come in and add their part.

In Jim Stirling’s office, you would do a drawing to a certain point but he always did the colouring in, and by colouring it with different kinds of hatches, he would transform it or take ownership of it. In a sense, those drawings are a sort of collaboration.

The way that most people work now, using the 3D model and then a lot of Photoshop, does bring back the sense of the individual touch, though. In the old days you would say, “Shaun is really amazing at 0.13 lines”, or whatever, and in our studio today, you know there are certain people who have particular graphic skills, a certain way of shading or building up drawings that you would go to. I like it that if you look at a presentation there are the recognisable marks of individual hands – or their hands fed through a computer – so the work that gets produced doesn’t all merge into a sameness.

LP It’s really exciting that some architects are looking to generate mechanisms for participation and engagement that can use collective types of drawing in order to involve people who aren’t in the industry in the design design process. On the flipside, drawing does provide that quite singular space, which is why I find it so comforting still. Even with something like BIM, where you have one central model that everyone is operating on at the same time, communicating the building reasserts the power of the drawing as, potentially, a singular act. You can draw something, three-dimensional form starts to emerge out of the page, and it’s just you and the form that’s being created. There’s still something incredibly direct about that, and I don’t see how that could be lost, even with all the wider changes in the industry.

SO I suppose it’s a nostalgic idea that the drawing is a thing – a sheet of paper. If I’m trying to find a straightforward plan of the first buildings we did on computer, it is very hard to locate. But I know where all the drawings of the Irish Film Centre are because they’re in three portfolios in the back of the office on tracing paper. There’s something interesting about the accessibility of the old way, the hard copy, versus the slightly worrying, “Where does it really exist?” aspect of digital drawing.

New directions

LP Some younger practices are working in a consciously non-photorealistic way in their depiction of projects. That style seems to have got a bit of momentum, and some slightly larger offices have adopted it as a successful way of keeping a slightly subjective gap between the imagery and the object.

I’m sure that they have just as detailed a 3D model as the starting point, but there’s this sense that we can still withhold certain types of information, or design the aesthetic of the information to be non-photorealistic so that we can take a very particular stance on what we want to show and communicate.

I think that because young architects are so comfortable and embedded in software, there will be a bit more of a critical pushback about the idea that everything has to be photorealistic, because, actually, the software can allow you to be expressive in many different ways.


‘Found Facilities’ by Luke Caspar Pearson, one of a series of drawings made in pen and marker on paper, and ‘Learning From Los Santos’, an ongoing research project by You+Pea into the facsimile Los Angeles that underpins the video game ‘Grand Theft Auto V’.

SO It’s also much easier now to produce images which are not drawings. I see students pinning up work which is really just lots of screenshots of a model. It does put a stronger pressure on people to know what it is you want to do and what the purpose and intention of a drawing is ­– more than when you had to actually make the drawing and would therefore think, to some extent, about what it means.

Architecture is a strong discipline and its own inherent logics will continue, and that includes thinking about how you represent ideas and the work you’re doing. It may be more challenging than it was when you had to start the drawing with a white sheet of paper and a pencil, but it will be very interesting to see where architectural representation goes in the next 10 or 15 years.