Yesterday, I presented at the London Roman Art seminar, which I attend somewhat regularly. Unlike the majority of my presentations, the Roman Art seminar was a 45 min presentation and I could layout my argument in more detail. The presentation was a summary of my PhD research with some hints at more recent work along the same lines. Those of you that read my posts regularly will no doubt recognise much of the presentation, although it does draw on much of my recent non-theory architectural work. The Lefebvre work is in the background and due to the audience not discussed in any detail. Instead, I focused on three case studies, which build an argument for the rethinking of internal and external noise as socially manipulated within the city of Ostia. In the presentation, I went off script and walked through the physics behind many of the slides. The Q&A reflects the oral version, which downplayed the literary material and was more focused on internal/external and public/private divisions. The paper is posted below and a summary of the Q&A is at the end.
Ear and Stone: Acoustics, Architecture and Art in Ostia
Jeffrey D. Veitch
University of Kent
My title is taken from Richard Sennett’s provocative 1994 book Flesh and Stone: The Body and the City in Western Civilization. I will not discuss this book in any detail and use it namely to point out Sennett’s reliance on sight as a quintessential Roman architectural virtue. For Sennett, Rome was governed by visual order with Hadrian being the epitome (87-123).
This reliance on Roman visual culture was already beginning to be questioned by studies of social practices, such as Roger Ling’s discussion of way finding in Pompeii (1990). What remained elusive was the auditory character of Roman cities and the ways in which Roman inhabitants experienced architecture.
Recently, the role of sound in Roman culture has begun to be explored in the epigraphic and literary sources. Initial studies have highlighted the character of social taboos in terms of trades (Bond 2016) and social history of professional musicians (Vincent 2016).
The sound specific studies draw from an increasing scholarly awareness of movement and multisensory approaches, like that of Eleanor Betts (2017), in shaping social space of Roman places and cities. It is in this context, and specifically Eleanor’s 2013 conference of the same title, that I first began to think about acoustics and Roman social relationships.
It is worth briefly introducing my own non-academic background in sound production, as it has led my interests in acoustics, social theories and architecture. Acoustics can be approached from a diversity of different fields, shown here. Prior to my undergraduate and postgraduate work in archaeology and ancient history, I worked as a sound engineer for live shows. For just under a decade, I worked for bands and venues producing, what is often called ‘mixing’, sound. This work centres in the ‘auditorium acoustics’ circle in the lower right corner. My first two years of university, I was in the electrical engineering department, before moving into the music department for a year and then transferring into the philosophy, history and religious studies departments, as there was no ‘classics’ department within the institution. So, a decade later, when approaching ancient sounds and Roman archaeology, I was thinking in engineering and arts terms, not life sciences or earth sciences terms represented in the diagram.
I had originally planned on writing a thesis on the role of social relationships in neighbourhood formation at Ostia, paralleling some work I had done on collegia there. In the first couple months of my PhD, I attended Eleanor’s conference, presented on street blocking along the Decumanus at Ostia and was reaching a dead end with mapping neighbourhood relations. While the there is a wealth of evidence for collegia social networks, neighbourhood groups, and certain architectural boundaries, these forms of evidence are rather challenging to place either geographically in the city or chronologically in relation to one another. The problems of excavation at Ostia, mostly undertaken by Guido Calza in the 1930s, are well known. The poor documentation and simultaneous reconstruction of the site leave a majority of questions unanswered, or raise more questions than they can answer. I had decided to concentrate on a series of studies that focus on the particularities of the site, especially in contrast to Pompeii. In the process of preparing for an end of year review, I realised that sound, used a tool for analysis could help navigate the challenges I faced. With this in mind, I turned to the standing remains in Ostia. Now, sound as tool often gets missed in presentations so, I want to spell out what it entails through my first case study of building materials.
The uniformity of building techniques at Ostia, even with or without plaster, creates an equal uniformity in acoustic properties. What create auditory hierarchies, in this case, are the dimensions of the space. Differences are evident between opus incertum, used at the Casette Tipo (3.12, 3.13), and opus reticulatum or brick-faced concrete, used in the Insula dei Dipinti (1.4.4). The mass of the wall materials produces transmission loss ratings of 65 dB(A) for opus incertum, 67 dB(A) for opus reticulatum or brick-faced concrete (Veitch 2017). While these are numerically minor differences, namely 2 dB(A), that is enough of a difference for the human body to hear the change.
This suggests that while the difference between opus reticulatum and brick-faced concrete would not be noticeable in terms of sounds, the use of opus incertum was noticeable. This suggests an experiential difference, a difference that could potentially reshape social relations through the wall, and a difference in the chronology of construction techniques. The choices and transitions in building techniques and styles are also choices in the experience of the spaces constructed.
In this way, the famous line of Augustus finding Rome a city of mud-brick and leaving it a city of marble reflects auditory experiences of changes, as well (Suet. Aug. 29). In the slide, I have converted the construction techniques into their absorption coefficients, measured in Sabins (named after Wallace Sabine founder of the field of architectural acoustics). The Sabin gives the amount of sound energy absorbed by the material, which then is converted into the length of time a sound takes to decrease by 60 dB. Turning to the remains at Ostia, the different construction materials and techniques can be converted into their auditory properties and the acoustic properties compared.
In this image, the various interactions between a sound signal and a wall are depicted. The sound (s) will strike the wall, reflecting some sound energy back into the room (A) and some energy being absorbed by the material (F). This will happen again as the sound signal encounters another material, until the sound signal passes through the wall (D) at a diminished sound energy level.
Acoustics is the study of the room based on the auditory properties. In this case, sound is not the object of study, but the methodological tool to analyse the characteristics of the room. In cotemporary design, the room’s acoustics are measured through the recording of sound in the room. Absorption coefficient is the amount of sound energy absorbed by a material, while transmission loss is the amount of sound energy that passes through the material. These modern measurements give numerical values for the reaction of sound passing through or reflecting off a given material. Thus, in Augustus transformation of Rome, the experience is one of less sound absorption and more sound reflected. Sound, in this instance, is the tool for measuring the materials and dimensions of the space, rather than the object of study.
The case studies chosen, namely streets and apartments, highlight the distinctive character of Ostia, especially in terms of building materials and acoustic properties. What these case studies help answer are the relationship between street noise, a product of street activity, and internal space. Or more broadly questions about the interaction between internal and external space, the relation between the concepts of public and private in Roman understanding, and the human experience of Roman architecture.
The density of, and potential for high levels of activity, evident in Ostia would affect internal spaces (see Laurence 2007: 107-9). Hence, do we see evidence for the arrangement of internal space to suppress, or dampen, noise (a modern question)? Or do we in fact need to rethink the boundary between streets and internal space and instead ask, what social function does noise play in Ostia? As Monica Degen, a contemporary sociologists notes, ‘Sound or its absence, can link or divide two separate spaces: inside a house the noise of outside traffic, such as police sirens and beeping cars, disappears or filters through, and questions what is inside or outside, public or private’ (Degen 2008: 44). In answering these questions, I move from apartment arrangement into the streets. Along the way, I will draw on some further research not included in my thesis looking at Mithraea for comparative examples within second century CE Ostia where visibility is restricted into the space.
When looking at the arrangement of apartments in Ostia the rectangular shape of the units leaves the long sides facing the street (most often) or internal space. In certain cases, namely the Garden House complex and the Casette Tipo, the apartments are surrounded by neighbouring buildings insulating them from certain streets. Of particular importance is the presence of windows along one façade, which in 14 cases front onto streets. Now unlike brick-face concrete, windows allow for higher levels of noise to pass through (measured in terms of SPL). It is worth commenting on the standard measurement of sound intensity, or loudness, which is the Sound Pressure Level, or SPL. The SPL is a standard unit, which is the base for modern auditory measurements. The human ear can differentiate between SPL measurements of at least 2 dB and can perceive sounds between 20 dB and 120 dB; over 120 dB the body feels pain and can experience immediate damage.
In this case, the acoustics of the internal room affect the level of sound audible in the street. That is to say, in certain rooms, namely the reception rooms at either end of the apartment, the acoustic properties show an ability to amplify the overall SPL levels, making the space sound louder, and enabling higher levels of noise to be heard outside. Now, the level of auditory clarity diminishes, in this case, making the particularities of exact words harder to interpret. We will return to the affect of auditory clarity at the end and the role of noise, understood as unclear auditory information.
At the scale of the city, it was the location and position of the apartment, in terms of the windowed façade, that suggest forms of ‘noise suppression’, a modern term for the control of sound. Here in the centre of the city, just off the forum and N Cardo Maximus, the Insula of the Painting faces into an internal area, not the street. In contrast, the two apartments in Region 5 form an ‘L’ windows facing onto the street (5.3.3, 4). The acoustic properties of the largest reception room (marked by red diamonds) can be analysed. What results is that 5.3.3 and 5.3.4 have broader frequency response and longer reverberation times, which suggest acoustic properties that make the space seem larger than its dimensions and the ability for higher levels of noise, especially high-frequency noise, to pass through the windows and walls. Where physical location was the only recourse to ‘noise suppression’ the analyses suggest the opposite: the apartments facing the street were potentially ‘louder’, than those facing an internal space. In comparison, the earliest Mithraeum in Ostia, ca. 162 CE, shared a wall with a row of shops and a small courtyard house (3.1.6; see White 2012). In this case, the opus incertum walls acoustically connected these neighbouring spaces, even with visibility complete cut-off.
A contradiction of conclusions can be drawn: first, location was the only form of control over sound, as certain regions or areas of the city could be louder; second, locations were of key social importance for inhabitants. Thus, ‘where’, location of the apartment (or Mithraea), and ‘what’ could be heard were the rhythm of Ostian high-status apartments (or Mithraea) social networks. In either case, apartments or Mithraea, the acoustic properties cannot contain the social groups within, sound emanates out into neighbouring space, creating a wider auditory geography than the physical walls, ceilings, and spaces of social activity.
The first conclusion needs a clarification, namely how locations within the city can be classified as ‘louder’, ‘noisy’, or more auditorially active. A variety of measurements have been devised for the possibility of street activity since Ray Laurence’s ratio of street length to doorway distance (1994a: 88-96), Space Syntax being the most common and complex. A simple look at the location of shops, in relation to the apartments, offers some insight. Shops surround the Insula of the Paintings, while the apartments in Region 5 are devoid of them. By comparison, Hannah Stöger’s Space Syntax analyses suggest that the streets, on which the apartments are found, are in the lower end of statistical movement analysis (choice and integration) (2011: 213-9). This suggests that these streets were not likely used in getting to other area of the city (integration) or as a potential destination for movement (choice). However, Stöger leaves out any connecting road along the north end of the city, thus creating dead ends where the buildings stop. The importance of a street along the river is evident in the placement of porticoes along streets perpendicular to the river, even lesser streets, and a topic I will be discussing at CA on Friday.
The reorganisation of the N Cardo and surrounding area between 116-120 CE highlights the continuation of the streets to the river and connections in the east-west direction. Heinzelmann’s geophysical survey of sections north of the Decumanus suggests possible streets along the river (not published yet), as well as Italo Gismondi’s plan in Scavi di Ostia I indicates an intersection by the river and a north-south street across from the Via del Sarapide (on the 1:500 plan). Connecting these to possible streets along the river seems reasonable and would, therefore, alter the integration and choice measurements for streets connecting the Decumanus and river.
In terms of street sounds, similar acoustic analyses can be done based on the building materials and dimensions of facades and pavements. In Ostia, the majority of streets were paved in basalt, which gives a uniform measurement for the paving. This overlooks the reconstructed aspect of street pavements in Ostia, which requires its own study.
This graph shows the acoustic properties of the three widest streets, all with porticoes along a given section. In this graph, a sound is hypothetical made at the middle of the street (0 along the bottom of the graph). The sound travels down the length of the street in both directions (descending lines from the centre). As the pavement and facades are constructed out of similar materials, the presence of porticoes and the dimensions are the differences shown in the graph (especially evident in the graph of the Via Epagathiana). The dimensions of the street were the primary difference in the acoustic properties, but in this final section I want to turn to another source of evidence for the particular sounds associated with streets and draw some conclusions about the relationship be internal and external space, and the concepts of public and private.
Latin literary sources make frequent reference to streets and sounds as displayed in this chart. The forum is by far and away the most referenced space associated with sounds, followed by via, vicus, semita. What the chart shows are the total counts of sound root words within 100 characters of the corresponding street term, pulled from the Packard Humanities Library. This is not surprising, as Cicero and Livy form the largest corpus of literary texts in the PHI. What this shows is an association between certain literary spaces, namely forum, via, vicus, and semita, and sound.
What, however, this kind of graphing allows for is the particularities of certain sounds in relation to the way streets are conceived within Roman literary sources. In general, ‘son-‘ words are associated with vicus, ‘clam-‘ words with the forum and ‘silent-‘ with the forum and vicus.
If we run the same network with a particular sound carried by people, namely rumours (rumor or fama), the trio of forum, via, vicus shift as via and vicus overtake forum. This association of certain sounds with particular places within the city of Rome reflects the experience of streets. As Ray Laurence discussed, rumours were part of the political communication between elite and non-elite in the late Republic (1994b) and, as shown here, were placed in certain streets by the literary sources. In this case, the misinformation associated with rumours could be seen as a diminished level of auditory clarity, noted in relation to the reception rooms within the apartments.
As noted, the arrangement of apartments was parallel to streets with a central façade and windows predominately along the street. I suggested that this physical arrangement of space created the contradiction of isolating noise through the geographical location within the city and acoustically linking internal noise with the street. As we saw, geographical location and arrangement were evident in the two sets of apartments, however there was no complete sound isolation, or sound proofing. Sounds from the main reception room could be heard in the street at certain apartments. This is in opposition to the way street noise is discussed today (mostly in terms of traffic noise). Today we are concerned with outside, traffic noise, being filtered out or separated from inside noise. What I suggest, in conclusion, is that Roman concerns, both positive and negative concerns, were associated with inside noise being heard outside.
This experience of the breakdown between inside and outside parallels the social role of rumours in streets. The street is associated with rumours and, as shown, acoustically linked to high-status apartments, in the case of Ostia. Now, references to rumours are primarily from the period before Ostia’s peak. Here, Augustus boast returns. The development of construction techniques, materials and the reconfiguring of Roman cities were experienced through Roman ears. The lack of clear sound isolation or development of sound reducing arrangements in the period between the late Republic and 2nd c. CE suggest that the experience of overhearing inside noise did not develop into a particular architecture, but it held a social importance, which breaks down the division of internal/external space and critically questions the continued utility of distinctions between public and private at the entrance to spaces.
M. Degen (2008) Sensing Cities: Regenerating Public Life in Barcelona and Manchester (London).
R. Laurence (1994a) Roman Pompeii: Space and Society (London).
R. Laurence (1994b) ‘Rumour and Communication in Roman Politics’ Greece and Rome 41: 62-74.
R. Sennett (1994) Flesh and Stone: The Body and the City in Western Civilization (London).
H. Stöger (2011) Rethinking Ostia: A spatial enquiry into the urban society of Rome’s imperial port-town (Leiden).
M.L. White (2012) ‘The Changing Face of Mithraism at Ostia’, in Balch and Weissenrieder (eds.) Contested Spaces: House and Temples in Roman Antiquity and the New Testament (Tübingen), 435-492.
I had a series of questions about basic physics of sound (does sound travel up or down at the same rate, if so, what about first floors?) and its relation to particular elements, mostly non-fixed elements (like furniture). I will summarise my response to these questions as a group, as many responses overlapped each other: Sound travels in all directions from the source of the sound. In the case of streets, the sound starts at ground level and reflects off the tarmac, as well as travelling upward. In this instance, sound would travel to upper floors, but it would be diminished as it passed through the building materials, same as the ground floor. First floor windows, in many cases, seem to be open with wooden shutters, rather than glass. This would allow certain levels of noise to enter unimpeded, distance from the sound source being the only effect on the sound level.
In terms of internal furnishings: yes, they alter the acoustics of the space and I did include them in certain instances. Room A in the apartments display signs of usage as dining rooms and I analysed the acoustics with three dining couches, using Pompeian examples, and nine adults (three on each couch as is typical of depictions of dining). Other furnishings like tapestries and carpets are evident in certain sources, and would change the acoustics, but as there is no direct evidence for them in Ostia, I did not factor them into the analysis. All these elements can be added to my database, but I started with the basic room.
I had some questions about transmission loss and its relationship to the width of walls. In particular, comparison was made to modern London where wall widths are, in comparison to Ostia, extremely narrow (difference between brick and mortar walls versus wood frames and sheet rock or wood): The transmission loss is measured not by the width, but by mass. So, wider walls are usually walls with greater mass (unless there is a hollow core, in which case you could calculate the transmission loss of each part individually). There are ancient construction techniques that would block out the majority of sounds, however, in the case of Ostia, these were not standard techniques. The use of the older castrum wall in several warehouses, is a case in point.
I had some further questions about particular sounds in relation to rooms and the layout of spaces, especially in terms of household activities. The first question was the relationship between day and night sounds, as certain carts were barred from entering Rome during the day time. This does provide a key piece in understanding the timing of noise throughout the day and night. Although I did not bring it up in my paper, the cycle of sounds throughout the day is one of the most important factors in assessing Roman inhabitants general perception of sounds. Another comment along these lines was about the level to which an average Roman would even care about noise. I think this is a valid point, which I agree with, but that does not invalidate asking the question. In fact, the lack of clear temporal reference to sounds within the literary sources (Martial’s complaint being the exception that proves the rule) suggests that these noises were not unusual, but in fact very usual and part of the lived experience of cities.
There were also a series of questions about windows: Roman window glass are small panes set within a framework, usually wood. The panes themselves are rather thin, 3-5 mm, and were usually not clear, but coloured. This adds an interesting element to the internal/external discussion, as these are not windows through which someone could see what was happening. The activity could be heard, as I discussed, but the participants would only be recognised by the sound of their voice. Window glass has a distinct reaction to certain sound levels, namely short and loud noises. In these instances the glass reflects a greater portion of the high-frequency sound, creating what is called a ‘slap back’, a sound of the energy bouncing off the glass. This effect happens to varying degrees depending on the sounds produced, but it is a particular effect to glass.
The discussion of activities within the apartments was based on a comparison with Pompeii. The Pompeian house is set perpendicular to the street, which effects the depth to which sounds penetrate the space. Several colleagues are looking at these types of questions (Hannah Platts, RHUL, has discussed this in relation to the Villa of Diomedes outside the Herculaneum Gate). I am interested, and this is potential future study, in looking at the arrangement of entry spaces in Pompeii, as well as first floors (likely Herculaneum, rather than Pompeii), to asses the level of sound penetration in comparison to Ostia. As noted in the paper, the distinctive parallel arrangement of Ostian apartments breaks down the internal/external divide in certain instances. Pompeii provides a different arrangement, as well as different building techniques, that are worth comparative analysis. Now, household activities within the apartments are somewhat limited in choice of location. The small rectangular shape leaves limited places for cooking and suggests that noise associated with these activities would filter through the apartment quite easily. Many of the internal walls are thinner than the exterior walls and would, therefore, allow more sound to pass through. By comparison, the Pompeii house arrangement allowed for a certain level of separation, or distance between reception spaces and household spaces. This is one of the few potential ‘sound isolating’ techniques available. By physically separating activities the sound had to pass through more space, which increases the dissipation, lower the SPL once it reaches the destination.
It was a great discussion and I am grateful to Will Wootton for the invitation, as well as to all those present for the rich engagement with my work.