Building for the Ear (from Chaco Canyon to Medieval Vorarlberg) (2/23/25)

An image of Chaco Canyon ruins from 2012

Note: The current blog post is in dialog with Primeau and Witt (2018), and draws on my own wanderings through Vorarlberg during summer 2024 and on Herbert Kaufmann’s Sakrale Kunst und Kulturstätten to understand Vorarlberg church placement.

Primeau and Witt’s study of “Soundscapes in the Past” asks “how people heard in their wider surroundings,” and answer the question in part with GIS measures. Their insight is that landscape matters and consider significant the “location of features within the built environment and performance spaces” (875).  They use the term “soundshed” – akin to watershed – to capture the way in which sound carries or is disrupted by the topographical features of a place.

They center their study on the rich archaeological site at Chaco Canyon, proposing that the Chacoan builders utilized terrain and topography as acoustical elements in their planning. They suggest that “certain features may have been placed at their locations so individuals may have heard events occurring elsewhere” (p. 875). In other words, the Chaco Canyon residents built with an eye (an ear?) toward the soundshed that surrounded them, choosing building locations and orienting openings to best use the acoustical features of a resonant landscape.

Primeau and Witt acknowledge the embodied nature of hearing, but importantly point out that larger elements in the local environment can shape these embodied perceptions. As Tilley (2008) has shown, surfaces, inclinations, textures and other elements can reduce or amplify sound. Given the nature of a canyon environment, echoic and non-echoic surfaces abound, both as elements of the built environment and of the natural surroundings.

The stark nature of Chaco Canyon’s building ruins of mud-brick and stone (shown in Figure 1 in the left-hand column) might historically have been softened by furs and fabrics, and almost certainly had sound-absorbing storage or even people – in other words by surfaces with less resonance than the  present day. The landscape too with its swales and swells, the hills and cliffs, and even the plant life each contribute in positive and negative ways to noise propagation, particularly at distance.

  Four Views of Chaco Canyon, July 2012

Primeau and Witt  believe, with Hamilton and Whitehouse (2006) that there can be an effective and measurable distance for interactions including speaking and shouting. Their newer methodology, however, seeks to establish more objective parameters than merely personal experience. They propose SPreAD: a System for the Prediction of Acoustic Detectability.

They observe that there are various element at play, distance attenuation being only one. Sound source height plays into sound’s ability to carry, as does the atmospheric absorption loss, which varies by temperature and humidity. Nevertheless, they assert that “Like visibility, audibility can be an actively managed aspect of the built environment, and one can question the relationship between sound and site in the landscape.”

For them, the presence of ceremonial sites on higher locations had significance, for it might mean the audibility of a ceremony’s start or end by individuals elsewhere in the Canyon community. Events in one place were meant to be experienced by individuals in another, they argue, and topographical placement support that.

For a medieval monastic historian, their conch shell examples were readily translated into the positioning of churches in Austria’s Vorarlberg. Such churches had a marked preference for locations on the hills that jutted up out of the local landscape; it is the rare church indeed that lacks a view (or that can be visited without a hike or a climb!). The inventory in Kaufmann (n.d.) makes the point effective: in image after image, it is easy to get a photo of the church from down below (as it were), because they are so frequently higher than the surrounding neighborhood. This can be confirmed in person; wear good walking shoes if you want to climb to see the building in person.

The parish church of St Nikolaus in Damuls, for instance, is on a hill; if it were waterside, we’d describe its placement as on a promontory:

  Damuls -- church of St Nikolaus: image July 2024

Already more prominent than the surrounding landscape, the position of the bells for Vorarlberg churches were even further elevated through built bell towers, typically added in the 13th to the 17th centuries. Elevation proves to be even more important than central location; many of the central churches of my study are several blocks away from the heart of the medieval “downtown,” such as it was. Clearly, the acoustical benefits of being high, along with the defensible ones, made these locations prime real-estate from the perspective of church communications. Neighbors near and far could readily hear the ringing of bells for starts or ends of service – along with weather warnings or peals of other sorts (war, arrival, general announcements). This meant that the topographical benefits of height outweighed the inconveniences of a further walk or an uphill climb. Just as at Chaco Canyon, events in one place in medieval Vorarlberg were meant to be experienced by individuals somewhat distant, and, again like Chaco Canyon, the topographical placement of churches and their bells support that.

Primeau and Witt’s study reinforces the idea that sound is not just a byproduct of environment but an actively shaped and managed aspect of spatial experience. Their concept of the “soundshed” and the methodology of SPreAD provide tools for assessing how people in the past may have structured their auditory worlds with intention—whether for ceremony, communication, or social cohesion.

Applying this framework to medieval churches in the Vorarlberg highlights how different cultures have used elevation to project sound across landscapes. These parallels suggest that sound, like sight, was a crucial factor in how such historical spaces were designed and experienced. Soundshed and soundscape design mattered as much to the late medieval church-planner as to the Chacoan builder some three centuries earlier. Both actively sought to manage audibility as an element of their built environment. Building for the ear in this way reminds us that sound was never incidental—it was an integral part of how people of the past shaped and experienced their worlds.

WORKS CITED:
Hamilton, Sue, Ruth Whitehouse, Keri Brown, Pamela Combes, Edward Herring, and Mike Seager Thomas. “Phenomenology in Practice: Towards a Methodology for a ‘Subjective’ Approach.” European Journal of Archaeology 9, no. 1 (2006): 31–71.

Kaufmann, Herbert, ed. Sakrale Kunst und Kulturstätten: Landesausgabe Vorarlberg. Innsbruck: Süd-West-Presseverlag, n.d.

Primeau, Kristy, and David E. Witt.“Soundscapes in the Past: Investigating Sound at the Landscape Level.” Journal of Archaeological Science Reports 19 (2018): 875-885.

It’s not time travel: sound modeling of 1622 St Paul’s (10/31/24)

1625 Woodcut of St Paul's (from Wall 2014)

Wall, John. “Recovering Lost Acoustic Spaces: St. Paul's Cathedral and Paul's Churchyard in 1622.” Digital Studies / Le champ numérique 3.3 (2014). https://www.digitalstudies.org/article/id/7245/

Sound modeling offers powerful insights about the past, but it’s not time travel. We can tell important things about our past and query our assumptions about it, but we never get to recreate an exact experience of what “really” happened. 

The NEH-funded Virtual Paul’s Cross project focuses on sermons as performance by examining sound as a component of the sermon experience. Using architectural modeling and acoustic simulation, they take a specific sermon from November 5th, 1622 – delivered by John Donne – and explore its sound signature. 

One of the key questions is “the audibility of a sermon delivered without amplification in a large open space for people positioned at different places in the crowd” – a question interesting for my own work on graveside ceremonies. Relying on a full team (“architects, visual and acoustic modellers, linguists, actors, recording engineers, and historians and literary scholars”) they put together a simulation – the fun part of scholarly productivity! http://vpcp.chass.ncsu.edu

Their assessment of the world of John Donne comes with important caveats: 

The purpose of digital modelling is not to give direct access to a world that is forever lost to us, but to enable us to organise and experience in new ways the data that come to us from the past and to evaluate from new perspectives both the scope and limitations of our understanding of that data. The outcome of a project using digital modelling is not an exercise in time travel… They are instead constructions based on interpretations of existing data 

That distance from the experience is one I have thought about as I have climbed hills surrounding churches of interest for my own work; without the pastoral provisions of the 15th century, how much even of topographically-shaped work is reflective (a carefully chosen pun!) of the soundworld my communities might have experienced? Since pasture and forest and field each have their own acoustic signature, the modern mix and familiar pathways of a 21st century hiker cannot replicate the historical past. But perhaps there are elements that DO point towards experience – rounding the corner with that glacially-deposited boulder is going to change the church-bell ring. 

In short, I agree with the authors that we use these replicated experiences “to reconceptualise the subject of our study, re-evaluate the usefulness of our existing approaches, and reconsider the kinds of questions we bring to the discussion.” Models help us think better. 

 (And I could use help thinking, for sure!) 

The St Paul’s team has the benefit of a treasure-trove of visual evidence, largely absent for those of us working at the level of town or country, and the authors have paid careful attention to the relative authority of various evidence types. And, since architecture has implied acoustics, they can move from visual parameters to acoustic models. Here, they follow Vorlander (2009), Longair and Boren (2010), and, expecially, Howard and Moretti (2010) on early modern church practices. 

As they note, large crowds, banners, and tapestries on festive occasions increased acoustic absorption. The practices of religion shaped the acoustics of religion, with impacts both on reverb and on musical clarity. 

After explaining why they chose their space-- the north east corner of the churchyard – the authors explain the ambient noises under consideration and the crowd sizes that they used in their modeling. 

They also give details about the kinds of things that the models “do” for them. They ask: What coheres (and what does not)? What assumptions are WE bringing to our experience? In part, the model becomes a useful filter for understanding our data – and even more importantly, its limits. 

One thing I’m noting just for my future attention: the authors distinguish between “representative rather than specific examples” – the absence of a detailed diary entry means that the generic category supplies information for the site that would not be available from the data inherent to the site itself. Thus, the ambient noise is representative randomly occurring sounds. (A background of silence would obviously have been inauthentic.) 

The discussion of their visual modeling is definitely worth a read https://www.digitalstudies.org/article/id/7245/, but here I’m just going to mention a few things about their sound model: 

1. an accurate model depends on 3 things; the space’s dimensions, the various disrupting forms (“geometric forms”) that are present there, and the materials of which they’re made 
2. in particular, they had to consider the materials of the space: stone, wood, plaster, brick, dirt, and the bodies and clothing of the congregation. This matters in my graveyard ceremony discussions, since so many material goods are mentioned as part and parcel of the ceremony itself. 
3. their model revealed that sound reflections from the buildings SIGNIFICANTLY amplified the speech – it would still be audible 140 feet or more away, instead of dying out at the 96 ft mark. 
4. the ambient noise is, they estimate, 35 Db – far less than the 45 Db of the modern urban environment 
5. bell-based pauses probably have remnants in the text, since a before and after would be necessary to stitch the performance across the bell-peal hole. That’s a really interesting stylistic feature and probably applicable to other kinds of texts as well! 
6. One has to imagine the voice, but a strong and measured cadence would be most effective given crowd size (and public expectation) 

As we are less than a week out from the anniversary of the Gunpowder Day sermon, I thought now a good time to review the process that the group took to a plausible reconstruction of Donne’s sermon as planned. 

But there’s one last twist to share: As every good hiker knows, outdoor delivery of program is beholden to the weather gods, and they did not smile on that Tuesday back in 1622. After all the planning (and reconstruction), that particular sermon had to be moved indoors as the storms rolled through. And that’s the kind of thing that REALLY mucks with your model!