Sharing with Amin Taha & Jason Coe of Groupwork
The award-winning work of the office formerly known as Amin Taha Architects often stands out for its sophisticated tectonics and surprising use of materials. The first project of theirs that caught my eye was One Six Eight Upper Street: a five-story mixed-use development on a slender corner lot in London’s Islington neighborhood. Using a three-dimensional scan of the Victorian building on the opposite end of the block, the architects milled foam with a robotic arm to create formwork for an in-situ cast terra cotta shell. Cross Laminated Timber (CLT) floor plates key into the structural exterior and large, single-plate glass windows are deliberately misaligned with the original window locations, creating a striking relationship between old and new. Another aspect of the office which further garnered my attention was its decision to shift to an employee-owned business model, renaming itself Groupwork in the process. Seeing both their designs and their method of working as quite topical, I sat down with founder Amin Taha and partner Jason Coe to discuss working collectively while maintaining a striking specificity in their work. Throughout the interview it became repeatedly apparent that the office’s unique approach and manner of working would not be attainable without honesty: honesty with each other as collaborators in a shared practice, honesty about profit-making in architecture, and honesty about the nature of remembering—or mis-remembering—history and its built environment. And honestly, it leads to damn good design.
Sheila Mednick (SM): I'll start out with logistical questions regarding the business model. On a basic level, why make the shift to employee owned? What do you see as the benefits? I understand that one of your first jobs was at Zaha Hadid’s office, a workplace that I think most would assume has a very hierarchical staff structure. Now you’re leading a collaborative firm. Do you feel like you’re venturing into uncharted territory?
Amin Taha (AT): We wanted to make that shift for some time. We’re moving away from the model where there’s one master and everybody else is in the shadow. That was something we were never interested in. I've worked for people like that and it was nonsense. They were the figurehead with a great amount of charisma and publicity, but the practice was entirely propped up by everybody else. We aren’t interested in senior partners holding onto every single piece of the company, we’re interested in sharing whatever we gain. Architecture, more often than not, is not the most profitable venture. But when the profits do occur, you may as well share them with the people who work hard and have been interested and dedicated for a while. We don’t want those who work here feeling like they’re simply employees but that they’re partners who share the profits, have an influence on the practice, and in turn responsibility. A limited liability company structure makes that very difficult as shareholdings need to be diluted to effectively sell a portion to the oncoming partner, which requires accountants and solicitors who need to be hired and paid. So we set up an employee ownership trust, which established that the company is something in which all members of our team are shareholders. Now the trust owns the company and the profits go to everyone. The subdivision of profits is determined by a constitution that we write collectively. We all sit together and write this document saying, "The longer you've been here and the harder you work, the more you get.” It’s a fairly flat hierarchy by comparison and that’s what we wanted it to be.
SM: I would have thought that an employee-owned model could be more cumbersome when it comes to expansion. Do you make decisions collectively about who to hire?
AT: It's not really that cumbersome because we aren’t a huge company. If we were bigger it would be a different story and hiring would be left to human resources. We offer interviews on the basis of qualifications and portfolio, then once over that hurdle we present the candidates to the team so that everyone has the chance to interview and chat more informally. After a candidate presents their portfolio or their recent work we ask each other, “What do you think? Is that somebody who can sit amongst us for the foreseeable future?” Then we decide collectively whether or not they should join us.
SM: With the title of Groupwork and the new business structure, do you see any changes in the nature of collaboration in the office?
Jason Coe (JC): Whilst the company structure has evolved, we’ve always been a fairly collaborative practice. Not long after I joined in 2010 we moved offices and changed from individual banks of desks to a very large dining table format, where everyone sat around one table so we could all overhear what was going on. I think that was one factor in improving exchange and communication in the office. And of course being a small office allows the opportunity for everybody to engage throughout the design review process, whether it’s on an informal hourly basis or through more formal reviews. So there’s always been that underlying collaborative structure which evolved into Groupwork. With the company switch we now have monthly practice reviews where we discuss the management and resources required for each project, which help everyone’s understanding of the business. Before those reviews would have been conducted with the practice directors behind closed doors. Now everybody is part of that conversation, so everybody understands the situation of each project and organization happens on an open table. I think that's certainly been a more fundamental change.
SM: When it comes to your working process, how do you divide tasks? Does each person share in the responsibilities of every project, or do you work more autonomously once it gets past a certain phase?
JC: If you take sharing and collaboration to an extreme where everybody can have a say there is certainly a downside in managing that process. For each project we have both a project architect and architectural assistants. But everybody is of course welcome to design reviews, which creates the opportunity to contribute and facilitates greater conversation. Being a smaller practice, all those conversations happen slightly more frequently and informally. Sometimes they happen if we're trying to make decisions on an element of design, but essentially we still have a project architect or lead who is steering the ship. We still need to have the overall vision in mind and a project architect who really knows the requirements such as program, cost, schedule. But everybody else in the office helps with decision making, and sometimes just the act of asking helps to make a decision.
SM: In working collaboratively, how are you able to you maintain specificity in your work? You’ve once said that style is a product of process. In focusing on the process more than the end goal, do you develop more common threads?
AT: Ultimately you have a brief—a program—that can be tackled from all different directions. What you have to work with is not just the materials and techniques, but a set of contexts. You have a social context, an historical context, the context of the program itself, and your own personal way of looking at things at that moment in time. Sometimes the chemistry between all those will produce something different from even the same person tackling it six months later. If you have multiple people thinking about the problem and you trust that everyone knows the craft of working—in whatever materials, environmental conditions, as well as the context and (what we may call) poetry—you trust them to approach it. Between us all we sit down and say, "Look, this is my approach. I've had a week to think about it and this is how I would do it." We all present to one another and look at the options and say, “Actually that one thing was better than the other," and we have a mature head on and decide which to proceed with. There might be some advantages to the other ideas but the overall poetry may not feel more specific, so we’ll put them in the drawer and leave them to revisit in the future.
SM: Let's shift and talk a little more about the architecture itself. I understand that your office is specific in its approach of focusing on tectonics and the narrative they produce, and the latter becomes somewhat prominent. How do you negotiate between the two?
AT: All architecture has a narrative with its materials being their vocabulary and their juxtaposition their grammar, or potentially their poetry. Sometimes the narrative is more dominant than an overt expression of structure or its process of making. The materials you use relate to that narrative but they're still put together in an architectonic way, which can be prosaic. Maybe there’s a perforated brass façade that’s a representation of the building that was there before, and it’s clipped onto the structure and floating above the ground rather than being rooted in it. That façade is not in any way supporting any structure, it exists solely as part of the narrative. It’s certainly expressed as lightweight and ephemeral—even ghostly, which becomes part of the story of the building. Considering the same idea, other architects in the office might say, “Instead of making that façade just a screen, why don’t we cast it in terra cotta and make it solid? Then the façade becomes the load bearing structure to the whole building.” Then there’s a narrative that initially might have been thought of as just material, but actually happens to be structurally architectonic as well. Instead of being ephemeral it becomes monumental, another type of memory to choose from. So sometimes there's a question where the narrative solution dominates over the structure and remains architectonic in that it expresses its assembly.
SM: On Upper Street, there are some obvious places of the new overriding the old: the windows and their locations, using in-situ cast terra cotta, and allowing certain inconsistencies in the formwork to be revealed. Did your choice of building technologies produce other unpredicted slippages from the historical reference?
JC: The idea was always to build it in a way that represents its construction. But also, as you mention, it's a deliberate juxtaposition to celebrate both the old structure and the new in order to create a relationship between the two. That was something we always anticipated and tried to control, but rather than a literal memory of the building, it could be a fictional representation of an existing building. It's not an exact replica, meaning areas of the construction slip in places, which is natural. We embraced this way of reinforcing the notion that the building is a misrepresentation of the original, but there were other slippages we didn't anticipate. The builders suggested casting their names into the rear of the building, much like how bricklayers used to carve their names into bricks to mark their completed work. So our builders called up the formwork supplier to ask that their names be quietly and discreetly put onto the back of the building, but it was much, much larger than we anticipated. I think it's a meter high in total with five or six names. That was very unexpected. Some inaccuracies also came from the translation of the formwork computer model into the routed polystyrene formwork. Some details became blurred, distorted or lost altogether.
SM: Through that process I would guess that you learn which aspects you have control over and where you don't have control.
JC: On that point of control, we often ask “Where do you have or not have control working with an in-situ material, such as terra cotta or concrete?” We've had a number of concrete producers who could construct perfect samples, but in reality it’s difficult to achieve complete perfection on site. From the early stages the contractor was quite concerned with how to achieve a perfect finish, and they were also worried with how they might correct any mistakes that occur in the process. We wanted to embrace the construction method and the resulting errors and inaccuracies as something we couldn't control. Rather than try to omit them, we embraced them and tied them into the design concept and architecture of the building.
SM: Did you see any errors that you wanted to fix, or did you notice an error that you wanted to amplify?
JC: There were a couple of occasions where we had cracks between the layers of day pours or the contractor tried to fill some holes and we advised them not to do so. Because once you try to repair it, it becomes quite obvious. So we decided to leave it and only repair it if it was a structural defect. There was a large crack on the top of the roof that the client insisted we fill, and we did so in a way that highlights the repair rather than trying to make it completely perfect.
SM: This sounds a bit like a live performance, doesn’t it? Sometimes if a musician makes a mistake while performing, rather than going back trying to fix it they continue on with the performance and might repeat the error so that it seems intentional.
JC: Yeah, I think that's a good analogy. From working with concrete in the past, we knew it was going to be difficult to achieve absolute perfection, so why not try to encourage those errors so that they add another dimension to the concept? Not just errors for errors, but further increasing the notion of mis-remembrance.
SM: I was struck by a photo of the CLT floor plates that were suspended in place before the exterior walls were poured. It was something I had never seen before. It's interesting how that process is apparent in the details, with seeing the way the floor plates key into the party wall.
JC: One thing that's interesting about the floor plates is the original structure was supposed to have concrete floors instead of timber floors but we substituted to CLT, which I think was a big improvement to the project. Our first engineer proposed steel beams and columns on the inside of the masonry party wall and at the other side on the terra cotta wall to support the CLT, which seemed to be a ludicrous amount of additional steel work. Another engineer said, "Of course you don't need that, we can rely on the CLT bearing against the wall directly and into the concrete." As a result the engineer was changed, so now the CLT bears into the walls on either side. I think the previous scheme would have required four or five steel beams spanning across the floor. If the CLT sat on these, it would have been totally unnecessary and it would have been a completely different scheme altogether.
SM: One of the things I noticed about Upper Street and the Costa Brava apartments is that there isn't a sense of tension between preservation and new construction. In these instances the old and the new can exist in the same place at the same time. Was that always a goal, or was it a byproduct of being honest about the nature of remembering and mis-remembering?
AT: I think it's honesty about the nature of remembering to a large degree. It's a reluctance too. If we're being specific about older buildings, there's a long tradition (over 150 years) of leaving the older parts alone. The newer additions might speak of the older part but they're not scrubbing away the older part to replace the tiling. Sounds Ruskin, doesn't it? So it's not far from what you suggest and I think all of us try to do that nowadays. It seems like the most aesthetic thing to do.
SM: But it does go a little beyond aesthetics because it gives a strong dialogue between the old and the new.
AT: That's why I would tend to describe my approach as lazy in that I get exhausted looking at some details, whether it's old and new or only new. It can be so laborious in terms of what you’re trying to cover up and display. You think, "I can't bring myself possibly to draw that, it would be so disheartening to spend hours doing it.” It makes more sense to draw the fewest lines, but often the fewest lines means you have to understand what that material is, how it's going to perform, and how it's going to fit. It means understanding whether it's just a new building or whether it's new and old together, leaving the old to speak for itself with the new doing its own thing. Obviously the intent is a bit more conscious than simply a brand-new building. I'm sure there's a better word than laziness but I use it because people understand it very quickly.
SM: Could you could think of it instead as the path of least resistance?
AT: Yeah but it begs a question: "Why do people take the path of more resistance and end up with something worse?" You'd imagine that the most generic details and generic architecture would be more efficient, quicker, cheaper, and make more money for the practice. But it's not, and it doesn’t. You could end up spending much more time on drawings while detailing the project. Instead we spend more time thinking about it and re-doing the same drawing to end up with fewer lines, rather than spending less time thinking about it and ending up with many more lines. I've been in those kinds of practices so I know that struggle at the drawing board thinking, “Why draw all these extra layers when it can just be left exposed?” Anyway, that’s one of my bugbears.
SM: I'll ask one more question about Upper Street. Sharing is becoming increasingly easier with advances in communication technologies. At Rice, questions have arisen about what copyright and crowd-sourcing will mean in the future. In one way, we can think of the architect of the building at the opposite side of the block—the one which was scanned—as effectively having shared their design with you for your reinterpretation. Would you consider that there was a kind of post-mortem exchange between you and the previous architect, and does Groupwork maintain the idea that one of your projects could become a similar site of exchange?
AT: I've been disparaging about the architecture of that building because it's a classic Victorian approach which quite honestly, is commercial. Here is a parade of buildings: shopfronts with accommodation above for the people who run the shops. However, unlike the rest of the street where it's normally blank shop fronts and blank façades with holes stuck in them, this block is pretending to be Palladian in the weakest possible way, which isn’t atypical of this kind of architecture. The architect could have done anything but here he did something that is actually very weak and cursory. We thought that was quite interesting because it exemplifies how all neoclassical (or Victorian or even Gothic Revival) buildings are always an interpretation of classical buildings with a vocabulary of neoclassical elements. Here we wondered, "Why don't we interpret the next step of neoclassical in the same way?” We took the building on the opposite end of the block, 3D scanned it, and tried to represent it as 1:1. But we didn’t use bricks or stone. We wanted to use a contemporary material and contemporary CAD modelling. The only thing we were worried about was that it was going to come out too perfect, so we let it be imperfect and made some deliberate errors. We developed a methodology that used machine technology to route the formwork and poured terra cotta into that formwork, which had nothing to do with the original Victorian architect and his methodology. We allowed our contemporary methodology to not only have a conversation with the last architect, but to interpret the next neoclassical vocabulary. For instance, concrete doesn't want to pour perfectly. By allowing some slippage in the formwork it was bound to pour into some gaps and make it messy, and that slight mess tells you something. The tooling machinery has routing components of a certain scale which create visible as fine lines in all the details. That immediately tells you it was done by something other than by the hand of a man, who would want to create a finer finish. It tells you it was done by a machine. Similarly, there are always errors when the CAD information is translated by the robot. These all tell you something of today's contemporary methodology.
SM: That brings to mind something I read recently. The identities of artists and architects tend to reside more in the details than in the overall composition. In a sense, the way we approach the smaller segments of a project can reveal more about our intentions. I think that's particularly true with Upper Street although the overall project does have a large, pervasive effect.
AT: Exactly, that's right. I had this conversation with the Architecture Journal the other day saying all buildings have a narrative, but some of the narratives are exactly what I was saying earlier on: expediency. It's still a story. The vast amount of architecture done in this country is either commercial or about commercialism, legitimately needing to make a profit for the company commissioning the project. It shouldn’t be burdened with anything more than a sensible construction, perhaps a pattern book methodology, the integrity of which should by default produce good buildings and in turn good urban landscapes. We managed to do that for a couple hundred years with builders who were developers with in-house draftsmen—until about sixty years ago. The venn diagram now overlaps architects (and architecture) with that past and unfortunately misguides expectations. A building immediately tells you that story, whether it's in the overall building or the detailing. You spot it as you wander around the building. You can read, it in other words. They're all legible and there's fine fiction in the classics.