good houses is an advanced design studio centred on one of architecture’s oldest and most open questions: ‘what makes a good house?’
throughout the semester, each student will design a double house as an autonomous architectural project, developed without a site or client. rather than responding to context, the studio asks architecture itself to generate its own conditions, working from rooms, domestic life, construction, atmosphere and materiality outwards.
the studio is structured around a deliberate ‘double eclipse’. students will read the writings of kazuo shinohara without looking at his buildings, while closely analysing the buildings of robert venturi and denise scott brown without relying on their texts. working between these two incomplete bodies of work, each project will develop its own independent architectural position.
representation will be treated as a design tool rather than a means of documentation. large-scale drawings, detailed plans and sections, physical models, material studies and discussions will all contribute equally to the project. the semester will culminate in a complete architectural proposal developed with the ambitions, precision and intensity of professional practice.
rather than searching for a universal definition of a good house, the studio aims to produce multiple, carefully argued answers through architecture itself.
studio ambition
housing conventions change slowly. more than any other architectural programme, domestic space is governed by consensus. conventions become habits, habits become standards, and standards become difficult to question. bedrooms remain bedrooms, corridors remain corridors, and domestic life continues to unfold within familiar arrangements. the studio begins with the suspicion that the house remains architecture’s most familiar, yet still least questioned territory.
the studio therefore refuses the generic. it refuses the catalogue of ready-made options — the standard plan, the default section, the room that exists because rooms like it have always existed. this refusal is not a provocation but a method: if the conventional answers are set aside, each project must construct its own understanding of what a good house might be. that understanding will be individual, argued, and tested; the studio will not provide it in advance.
designing houses and housing is more relevant than ever. it is a universally understood programme, but one increasingly confronted with new forms of living, new expectations, and new demands. despite its apparent familiarity, housing remains an open field for speculation, experimentation, and architectural expectation.
the studio will also attempt to reproduce, as closely as possible, the conditions of a real-world architectural practice. the classroom will operate as an atelier, where projects are discussed collectively, revised continuously, and developed with the seriousness of to-be-built works rather than that of academic exercises — with both pragmatism and optimism. drawings, models, conversations, deadlines, and decisions will all contribute to a working environment that reflects the realities of the profession. the projects may remain fictional, but we believe fiction to be a profound condition of architecture in any context, direct commissions and academia alike.
the ambition is neither innovation for its own sake nor nostalgia for lost models. rather, it is to ask a series of simple questions: if housing evolves so slowly, how much architecture can still happen inside the house? how can it happen? and why does it matter?
objectives
through individual projects, the studio aims to conceive, test and discuss possibilities for the house as an architectural question and, by extension, for housing itself. rather than treating the domestic realm as a fixed typology, the studio will approach it as a field of enquiry, where familiar conventions can be reconsidered, displaced and reassembled through design. the studio will not be searching for a consensual answer to a non-consensual question. instead, it will cultivate multiple positions, encouraging each project to construct its own architectural argument — its own definition of what a good (double) house is.
representation will accompany the work throughout the semester, deployed as a plural field: plans, sections, elevations, models, perspectives, collages and details will serve simultaneously as tools of analysis, speculation and design. representation will not simply document decisions already made, but actively participate in their production.
within this plurality, two instruments will hold the centre. the first is the drawing that gathers everything at once: a full plan or a full section carrying all the details of the project within a single document, from the general disposition of rooms to the profile of a handrail. by refusing the conventional division of labour between 1:100 and 1:5, these drawings will deliberately challenge notions of scale, forcing every decision — spatial, constructive, material — to coexist on the same surface. the second is the physical model, understood not as a miniature of the building but as an instrument for its space, atmosphere and materiality: colour, texture, light, weight, junctions and finishes, tested directly rather than described.
this double instrument will lead the studio into the material culture of architecture. colour, texture, junctions, finishes, furniture, fixtures and the smallest details of domestic life will all become architectural questions. every door, window, wall, stair or threshold will be treated as a project in itself. rather than treating details as the consequence of larger ideas, the studio will understand them as the place where architecture acquires character, atmosphere and meaning.
studio-wide discussions and desk critiques will be complemented by a field trip and a series of external contributions, bringing additional perspectives into the conversation.
brief
the semester is organised around a single, individual exercise: the design of a double house. neither entirely individual nor properly collective, the double house occupies a liminal position between autonomy and coexistence. it introduces questions of proximity, difference, sharing and reciprocity without relying on the repetition and institutional logic of larger housing developments. it is the smallest possible form of housing that is already more than a house — and, precisely for that reason, a territory where the conventions of both can be questioned at once.
the project will deliberately begin from the inside out. rather than conceiving the house as an external form to be subsequently subdivided, the design process will start with the room, its elements, its atmosphere, proportions and purpose, before progressively constructing relationships between rooms, between sequences of spaces, between the two halves of the house, and ultimately between the building as a whole and whatever circumstances might surround it. the house will emerge as the consequence of domestic life rather than as its container.
the exercise will be developed without a site. in 1964, kazuo shinohara proposed what he called ‘original houses’: houses conceived not as responses to a client or a plot, nor as ‘one size fits all’ solutions, but as autonomous works — limited editions, in the manner of artworks, waiting to be acquired. the studio adopts this attitude. each double house will be conceived as an original: complete, precise and specific, yet unattached. in some projects an ideal context may eventually emerge from the ideas of the house itself; but the context will follow the project, never the other way around.
rather than beginning from a blank page, students will work through the careful accumulation of architectural knowledge. throughout the semester, the studio will assemble a collective library of precedents drawn from two partially eclipsed bodies of work, from which ideas, plans, rooms, stairs, windows, columns, walls and attitudes will be extracted, catalogued and reinterpreted. projects will emerge through selection, editing, recomposition and refusal — understanding architectural design not as invention from nothing, but as the precise transformation of already existing intelligence into autonomous positions.
a double eclipse
instead of a site, the studio offers two universes. both will be encountered deliberately incomplete — each with one half of its body of work placed in shadow, hopefully revealing in the process things one couldn’t see in full light.
the first universe is textual. in the mid-1960s, kazuo shinohara theorised the house with a radicality that remains unmatched: the ‘house as a work of art’, the house as a potential for critique of society, the room as a space capable of carrying meaning on its own. the studio will read these texts closely and take them seriously as instructions for design. but shinohara’s buildings will remain outside the frame. his theorisation will be studied without the houses that accompanied it, so that his ideas can be tested against new architecture rather than illustrated by his own.
the second universe is built. robert venturi, denise scott brown and their acolytes produced a body of constructed work — above all houses and buildings of comparable scale — of extraordinary density: plans of deceptive simplicity, facades that argue, ornament entangled with construction, ambiguity handled with precision.
the studio will study these buildings closely, room by room and detail by detail. but their writings will remain outside the frame. the buildings will be encountered without the filter of the theories that made them famous, so that the architecture can speak before the argument does.
a double eclipse, then: a theory without its buildings, and buildings without their theory. the two universes are neither aligned nor opposed. they share unexpected coherences — the conviction that the house is a legitimate site for the most serious architectural thought, the refusal of the generic, the attention to the room and the element — and they diverge profoundly in temperament, language and method. the studio will learn from both, and equally from the misalignments between them.
the goal is not synthesis, and certainly not imitation. neither universe is to be inhabited. by working between two references that never quite meet, each project will be forced to construct its own standpoint — autonomous from both, yet informed by the coherence of each.
context
the studio deliberately operates without a site. the objective is not to escape reality, but to approach it from within the project itself, shifting the emphasis from what architecture responds to towards what architecture is capable of producing. landscape, neighbourhood, planning strategies and contextual narratives will remain in suspension — not postponed, but genuinely absent as points of departure. each house will be conceived, like shinohara’s original houses, as a complete work awaiting its circumstances.
this absence is not a void. the double eclipse — shinohara’s theorisation on one side, the built work of venturi and scott brown on the other — will function as the project’s surrogate context, replacing the role normally played by a physical location. students will repeatedly return to these two bodies of work, extracting spatial, formal and material observations that will inform their own proposals and ground them in an existing architectural reality.
attention will first be directed towards the organisation of rooms and elements, the construction of relationships, the negotiation between the two households of the double house, and the spatial intelligence embedded within domestic space. questions of use, proportion, adjacency, circulation and atmosphere can therefore be examined on their own terms. if, in the course of a project, an ideal context begins to suggest itself — a slope, a street, a climate, a neighbour — it may be welcomed and articulated. but it will be a consequence of the architecture, deduced from the house rather than imposed upon it.
in this sense, the house itself becomes the project’s territory. we will approach it with curiosity, precision and doubt, treating it as a sufficiently vast field of enquiry in its own right. if successful, the ideas developed within its walls will prove capable of extending beyond them, ultimately informing the realities of architectural practice.
questions
there are no recipes, nor predetermined results on the horizon of the semester. we will discuss, agree, disagree, and move forward through questions. these are not all of them, but some we propose to begin with and perhaps return to later.
the double: one house or two? equal halves or unequal ones? mirrored, rotated, stacked, interlocked, or simply adjacent? a shared wall, a shared garden, a shared room, or nothing shared at all? at what point do two houses become one building? at what point does one building become housing?
program: how does one invent it? how much should a house anticipate life, and how much should it accommodate the unexpected? defined by rooms, by users, by activities, or by none of these? fixed or adaptable? precise or ambiguous? how much programme is enough?
plan: familiar or unfamiliar? order or chaos? understandable at first glance or slowly discovered? geometry or intuition? does the plan come first? practical or narrative? can a plan become a worldview?
section: is there still room to reinvent the domestic section? slab to slab 240cm? always parallel? double heights? split levels? rooms stepping over one another? flat floors or topographical interiors? is a section merely a vertical extrusion of the plan?
architectural language: generic or specific? historically referenced or deliberately detached from history? invisible or highly present? abstract, and if so, what exactly is being abstracted? quiet or loud? serious or joyful? why not both?
arrival: how does one enter a house? immediately into living space? through a sequence? through compression and release? is the threshold ceremonial or casual? protective or generous? what is the first room? do the two houses arrive together or apart?
ground floor: public or private? elevated or grounded? a piano nobile? a place of encounter or retreat? how does one negotiate privacy without isolation? what relationship should the house establish with the world outside?
roof: sloped and tiled? flat and dry? occupied or ignored? a technical surface or an inhabitable room? a terrace? a lookout? a hat? a haircut? one roof or two? or three? something else entirely?
car parking: necessary? how much space should be devoted to a machine? underground, above ground, or nowhere at all? can it be used for something else later? what happens when the car disappears?
interior circulation: how much is needed? corridor or no corridor? central or peripheral? room after room? can circulation itself become inhabitable? can one eliminate it altogether?
living spaces: are they still necessary? should they be larger than everything else? smaller? central? transversal? open or enclosed? what is the actual difference between a living room and any other room?
kitchen: a room of its own or part of something larger? small and efficient or generous and social? should cooking and eating happen together? can the kitchen become the true centre of the house?
bedrooms: how many? why? all equal or all different? closed or open? permanent or interchangeable? to what extent is the bedroom a necessity rather than a convention?
furniture: architecture or decoration? fixed or movable? integral or autonomous? something to design or something to ignore? how much of domestic life is determined by objects? bathrooms: how many? how large? related to what? complete or fragmented? shared or private? naturally ventilated? can they become more than technical rooms?
doors & windows: are they different things? should they be? transparent, translucent or opaque? one per room? many? all equal or all different? what kind of life do they produce?
facades: silent or expressive? silent or expressive? ordered or accidental? deep or flat? classical or indifferent to style? coloured or neutral? should a house communicate anything to the outside world? should the double house admit that it is double?
materiality: interior versus exterior? economy of means or abundance of effects? light or heavy? rough or smooth? which materials age well? which ones make domestic life better? can sustainability be spatial rather than technical?
context: what happens when there isn’t one? if the site disappears, what remains of architecture? can a house generate its own context? can it become its own world?
studio trip
the studio trip will function as the 1:1 scale counterpart to the studio’s exploration — and as one half of the double eclipse made tangible. conceived primarily as a journey through the built work of robert venturi, denise scott brown and their acolytes, it will trace a route from new haven to washington and back, visiting a constellation of buildings that have shaped certain strands of architectural culture over the last fifty years. projects such as the vanna venturi house, guild house, and the university of pennsylvania laboratories will become one of the studio’s tangible reference points, establishing a shared imaginary against which the projects can be measured.
the ambition is not simply to visit canonical buildings, but to experience them as architects — and, deliberately, without the mediation of their authors’ writings. the trip offers the opportunity to encounter what neither drawings, photographs nor theories can fully convey: the thickness of a wall, the depth of a reveal, the precision of a pattern, the ambiguity of a decoy, the relationship between ornament and construction, and the atmosphere produced by colour and material. these observations are not secondary, but the substance of the studio itself.
travelling together by bus reinforces this mode of enquiry. rather than isolating destinations through flights, the journey reveals the territory between them — and the territory will not be empty.
along the route, the itinerary will make room for other relevant and worthy visits: buildings, rooms and places that resonate with the studio’s questions without belonging to its central references. long conversations, shared meals and repeated encounters with the work will create a collective space for discussion. the trip becomes an extension of the studio.
more than a collection of precedents, the buildings of venturi and scott brown will serve as one of our guides throughout the semester. they will become, in many respects, the substitute for the project’s missing site: a common architectural ground from which ideas, fragments, details and attitudes can be borrowed, transformed and reacted upon. the objective is not imitation, but reaction and inspiration — developing the capacity to observe architecture closely enough that its fragility can inform new work.
lenses as tools
no architectural representation is ever neutral, passive or objective. beginning with a plan inevitably produces a different project than beginning with a collage, a detail or a photograph. the way a perspective is composed, furnished or left empty is equally consequential. a red column is not the same as a white-and-blue column, whether in a built space or in the promise of one. every drawing is already an architectural position.
throughout the semester, representation will therefore be treated not as the description of a project, but as one of its principal instruments — a plural field in which plans, sections, elevations, perspectives, collages and details will be selected, tested and deliberately manipulated according to the questions each project seeks to address. changing the mode of representation will often become a way of changing the architecture itself.
at the centre of this plural field stand two elements. the first is what shinohara called a detail plan: a complete plan or section that gathers every detail of the project within a single document. general arrangement and construction, disposition and junction, the whole and the fragment will coexist on one surface, challenging the conventional hierarchy of scales by which architecture is usually divided into separate, non-communicating documents. these drawings will demand that decisions normally deferred to ‘a later stage’ be made now, and made together.
the second is the physical model, dedicated to what drawings struggle to hold: spatiality, atmosphere, materiality. colour, fixtures, weight, surface and finish will be tested directly, at scales large enough for the model to operate as a de facto construction rather than an illustration. the model will not represent the project’s appearance; it will investigate its presence.
drawing and model will not divide the work between them but continuously inform one another: the detail plan pulling the project towards precision and coherence, the model pulling it towards matter and experience, each correcting the abstractions of the other. one of the ambitions of the semester is to develop a conscious position towards the inevitable partiality of every representation — understanding it as a cultural and architectural act, capable not only of communicating ideas but of producing them, ultimately shaping the projects as much as the projects shape their representations.
printed/digital
all presentations will exist simultaneously as a physical model and as a single drawing file. neither will be treated as secondary. the model will constitute the project’s physical reality, while the file will operate as its evolving archive. each will constantly inform and correct the other.
a single layout file will be provided at the beginning of the semester and must be used throughout. with a fixed horizontal a3 format, it is intended for full-bleed pages containing one drawing, image or piece of information each. the very same file will be used for every discussion, pin-up and review, whether printed or projected, with no parallel versions or reformatted presentations. at the end of every week, an updated copy will be uploaded to a shared dropbox folder, constructing a collective record of each project’s evolution, successes and dead ends.
the physical model will accompany this process continuously, growing alongside the file rather than appearing only at review moments. every discussion will therefore revolve around the dialogue between the two: the model as the project’s gravitational centre, and the drawings as complementary ways of observing, testing and communicating it. students are free to print whichever pages they wish to sketch upon. crayons, scotch tape, magic mouses and 6b pencils are all equally welcome. together, the accumulated representations and the model should gradually reveal the difficult coherence that every project seeks to achieve.