Space is one of the essential elements of architectural design. It constitutes the vector that amalgamates architectural content, fusing the role of perception with the specific solution of a series of programmatic functions. In tangible, tectonic terms, space forms a framework of reference that usually corresponds to the three Cartesian spatial dimensions. The temporal dimension, conversely, is viewed differently, being integrated into architecture as an immanent layer. Albert Einstein’s contributions with the Theory of General Relativity saw the introduction of the temporal dimension with greater complexity. Einstein outlined concepts such as space-time deformations caused by the effect of gravity and the indissociable interrelation between Cartesian dimensions and time. In relation to these concepts, in his work The Elegant Universe, Brian Greene writes: ‘Accelerated motion not only results in a warping of space, it also results in an analogous warping of time.’ Accordingly, we understand that the quality generated by time extends beyond linearity. This consideration is defined by a necessary revision of the space concept to integrate the temporal dimension—specifically nonlinear time—when it is intertwined with the intangible telematic layer we now find in everyday life. This revision calls for the inclusion of related layers taken from different disciplines, as in other fields of knowledge. Below, by way of a guideline, is a classification of this revision, describing some of the unconventional spatial typologies in today’s context.
These are purely digital, immersive complex spaces, also known as metaverses. The term ‘metaverse’ is a portmanteau of meta- and universe that was coined in science-fiction narrative. First introduced by Neal Stephenson in his novel Snow Crash, a metaverse is a parallel virtual world where people interact socially as avatars and which is a metaphor of the real world but without its physical limitations. The most popular example of metaverse is the Second Life virtual environment, which reached its peak of popularity a few years ago. To experience these types of spaces, users initially needed special devices like bulky virtual world headsets and elements for gestural interaction using cybernetic gloves or harnesses. Virtual spaces are very useful for simulacra and testing in numerous sectors (aeronautics, engineering calculations, safety protocols, etc.), offering simplification and optimization of costs as opposed to ‘physical’ simulations.
One branch of virtual reality has ceased to rely exclusively on virtual worlds to juxtapose digital environments with elements in tangible perceived space. This branch is known as augmented reality (AR). At the same time, this evolution has eliminated active interaction devices or replaced them with others that interpret action in tangible space. In augmented spaces, physical space and interactive system are hybridized, resulting in a mediation of a given physical, tangible space, ‘augmented’ with a layer of dynamic data. This augmented layer processes dynamic or live information in the form of text, graphics, 3D models or other digital entities.
[Fig. 2.1] [Fig. 2.2]
Locative spaces interconnect geolocational information (georeferenced positioning on the earth’s surface) in a space or territory. Locative experiences transform a given perception of a territory or place using global positioning data (GPS), an interactive system that is generally aural. Imagine you are planning to hike around a large natural park. Using a GPS data receiver, a device to interpret this data (an application designed for this purpose) and a portable sound system (such as headphones), your perceptual experience of the route will be very different. The experience will be augmented by overlaying contextual information, sound or other types of data with the perceptual layer of the environment (usually sight, hearing, smell and touch). It represents a new way of experiencing the territory, in which the juxtaposition of layers of sound and information generates a new spatial and relational perception. Locative spaces are usually developed in creative projects that experiment with narrative.
The historical precedent of intangible space is telephonic telecommunication space. Sound, specifically the human voice, is used to construct an intangible space that is dimensionless in Cartesian terms and connects two remote spaces which share the same temporal dimension. The introduction into daily life of the telematic layer, thanks to the Internet, calls for a recontextualization of the ‘space’ generated by these types of media. They are spaces with no three-dimensional topology, generally represented two-dimensionally by interfaces, apps or programmes, and which virtually constitute a space that is formless in the usual space-time dimensions. These intangible spaces host relations between different persons who are not in the same physical space but engage in live interaction, creating a common space of communication, relation and human activity. Examples of this spatial typology are the classic chats or, in recent years, some social networking tools that allow this kind of interaction. It is noteworthy that in chats, the compartments between different people were called rooms in an allusion to the inherent spatiality of these communication channels. These intangible spaces share a temporal dimension thanks to the synchronization of remote physical spaces that may be located in different time zones, anywhere in the world.
In the field of visual perception, holograms are a type of representation that combines the two dimensions of the holographic support with a three-dimensional perceptual effect. In the field of sound, holophones are analogous to holograms. Sound travels in three dimensions through space, making it difficult to zone spaces unless they have physical sectors, partitions and soundproofing. However, in recent years work has been done using techniques that generate a perceptual spatial effect zoned using sound, with no need for partitions. The best known is wave field synthesis (WFS), a holophonic sound technology in the form of a special installation with a large array of loudspeakers (often as many as a hundred) arranged in series around the perimeter. WFS plays a given composition in which sound is manifested as cubic, spherical, longitudinal subspaces, etc., within an initial environment. WFS is a complex technique since it challenges the propagation of sound by cancelling it out in certain parts of the environment. With the use of high precision, complex calculations, a sound can be cancelled in specific areas by means of phase inversion.1 Phase cancellation of a sound is the effect of superposing a given oscillation and its inverse frequency, which cancel each other out. Another type of holophonic technique involves binaural listening using headphones. It is therefore limited in spatial terms, as it requires an extra device that is not needed in the case of WFS.
Spatialization involves multichannel devices, where the whole is greater than the two stereo channels. With this technique, it is possible to move sound using monitors distributed around the space. The result is a surround sound that can express and simulate compositions based on different focuses of sound, as occurs in real listening. Unlike the example of wave field synthesis, and though formally it may seem like an installation, spatialization pivots and moves sound through its various sources rather than zoning it.
Telepresence spaces combine telecommunication and teleconferencing technologies, and adapt them to a physical space. They create remote environments that mirror each other: on one of the walls of space A, a remote space, B, is projected in real time, and vice versa. The result is a specific physical space plus its remote extension, which in some cases coincides with the topology of each of the spaces, resulting perceptually in an interesting visual and spatial continuity.
Sound as a generator of spaces
With regard to spatial typologies, these considerations have to take into account physical dimensional space, overlaid with one or more technological layers. Accordingly, the latter are regarded as hybrid spaces, outside the orthodox classification of three-dimensional space. Creating a fusion between the intangible information layer and the tectonic is more than a postmodern technological anecdote; it is a starting point for resituating architecture in today’s context and bringing it up to date, like most professions that have felt the irreversible impact of the digital overlay.
Historically, architecture has worked with processes in which the abstraction and intangibility of the ideas that generate a design are materialized physically. However, the above examples show that spatial construction is not necessarily dependent on a physical environment and, although it involves other disciplines, this does not detract from it. On the contrary; at today’s critical moment of recomposition of human professions and activities, hybridization is a vital element in evolution, and architecture is no exception.
Recent years have seen the experimental introduction of concepts that resituate the bases of orthodox architecture methodology and design: ephemeral architecture, intangible architecture, particle architecture, generative architecture and media architecture are just some specializations.Though some of them may seem anecdotal, fake or superficial, all produce hybridizations with other fields of knowledge, which should accordingly be analysed differently to their original context. In media architecture, for example, it is necessary to consider the media, technological and communication layers that overlay the original architecture context.
Sound is an intangible, aural manifestation. It is intangible because the scale of reference falls outside the visual range; we are unable to see the movement of the colliding air molecules (or any other kind) through which sound travels. It is aural because this manifestation is captured strictly by hearing. It is also important to remember sound’s capacity as a mechanism for the mental construction of images. Sound, then, can generate not only parameters of the observer/user’s spatiality, but also metaphorical meanings in that space.
The sound-space relation usually addresses the idea that a given space has a sound in itself. A clear illustration is given by wind instruments, where the sound depends on the force applied, but especially on the resonant cavity. If the geometry, topology and scale of this space changes, the sound produced changes accordingly.
Something similar occurs in architecture spaces, though it is unusual to find a building that is treated as a large-scale instrument. An interesting example of this is Teshima Art Museum designed by Ryue Nishizawa, on the Japanese island of Teshima. Here, the architecture becomes a resonant cavity for the interactions of its natural surroundings such as the wind or the sound of the sea filtered by vegetation. Another example is the Morske orgulje (Sea organ) by Nikola Bašić. This is an architectural object in step form, located on the seafront in Zadar (Croatia). The installation interacts directly with the waves, which, by compression, activate pneumatic tubes of different lengths and diameters, emitting different tones.
The above comments are, then, examples of how space can generate sound. But can a sound generate space?
Some of the typologies outlined in the introduction show that sound is very important in the structure and identity of a space, place or territory. At the urban and/or territorial scale, locative spaces provide an interesting example. However, a more rigorous analysis of the role that sound can play in articulating space reveals the mechanisms described in ‘Sound spatialization’ and ‘Holophonic spaces’ to be more paradigmatic. In these cases, sound can delimit spaces, in the same way that walls and partitions do in buildings. These mechanisms can be more or less visible as physical objects, though experiments in this space can produce perceptual limits of spatial positioning with a high degree of aural and perceptual clarity. In wave field synthesis (WFS) installations, it is possible to move freely through spheres, cubes or other forms in which a type of sound is reproduced with great clarity and spatial definition.
Another interesting point as regards the creation of space by sound is conceiving of sound as a construction material that allows other spaces to emerge. However, as sound waves are in constant motion, the space they define will be dynamic. This space cannot be fixed visually or recognised from the outside; it is the body that feels these audible spaces. This is, then, a form of corporeal, haptic hearing, where acoustic perception occurs not just in the ears but in all parts of the body.
An important figure in the field of sound art is the architect Bernhard Leitner. When Leitner began to dedicate himself completely to this field, he thought about using sound as a construction material. He was first fascinated by experimental music notations and then began to design sound spaces, having realized that hearing music in a space is rather different to the formalization of a space by sound. His entire oeuvre manifests this attempt to assimilate sound as a tectonic material—that is, to work with sound in architecture and sculpture.
[Fig. 8] [Fig. 9] (Both images Copyright: Archiv Leitner / courtesy Bernhard Leitner)
Leitner’s work is pioneering in the conceptual area where architecture, space and sound come together. It is comparable in importance to the contributions of Iannis Xenakis with stochastic music.2 Xenakis essayed this technique in the popular Philips Pavilion 3 designed jointly with Le Corbusier for the Brussels’ World Fair in 1958. While Leitner experimented with sound as an articulator of space, Xenakis connected musical composition with the fields of computation, mathematics and systems theory. The definition and connotation of space are usually separated according to professional domain: space in architecture, in astrophysics, in neuroscience, in artistic creation, etc. Yet when major breakthroughs occur, they are defined by the introduction of other frameworks of knowledge. In general, architectural space is referenced by its three Cartesian dimensions, X, Y and Z. However, in recent decades physical sciences have set out to construct theories (such as superstring theory) that introduce more spatial dimensions defined in very different scalar fields.
[Fig. 10] clearly illustrates this point.
It would be very interesting to apply this change of scale and dimension to architecture praxis in order to include scales that are not normally considered, such as those developed in sound phenomenology and the composition of matter. By integrating these scales, it would be possible to expand the compositive, perceptive and tectonic aspects of architecture.4
In the field of physical sciences, another highly significant consideration, referred to in the introduction is Albert Einstein’s contribution of the Theory of General Relativity. He added the temporal dimensión to an understanding of space as a topological surface with curvatures and deformations. From this association he drew the concept of space-time, as the elementary matrix underlying the composition of the universe. This matrix is deformed at different points by the action of massive bodies (matter) distributed throughout space. Gravity can, then, be understood as the distortion that these bodies generate in the space-time matrix. Einstein’s consideration of space-time as an indissociable concept articulates an important connection with the sound phenomenon.
Sound cannot exist without time. Nor does it exist without a mediated space with a potentially variable density to propagate vibration. The propagation of sound is directly proportional to the density of matter. Sound travels faster in solids and slower in gases, because the cohesion between atomic and molecular bonds is greater in the former. For example, the speed of sound in air of 20º is 343 m/s; in water of 25º, 1593 m/s; in wood, 3700 m/s, and in aluminium, 6400 m/s. Propagation speed and transmission medium are of great importance in relation to space. In this way, spatial perception is inversely proportional to propagation speed: the greater the speed of propagation, the smaller the spatial perception. The artist and researcher Edwin van der Heide explains this clearly in an interview conducted for the ‘Sonic Spaces’ thesis.5
[…] ‘Sound in water travels at a completely different speed than in other matter (air, solids). The difference in speed allows us to think of it as having a different scale. Sound in water is five times faster than in air. This means that if we were to listen to space, it would be approximately five times smaller—simply because it goes five times faster, everything is reduced by five.’ […]
As this quote shows, sound, space and medium are directly connected. So, what happens in the digital medium? Obviously, rather than a medium in which sound is transmitted, it is a medium in which to manipulate and transform sound. This concept therefore has different connotations.
In any case, the digital medium is an interesting framework in which to understand the radical breakaway from the ‘rules’ that structure the material or physical world. Today, the digital overlay that envelops the tangible world is capable of shaping intangible matter (data) into numerous concretions or formalizations that are not a priori related. This was not the case in the pre-digital era, which was exclusively tangible. Accordingly, each object is specific and generally has a function, changing the function or specificity of this object calls for a difficult process of transformation; it is not impossible to turn a vase into a chair, for example, but it is certainly difficult.
In short, the potential mutability of any digital entity is due to the absence of matter or body. For example, a digital image can easily be converted into a sound file using computer software. However, we can hardly listen to a painting by connecting it directly to a set of speakers, as these objects are very specific. To do so, we would need an intermediate device that analyses, interprets, converts and reproduces them. According to the above contextualization and consideration, in the digital context it is feasible to convert sound into a three-dimensional form. Using software developed at the MIT (SoundPlot), a series of sound samples or fragments were rendered into 3D mesh models. This type of conversion is called sound morphism. The results are shown in the images below.
Consequently, some three-dimensional models can be built physically, at the small scale by means of 3D printing, or at larger scales using traditional design techniques such as plans, diagrams or notations. In the not too distant future, 3D printers could be capable of dealing with larger scales of architecture. If a model like the one described here were built at the human scale, the form modelled from sound would generate one or a series of spaces. The process could be summarized by the sequence: sound >> form >> space.
In conclusion, this would be a method by means of which sound produces spatial conformation. Though still very experimental, this could introduce interesting elements into architecture projects, such as creating harmonic sound patterns and turning them into harmonic spaces. A harmonic sound is made up of different waves that are geometrically related—that is, in a principal oscillation, there are other superposed oscillations corresponding to a half, a third, a quarter, and so on, of the original principal oscillation. The waves that shape a given sound are defined by metric dimensions (wavelength). There is, therefore, a direct link between sound phenomenology and the metric dimensional form it takes.
The digital introduction of sound establishes a sound-space relation that offers numerous possibilities. This variability allows the inclusion of sound as one parameter more in the design process. Today, this is still a very experimental and little explored method, which should nevertheless be taken into account in terms of the relations that sounds establish with resonant patterns—in other words, the relations between sound and form.6 [Fig. 12]
To conclude, in response to the original question of whether sound can generate space, this article offers two answers.
Firstly, sound can directly shape a space by means of its physical quality. In WFS techniques, sound directly, physically articulates space, albeit by means of highly complex digital calculations carried out by a computer cluster and an array of speakers. Secondly, as shown in the 3D renders sound indirectly shapes space in a more algorithmic, procedural fashion.
The sound-space relation opens up an interesting field of research that could influence future architectural processes. Numerous projects and works have been carried out in this field in recent years, in which architects, designers, researchers and artists explore the potential of sound for articulating and shaping space. In chronological order, below are examples not mentioned in this article:
Soundcube (1969, 1970, 1971), Sound Space TU Berlin (1984), Le Cylindre Sonore (1987), Sound Gate (1990), Sound Space Buchberg (1991/98), Blue Vaulting (1994/2007) and Tuba Architecture (1999) by Bernhard Leitner. Distant Trains (1984) by Bill Fontana. Sound Garden (1983) and Listening Vessels (1987) by Douglas Hollis. Audio Grove (1997) by Christian Moeller. Swiss Pavilion in Hannover (2000) by Peter Zumthor. Spatial Sounds (2000) and Pneumatic Sound Field (2006-7-8) by Edwin van der Heide. Water Pavilion (1997-2002) and Son-O-House (2004) by NOX/ Edwin van der Heide. Whispering Garden (2005) by NOX/Hanna Stiller/ Edwin van der Heide. Elastika (2005) by Zaha Hadid Architects. Tessel (2010) by Lab [AU]. Deep (2010) by Finnbogi Pétursson. Panels (2011) by Paul Devens. Test pattern (2011) by Ryoji Ikeda. Sound of a School (2011) by Haugen/Zohar. Lullaby Factory (2012) by Studio Wave.
 Phase inversion is a method of cancelling out a given acoustic signal. The following example illustrates this method: using a saw-tooth wave oscillator,
When the phase is inverted, the wave has the following form:
When the two are superposed synchronously, the outline of the former cancels out the latter, and vice versa. When the two signals are superposed, the result is zero—that is, there is no oscillation (silence).
 Iannis Xenakis was also a pioneer in introducing elements of complex mathematics (variations, combinations, permutations, probability, game theory, etc.) into his musical compositions, and algorithmics, engines of growth and generativity into the development of musical time. This experimental genre is known as stochastic music.
 This historical precedent, integrating architecture with perception and spatiality by means of sound, was one of the first pieces of hypermedia architecture. Tens of speakers were set into the walls of the pavilion as a direct adaptation of the composition Metastaseis by Iannis Xenakis, which, for its time, was a highly unusual exercise in hybridizing space with sound.
 Today, new “smart” materials are being researched for direct application to architecture and industrial design. Another example is the work of the Mediated Matter research group of the MIT in Massachusets, coordinated by the architect Neri Oxman, which experiments with growth and self-healing of materials.
 Interview to Edwin van der Heide by Bea Goller, Rotterdam 2012 for her Phd Thesis Sonic Spaces at the ETSAM. Not published yet.
 A cymatic device uses a sound source applied to a vibratory mechanism (normally a transducer or a modified speaker linked to a surface) to visualize resonant patterns on the surface.
Greene, Brian. The elegant Universe (New York: W.W.Norton&Company, 2003), p.53-85
Article written by Bea Goller Text licensed with a CC license > by-nc-sa 2014 All images except [fig8+9] by Congoritme CC license > by-nc-sa 2014