Collaging atmosphere: Exploring the architectural touch of the eye

Introduction: Deciphering CGI

A common way for scholars to debate the recent technological transformation of architectural and design firms has been to study the role of computer-generated imagery (CGI). Particular attention has been paid to the effect that CGI and other algorithm-driven software may have on urban development (Graham and Marvin, 2001; Kitchin and Dodge, 2011; Sheller, 2009) and how CGI is part of networks of offices, computer screens, software programs, architects, graphic designers, clients, etc. (Rose et al., 2014; Yaneva, 2009). Against this background, CGI has been criticised both for its fraudulent alluring properties used for marketing purposes (Jackson and DellaDora, 2009; Kaika, 2011) and for debasing the experience of architecture: ’Computer-imaging tends to flatten our magnificent and multi-sensory […] imagination by turning the design process into a passive visual manipulation,’ architect Juhani Pallasmaa writes (2005: 12).

There is a tendency, however, to treat CGI in a generic way with reference to ‘visualisations’ (Biehl-Missal, 2013; Degen et al., 2017; Melhuish et al., 2016; Rose et al., 2014), and lack of specification may easily obscure the image of what each kind of CGI does, when and how. Technical drawings, plans, sketches, diagrams, renderings and collages are all types of CGI, each making their important contribution to the diversified alignment of specialist fields engaged in urban development. Thus, in order to avoid confusion with the common understanding of visualisation as simply being the act of creating an image, I will henceforth strike a blow for a specified and differentiated investigation of architectural image production.

Fieldwork among architects, visualisers and clients has already convincingly argued that CGI in the form of renderings is far from always ‘flat' or 'disembodied', as Pallasmaa would have it, but does in many cases evoke ‘digital atmospheres’ (Degen et al., 2017: 4) or simply atmosphere deriving from the aesthetic, emotional and corporeal effect of the visual encounter (Biehl-Missal, 2013; Stenslund and Bille, in press). However, even if the kind of CGI debated in the mentioned works above remain composites of many layered elements ‘using a wide range of graphic effects’ (Degen et al., 2017: 3) processed by several software programs such as AutoCad, Adobe Photoshop and 3 D Studio Max (Houdart and Minato, 2009: 88–89), they all belong to the type of CGI that I call renderings.

Renderings are ‘productions aimed to pitch a masterplan to an external audience in the early stages of a design process’ (Stenslund and Bille, in press). Later, throughout the design process and by submission of tender, they serve as out-of-house presentation tools made with the intention of persuading investors and clients that the absolute best project proposal is revealed through the rendering. It is no coincidence that this kind of rendering in architect-speak is also called ‘the money shot’ – a ‘shot’, some would say an ‘interface’ (Rose et al., 2014), that seeks to present ‘what [places not yet built] will look like when complete’ (Rose et al., 2014: 386). In renderings there can be a ‘push-and-pull between technical accuracy […] and atmospheric evocation of what it would ‘feel’ like to be in this new place’ depicted by the image (Melhuish et al., 2016: 229). This means that the rendering designers repeatedly find themselves in a limbo between, on the one hand, satisfying often competing interests such as demands made on building regulations, financial constraints or requests, clients’ expectations or professional artistic interests, and, on the other hand, a wish to devote oneself to the creative, unrestrained design process that ideally allows a design studio to follow its vision, which can in some cases involve the production of atmosphere.

It is no wonder that renderings tend to garner most attention. Due to their extended outreach, they meet the public’s interest and therefore also – naturally enough – that of many researchers. However, since the various types of CGI are produced with different intentions, and since the above-described tension between competing interests is not the reality for all kinds of CGI, it is critical to differentiate the field of CGI. This article will focus on one type of image that in some aspects resembles the rendering, yet in others it differs radically. This is the collage, and both renderings and collages unite under one hat by both being graphic productions aimed at evoking atmosphere. However, with renderings targeted at an external audience and collages, as we will see further on, directed towards in-house collaborators, they also differ in crucial aspects. While today’s architectural visual production is often vilified, I seek to re-evaluate the significant contribution of the collage that exceeds the visual communication about a place by evoking an atmospheric and synaesthetically felt sense of a place.

The computer-generated collage has received little attention in the scholarly literature on the architectural process. From Shields, one will learn that the collage has served as a deliberate artistic method in the architecture of selected renowned architects, showing what she calls a ‘collage attitude’ or ‘collage mindset’ (Shields, 2014: 212, 220). Considering, for instance, the marvellous and massive collection of architectural collages made by architect and ‘collage artist’ Nils-Ole Lund (1990), one will soon recognise the collage’s ability to stimulate the imagination and to offer a critical voice that may manage to transgress conventional architectural practices. But what does the tearing apart and layering of different previously unrelated fragments ‘from different levels of reality’ (Houdart and Minato, 2009: 86) do to the process of developing urban design? It is said that methods involving collaging, montaging and assembling ‘convey atmosphere’ (Shields, 2014: 114), yet the riddle remains how exactly this happens, and this calls for ethnographically informed knowledge about processes of architectural image making.

The article draws on nine consecutive months of fieldwork in SLA – a renowned urban design studio with departments in Denmark and Norway and with completed projects in Europe, Asia, the Middle East, Africa and North America. During each day of the fieldwork period, I participated as an observer in internal and external meetings, interviewed employees with various educational backgrounds (interns, graphic visualisers, architects, landscape architects, horticultural scientists, a biologist, a few anthropologists, lead architects, communication officers, partners) in the studio, during inspections or at visits to sites, sometimes doing walk-along interviews asking for their re-enactments (Pink and Leder Mackley, 2014).

In principle, I was unrestrictedly welcomed to join the working day at SLA, and I was treated like any other employee: within the open-plan office I was given ‘my own’ desk twice. Removed only once to another floor due to ‘ordinary rearrangement’ I was not at all facing the same challenges – or given the same opportunities – as Yaneva (2009) was in her fieldwork by Rem Koolhaas, where no employee would have their own workstation. Equipped with a PC and a freshly created e-mail account consisting of my initials (ast@sla.dk), the task of establishing contact with key informants was much easier. As I gradually became involved in more and more projects, I often asked to follow the dialogue with clients and external partners by being cc’d in their email correspondence. Throughout the fieldwork period, I was allowed to access all projects under preparation, including projects of the past accessible through archives.

The fact that I was well received at SLA is of course of great importance to the quality of the ethnographic material that I am able to present and draw on. It is the little things that often matter most. For instance, I was invited to join their lunch programme and a considerable portion of my field notes comes from the informal conversations over lunch. I was also told to help myself to whatever I wanted from the kitchen, where there was an indulgent supply of drinks and snacks that would make even the most disciplined employee feel tempted at some point during the day. Hence, chatting in front of the coffee machine contributed greatly to the insight I was given by some 75 employees in the office at that time.

The material presented in this paper goes across several projects and rests on interviews and observations that tell of a typical use of collages in SLA. What marks SLA is a stated goal of creating aesthetic experiences in the form of atmosphere (Andersson, 2014). Within the studio, aesthetics and atmosphere are often referred to as one and the same, both belonging to ‘the felt’ features of a place that need to complement its ‘measurable’ aspects (such as scale or utility function). The findings presented are naturally not generalisable to all urban design practices, but suggest the applicability to businesses of that type using the aesthetic experience as a starting point for any design solution. All contributing informants appear anonymised with names and sometimes gender that do not match reality.

After the next section, which offers a conceptual clarification of the collage in urban design, the article is divided into three interrelated parts. The first part demonstrates how collaging architects address the distinctive character of a place through a physiognomic approach. In the second part, it is made clear how the collage selectively depicts an overall felt sense of a place, thereby leaving other ways of addressing space to other types of CGI. The concluding remarks sum up and invite further reflection on the important epistemological contribution of the collage within architectural practices. Finally, the limitations of who can ‘read’, sense and feel the atmosphere of a collage are discussed, with suggestion for furthering the research agenda on atmosphere through CGI in the future.

What is a collage and how does it link to the atmosphere of urban spaces?

‘Collage’ literally means ‘glueing’ or ‘sticking together’. It is a way of making an image in which various cut-out materials – for example: paper, cloth, photographs or paintings – are stuck onto a plain, two-dimensional surface. Unlike a montage in filmmaking, a collage is a still image, and it has no 3D objects protruding from its surface like, for instance, an artistic assemblage does. A collage stays ‘flat’ on paper, linen or, more recently and in the case of interest, pixels on the computer screen. It is no more than 10–15 years ago that collages made by SLA were created using paper, scissors and a photocopier. Today, however, when designers collage future prospects of the cityscape, it happens primarily on the computer screen and occasionally it will appear on a printout. Collages are typically made in PhotoShop – a software program that allows the manipulation of a number of images separately (for example: images of paintings, landscapes, sky, earth, stones or other materials) that can then be layered on top of each other, as if it were a bulletin board with cut-out pieces to be moved around above, below, next to or across each other. In this sense, it resembles a photomontage – the composite image of elements from separate sources.

Since Benjamin’s Arcades Project, especially, there has been a vivid debate on the aesthetic principle of a montage – literary and photographic – with which the architectural collage shares similarities and differences. Both genres still seem to carry the intention, as first formulated by Benjamin, of showing rather than saying something (Benjamin, cited in Pred, 1995: 11). Whereas drawings ‘speak’ with collaborators through ‘a lot of technical terminology and detailed hatchings’, Harry – an architect in SLA – says that graphic presentations like renderings and collages are made for non-specialist clients in order to ‘show’ rather than tell. According to Harry, this shift of interlocutor followed by the adjusted choice of CGI – from drawing to presentation – allows an enhanced focus on the atmosphere as the subject of conversation.

The atmosphere of a place can, following New Aesthetics (Böhme, 1995), be defined as the felt sensation of a place. Atmospheres are ‘tuned spaces’, Böhme writes (2006: 25), that require our human bodily presence in order to convey themselves and be conveyed. The alternating description of the conveyance of atmosphere (to convey and be conveyed) expresses a phenomenological ambition to move beyond a subject-object divide. In practice this means that even if atmosphere is omnipresent – it incontestably appears everywhere – the way it appears depends on the person who attends to it. The strength of approaching the study of collaging atmosphere in this way is how a place appears in experience via the collage. Pivoting on atmosphere as something that is co-produced within the intervening space between the environment and the human state of being, the distance naturally dissolves between, for instance, the collage and a given ‘recipient’. Thus, even if I sometimes refer to the ‘recipient’, this person is not a passive receiver of any kind of message sent from designers via a collage, but rather a felt and sensed impression of a place is co-produced within the encounter. This also means that when I tune in on designers working with atmosphere in the city, it is never claimed that urban designers can wilfully create atmosphere either on-site or in a collage – but they do co-create atmosphere, I suggest, in concert with the geographically definable place within the city, with all its characteristic features, as will be illustrated.

I now turn to the way architects explore the physiognomy of urban space through their making of collages. This happens, as we will see, not by a visual depiction of each and every constituent part of the place, but by the careful selection of typical parts that communicate its atmosphere.

A physiognomic approach to urban space

Lily, who had worked intensely with collages throughout the sketch proposal phase for a project that was now approaching its end, showed me how she had sought to ‘translate' some of her collages into renderings: ‘Look here’, she said, handing me some print-outs:

I think I managed very well – in this case at least – to have the atmosphere converted. But you see what I mean by the collage catching a much more direct hold of the atmosphere, right? Anyhow, the renderings tend to appear more far-fetched because we have to take the place into account. In the collage, in turn, you can still feel the roughness.

The answer was none of those. I have yet to meet an urban designer who would be satisfied with a generic project solution. What normally would mark a ‘good project’ in SLA – projects that they would vouch for in public – would be the attentiveness to the unique characteristics of the place under renewal. Showing a sensitive attitude towards the ‘place identity’, as they would call it, helps to develop ‘customised’ visions for clients and users. Lily herself tells how, in order to produce graphic presentations, she would usually leave the studio for half a day or more to collect ‘data’ on site (Figure S2 in Supplementary Material). I witness how several of Lily’s colleagues are doing the same thing; they leave their desks in order to return with photographs of everything from soil or coating of foundations, large stones, old rails, reinforced concrete rubble, rubber tyres, masonry or abandoned buildings, plant species, trees – all of which the architects have a ‘trained eye’ (cf. Grasseni’s ‘skilled vision’, 2004) to detect and ‘translate’ into a ‘material narrative’ to be developed conceptually throughout the design process.

Even if Lily’s proud claim about her site-sensibility would make any ethnographer choke, the studio at least seeks to appear observant of the material aesthetics of the site. As one of Lily's colleagues explains when I ask about their procedure for data collection:

Basically, the way we gather empirical evidence is shaped by the architectural method. So it's photo registration, rather than fieldwork. […] It's not about talking to people […] If we were to do interviews it would take a lot of hours and that’s expensive. […] It’s not that we don’t want to. Or, of course it’s a pity that people [in-house] don’t see the value of talking with locals, but actually what should be regretted, I think, is that clients don’t ask for such service.

Part of urban designers' skill in detecting ‘photogenic’ materials to be brought into a collage rests on a trained eye for what Böhme calls ‘the ecstasies of things' (2001: 131). Böhme refers to his idea about ecstasy in order to demonstrate how things and ‘half-things’ (e.g. wind, sound, smell and light conditions) within our surroundings are anything but passive entities and occurrences, as they ‘extend beyond themselves […] in complex relational interactions with other entities’ (Dorrian, quoted in Böhme, 2017: xxi). By definition, what are termed ‘half-things’ differ from things because of their less stable presence: they seem to come and go and still it makes little sense to ask where they have been in the meantime (Schmitz, 1994: 80). So, for instance, when SLA choose to show in a collage four boys standing in the middle of an open steel structure facing the bay with each of them holding bath towels in the air that are taken by the wind, then the collage communicates ‘a meeting with the wind’ and not, say, how this place might be particularly good for putting out your washing to dry. It is the ecstatic way that things and, in the case mentioned, the wind as a half-thing mark how we feel about a place that the collaging designers pursue.

The way that urban designers produce a collage can be seen as a token of their delicate sense of the ecstasy of things and half-things. But I see their work as even more than that, which is why I make use of a related concept from Böhme to better analyse the collaging of atmosphere. This is the concept of physiognomy, of which a reinterpretation can be found by Alexander von Humboldt (1844). Physiognomy was traditionally used to denote a human being's ‘inner’ character with reference to their ‘outer’ facial appearance, but Humboldt both de- and re-constructs this tradition anew, relying in turn on the work of Johann Wolfgang von Goethe, Carl Gustav Carus and Herbert Lehmann. I will here stick to Humboldt, who carries out a physiognomic reading not of humans but of plants and landscapes, in which he considers the artistic and graphic representation of nature.

Humboldt argues that the very shape of a plant evokes a distinctive feeling of a place – which Böhme interprets as atmosphere (1999). According to Humboldt, it is the job of the graphic artist to suggest this sensory impression of both types of plants and types of landscapes so that ‘their moods’ can be experienced by an outsider – that is, people not familiar with the plants and the landscapes represented by graphics. The ambition of such physiognomic landscape communication is, however, not just to pass on the individual experiences of graphic artists but about creating a fundamental cognition or knowledge: ‘The subject of physiognomy is not […] the individual experience, but the general one, the typical one’ (Böhme, 1999: 100, my translation). Hence, there is a difference between ordinary descriptions of landscapes and a landscape physiognomy which is about generalisability, and it is the task of the artistic designer to address the typology of landscapes through suggestions that reveal, for example, the Nordic feel of Scandinavia, the Mediterranean touch of the Italian sky, etc. Humboldt writes:

Who doesn't feel […] differently in the dark shade of the beech trees; on hills covered by single fir trees; or in the grassy field, where the wind whispers in the trembling arbour of the birch? Melancholic, serious, or happy pictures are evoked in us by these patriotic plant forms. (Humboldt, 1806: 13, my translation)

Back in the office, the designers worked on identifying, naming and categorising the objects photographed, and at some point, the aesthetic potentials and felt expressions that the designers wanted to extract from the materials were discussed over and over again in order to be discarded or pursued further (Figure S3 in Supplementary Material). The procedure involved some careful choices of image-making based on the atmosphere they sought to support through their design. The urban designers would address the constellation of constituents that were telling of ‘the feel of the place’ in order to make them appear in the collage, suggesting the atmosphere of a place through the extraction of things and half-things. As Yoko, architect and specialist in graphics, said to me, ‘it’s about the larger whole, not the parts in themselves.’ Without her knowing, she would then speak along the lines of Humboldt’s total impression of a region (1806: 11).

This way of collaging through a delicate typological understanding of the correlation of felt sensations radiating from a selection of visual surfaces of a place I consider a physiognomic exercise that goes far beyond the acknowledgement of the ecstasies of things. The physiognomic approach to urban place is a ‘reading’ of material surfaces decentring or ‘leaving’ the objects like ecstasy and – via the collage – meeting, touching and merging with a receptive viewer’s felt sense of the given place under redevelopment. When the ecstasies are selected, torn apart and put together in new constellations tied to place-identity, I suggest we talk about physiognomy. ‘The collage’, says Harry, ‘serves a bit like a stock cube holding the essence of the distinctive characteristic of a place’. Harry's metaphorical way of describing the collage ‘holding the essence of a place’ tunes directly into Humboldt's thoughts on landscape physiognomy. To Humboldt, however, essence would not be tied to an inner, hidden or underlying being that can only be addressed hermeneutically. Rather, the essence of a place is a surface character; it can be sensed and felt aesthetically – not as a principle of beauty but as the principle for perception and sense cognition. That is why I tend to consider the physiognomy of the collage as an aesthetic kind of physiognomy.

I began this section with an apparent paradox: the less precisely an image addresses a place, the more atmosphere it seems to evoke. As Lily said, she would capture the atmosphere much better through collaging than through her photo-realistic rendering. As I have argued, this does not imply that collaging is detached from the place on-site, but rather that the method by which collaging designers make a place appear is very different from the way a rendering is made. I have argued that what the designers bring home from their on-site visits in the form of photographs holds the physiognomic potentials of things and half-things that can help give others the impression of what it would be like to be in a place in terms of a bodily, material and sensory presence, described more thoroughly below.

Selective collaging of visual excess

In SLA, there would be no mention of physiognomy and only occasionally you would hear employees talk about atmosphere. Instead, more often, words like ‘felt’ qualities that are complementary to ‘measurable’ parameters would be used to describe redevelopment projects. ‘The felt is the emotional and sensed,’ Jonathan from the communication team explains. ‘For us’, he adds, the felt ‘is about adding value and creating quality of life for people’. In the studio, the urban designers would work with ‘the felt’ when working with the aesthetic experience of atmosphere, while working with ‘the measurable’ would mean addressing scale, quantity, and relational placing of things, trees and buildings, for instance. While technical drawings are particularly good at addressing the measurable aspects of a place, graphic presentations are extraordinarily good at addressing the felt – or atmospheric – dimensions. However, as the reader will know by now, there are different presentations: where renderings must live up to high expectations attending to almost any dimension of a place – both measured and felt, the collage will often selectively address aspects of a place to be felt. This observation will be expanded upon below.

Although renderings do not show reality but merely an imagined future reality, the expectation is nevertheless that they will match reality and not embellish it; they are produced in order to enter a dialogue with external parties about a future cityscape. It is a different case with collages, where standards of reality might still matter, but they matter differently – that is, atmospherically. To exemplify, in a collage (Figure S4 in Supplementary Material) made for the development of a place in the North Port of Copenhagen, some large heavy rocks are placed in the foreground, the sea is rough and rain is pouring down from a cloud-covered sky. The sea surface seems almost beaten by the rain. In reality, this spot may not look exactly how it is depicted in the collage: the stones might be oversized and they are not to be found quite that close to the cargo quay; the wind would normally come from a different direction, which makes the scenario in the collage, with waves hitting the dock, a rare sight. But this is not really an issue here.

In this collage, the designers have deliberately left out some ontological layers of the place – the wind speed calculation, for instance, is ignored – and with the collage they allow themselves to focus on some single dimensions of the place that express its atmosphere. Hence, what the designers aim at in this specific case is to evoke the ‘roughness' of the place, and the roughness discloses itself when the recipient encounters rocks, wind, rain and sea in the collage. One should therefore no longer consider the ‘mere’ appearance of the collage to the eye. ‘The invisible is the firm component of architecture,’ as Hasse quotes the contemporary artist Jörg Sasse (Sasse, 1997: 2, quoted in Hasse, 2012: 43), suggesting that the collage seems to have an associative surplus that can bring forth ‘more’ than what meets the eye. For example, the visual encounter might enable a sense of touch, smell or hearing (Morselli, 2019; Stenslund, 2017).

In the collage, rather, one should be able to imagine how it would feel to be in this place: feel the blowing of the wind, the bite of the cold, the roar of the waves, the play of dim sunbeams far out on the horizon and the grip of the concrete under your feet. It is qualities like these that roughly outline what we may understand as atmosphere: the felt sensation of being in a place, with things and half-things that can no longer be differentiated from the person who experiences them – not even if this experience is brought about by a computer-generated image like the collage. As Pallasmaa says – without, however, recognising the value in the digital production, the collage invigorates the experience of tactility (2000: 80). Or, as I argue, the collage transcends the visible through its way of gesturing atmosphere – it points to something other than what is seen by the eye – just like other kinds of images may manage to do. What one sees is therefore not necessarily what one can see, but something fairly different that should be picked up from the image by multiple ways of sensing.

It might ease understanding of the selective way that a collage addresses urban space to consider the place as analytically divisible into different ontological layers: a mathematical, a symbolic, a social and an embodied space, as Hasse suggests, that can be lumped together in a situational space (2014: 21–42). Through such an analytical lens, a rendering will ‘speak’ to as many layers as possible – it will seek to address the complete situational space: the mathematical space, which can be measured and proportionally arranged; the symbolic space, that can be intuitively interpreted according to cultural production of meaning; the social space, that acknowledges how taste, style or activity appeal to different segments within the population, and the embodied sense of a place, which designates felt impressions of a space (Hasse, 2014, pp. 21–39). Unlike renderings, however, the collage clearly does not deal with space in all its detail, but rather quite the opposite: it seeks to evoke an overall impression of a space.

Hence, when issues pertaining to a computer-generated image are negotiated concerning, for instance, camera angle, scale or zoom adjustments – downscaling (zooming out) in order to show more of the landscape and surrounding area or conversely scaling up (zooming in) in order to show detailed sectional views (cf. Yaneva, 2005), this is a matter of engaging in a space in terms of measurement. Calculable solar studies, shadow diagramming, density studies and approximate location of construction work are measurable properties as well, and such ways of approaching a place by scale are very much reflected in renderings but absent in a collage. Moreover, whilst not excluding that collages can communicate through symbols, this is not its supporting language either. Collages are not primarily to be read but sensed and felt, I argue. This is due to their strong presence effect (Gumbrecht, 2004) that involves the eyewitness to a degree that exceeds pure vision. A collage is ‘ready to hand’ in the sense that it is felt intuitively without taking any notice of intellectual, objective or educated ways of evaluating the architectural design at arm’s length. Again, by way of comparison, when urban designers choose cut-in people with a specific ‘style’ for renderings or when they present furniture by selected brands, they normally do so deliberately, expecting an allegorical reading by a recipient reacting to a language of symbols, social value and taste, thereby addressing what can be defined as the symbolic and social space. However, when the collage seeks to make one feel a place, its aim is slightly different. Collaging atmosphere favours the aesthetics.

The aesthetics account for only part of the complete space ontology, however, and for that very reason, it is prudent practice for an urban design firm to refrain from sharing such a partial piece of the complete design solution with clients until that very moment within the design process when the aesthetics can be complemented with other ontological layers of urban space. This is because it is often not the aesthetics that forms a client's main incentive to seek out a design company for advice, but matters of user behaviour, economic development or environmental issues. Hence, whether clients seek advice to address problems concerning, for instance, rainwater management, CO₂ reduction, noise reduction or crime-fighting efforts, urban design solutions naturally need, first and foremost, to solve such issues on demand with high utility value. In such situations, it seems fair to ask what purpose a collage with felt qualities of a space may serve.

In a company like SLA, which rests its vision and practice on the conviction that everything has two halves: a rational and an aesthetic one; a measurable and a felt one; one of direct utility and one of indirect utility – or better, amenity value, the intention is not only to solve a problem for a client but always to include within the problem-solving an aesthetic surplus that adds something to a place that can only be sensed. A simple example would be that by planting aspen (Populus tremula), one would not only remove attention from unwanted noise nuisance (the utility issue) but also add an aesthetic value by introducing the attractive sound of trembling leaves (the amenity value of the aesthetic surplus). But the aesthetics, the felt and sensed qualities, of the design that is communicated through the collage form only part of complete picture. ‘Collages cannot stand alone’, Yoko once said to me, considering the difficulties associated with using them as communication tools with clients and stakeholders. Collages serve a great purpose as in-house work-in-progress tools enabling designers to focus passionately on the aesthetic elements of their design, but as external communication tools, their function is more questionable and uncertain.

Although intended primarily for in-house use, I would occasionally witness how SLA would nevertheless bring collages to meetings with clients, with unfortunate results. At one point, it came to a conflict where the client did not feel that SLA came up to the standard expected. SLA had presented collages in an ‘identity analysis’ which reflected a thorough observation of the history of the place, its distinct characteristics and the atmosphere to be incorporated into the subsequent design. What unexpectedly happened, however, was that the client was most annoyed by this and dismissed the work of SLA as ‘fluffy’. The disappointed client confronted SLA in an e-mail:

We always wanted you to go into the task and explore different options [in relation to ground plan, heights, parking space]. So why do you choose to prioritise vision, concept and identity over sketches? And this even directly after a meeting where we’ve pointed out our need for no overall narrative. […] It’s very difficult for me to understand why you’ve failed to investigate what we ask of you […]. All we need are good solutions. Nothing else. And we must be sure that these are good solutions and not just a random line.

Concluding remarks

This paper demonstrates that there is knowledge to be gained from a differentiated understanding of CGI. The mediation of urban sites by different kinds of images varies, and at a time when there is much controversy about the way architects and designers visualise their ideas, it becomes important to better distinguish between types of images. The paper demonstrates that architectural image making is a rather complex affair embedded in professional ways of seeing. People do not see the same thing in the same image, and examining the collage as just one of many forms of CGI, it shows how collaging atmosphere relies on skilled physiognomic landscape communication, based on a designer’s ability to carefully select – take apart and combine anew within the collage – the key materials of a space that can make even an outsider feel its atmospheric vibe. The architects’ physiognomic exploration of space through collaging requires the skilful attendance to its multisensorial appearance. Hence, collaging atmosphere makes vision work together with other senses in order to produce ‘the stock cube’ of a place – a metaphor that serves as a reminder about how the ‘quick and easy’ selective and fragmented way of making a collage can serve as a ‘flavour enhancer’ that intensifies and develops the architects sense of a place in terms of atmosphere.

Both renderings and collages aim to evoke atmospheres, but they address different audiences with different types of atmospheres that are more or less easy to understand. Renderings are produced with intentions of being easy to digest and they must hold a broad appeal. Aimed to humour what SLA employees would often call a ‘many-headed client’, insinuating the monstrous phenomenon of being confronted with the wishes of several clients at once, the professional renderings would need to assuage potential worries or concerns of all parties, be they citizens, politicians, traders or proprietors, by providing answers, and thus ensuring appeal to the greatest common denominator.

Collages, in turn, are tricky and it seems a matter of professional pride when architects and designers dismiss renderings. Different adjectives are used when employees in the studio speak about renderings and about collages. Renderings are predominantly ‘smooth’, ‘polished’, ‘finished’, ‘realistic’ and carefully composed, whereas collages are ‘rough’, sometimes very ‘abstract’ and made ‘in a jiffy’. Both renderings and collages are often very ‘beautiful’, but beautiful each in their own way. According to the design specialists that I have spoken to, a good collage can be made by a single employee in under an hour, whereas the professional rendering will usually require weeks of preparation with multiple parties involved. Renderings obviously are far more time-consuming, much more expensive and they are ‘smooth’, I argue, in the sense that they seek to satisfy all parties as well as possible.

The paper has highlighted the role of the collage in urban design practice as very different from a rendering. From the outside and at first glance, the collage may seem insignificant. Collaging atmosphere is therefore easily disqualified as an artistic ‘fluffy’ excess that does not answer a call directly. A practical implication of this would be the cautious handling of collages in marketing, which, as the paper has shown, has turned into a risky business. ‘Clients have far too many expectations’ in the sense that they wish to see how a project turns out to be ‘in reality’, says Yoko. Based on insight from my ethnographic study presented above, I suggest a different argument: collages are artistic, and they are less accessible and ‘readable’ to non-architects, graphics or designers, so it is necessary to raise awareness of its limitations as well. As I learn from Yoko, collages serve as epistemological tools initiating a process of recognition during the design development:

When I have a question I want to investigate, I make a collage. And then it will start asking questions in return that I will need to respond to [for instance] by way of describing the collage to my peers. I then try to put some words on the collage […]. Through this process I become more distinct in the way that I formulate the concept, and I get better at putting words to what we decide to draw.

As I have shown, however, collaging atmosphere involves a predominantly material and ‘synaesthetic’ focus on things’ surfaces, and the only people who enter into dialogue with the collaged atmospheres are the architects themselves. As I ask Eric, an anthropologist at SLA, about the company’s dialogue with local citizens, he explains that most architects in the studio ‘think it's a bit tiring and ‘folksy’ to have interviews and empirical knowledge of local culture:

You need some form of friction to create something new, which touches on architects’ self-understanding as being almost artists. There is a cliquish kind of aesthetic among architects, where they practically communicate more with each other than with others. Obviously you want recognition from your colleagues and not from the municipality or the local people.

The atmosphere of a place is collaged by urban designers in order to create a fairly small refuge for architectural creation at a distance from outside interference. It serves as an epistemological tool of crucial importance to the vision of urban design – a vision in terms of atmosphere that moves far beyond sight into the multisensory and embodied awareness of how places are experienced. Collaging atmosphere holds a critical potential in testing the conventional thinking of designers, yet, with greater awareness of its limitations being selectively material and aesthetic, and also of it holding little or no ethnographic knowledge of local culture, it can be well complemented by other types of CGI of which we still deserve to know more about in terms of atmosphere.

Supplemental Material