When visiting a deteriorated building or ancient ruin, one may think the place is lifeless. But that is not the case when it comes to sound. Though the energy embedded in a sound dissipates soon after propagation, the conditions for creating that sound are not necessarily lost. Conceivably, a given sound could be experienced today much as it was in the distant past.
My first experiment with this concept is under way in Arcadia, Greece, using a combination of binaural recordings and psychoacoustic analysis tools. I am an architect and heritage specialist by training; as a graduate student and architecture and historic preservation professional, I helped to document the ruins of a sanctuary dedicated to Zeus on Mount Lykaion. Ancient worship there started as early as about 1600 BCE at the southern peak of the mountain and grew to include athletic contests and burnt offerings as part of ritual practice. Mount Lykaion is a prominent peak in Arcadia, and its location ensured that the smoke of offerings could quickly reach heaven-bound Zeus while remaining visible in the surrounding region. Today only the foundations of some of the buildings and structures survive.
Almost nothing is known about the daily workings of the site in antiquity besides the uses ascribed to each building. The athletic games took place in a hippodrome and stadium placed on flat terraces about 217 meters below the peak. Those venues were surrounded by a cluster of structures including a stoa (a long, narrow colonnaded building for public gathering, offices, or shops), a fountain house, and an athlete’s corridor to the hippodrome.
Acoustic analysis
During my initial fieldwork documenting the ruins, I sat still for hours at a time, drawing the cut limestone blocks and foundations around me, while groups of archaeologists and surveyors traversed the site doing their own work. At times I would hear clearly people speaking from 75 meters away, only to lose the thread of their conversation as they moved much closer. I began to wonder: Were those moments of acoustic clarity not simply a product of the irregular limestone mountainscape but also part of the sanctuary’s design for ritual activity? Was there a reason I could hear my colleagues so well from the ceremonial seats but not from the public fountain house?
Such questions required a new approach to site analysis. I needed a way to consistently compare sonic experience from different recording locations. As an architect, I wanted to see the results visually as well as experience them aurally.
In my work I have employed a binaural recording and analysis system developed by HEAD acoustics in my work. The portable SQobold binaural recording system has microphones in the earpieces of what otherwise looks like a pair of headphones. Its recordings capture a near-perfect impression of one’s sonic experience, from the direction of specific sounds to the subtle reflections off one’s shoulders. The system also records a bounty of data, from actual decibel levels to psychoacoustic measurements such as perceived loudness and sharpness. Those measurements can later be analyzed, visualized, and compared.
Providing comparable measurements requires identical conditions to be recorded. To achieve this on-site, I set up a speaker at human height in a predetermined location—for instance, at the entrance of an ancient building or on a prominent overlook. The speaker plays recorded sounds at a set volume. Wearing the SQobold headset, I then make recordings of how those sounds reach various points of interest around the site. I note meteorological data such as wind velocity, temperature, and relative humidity, as those variables may affect sound propagation and perception. I then move the speaker to another location, and repeat the process. So far, more than 700 recordings have been collected, with about 200 of human speech. Control sounds include a popped balloon, a sinusoidal sweep, and a series of pure tones randomized within the human hearing range.
Selective hearing
Detailed analysis of the recordings is just beginning, but already some interesting patterns are emerging. First, I have confirmed that a hill in the middle of the sanctuary is an exceptional location for people to both hear and communicate with others across the site. If a person were to stand on a specific spot about a third of the way up the hill, he or she could speak with someone located in the middle of the hippodrome below (around 100 meters away and 25 meters lower in elevation) or in the seats (around 90 meters away and 5 meters lower in elevation). In contrast, the athlete baths and the public fountain houses are essentially cut off acoustically from the majority of the site.
The proposed processional way—an easily traversable ridgeline that points from the hippodrome toward the altar above—has also proved to be an intriguing acoustic feature. I have found that, surprisingly, sounds emanating from the ridgeline are not heard very clearly through most of the site. To reach a wider audience, a speaker could move a few meters downhill of the ridgeline. However, equally surprisingly, a listener at the altar can faintly discern sounds from the ridgeline 570 meters away, despite a difference of 187 meters in elevation.
What do these types of relationships tell us at a site where the buildings are so thoroughly in ruins? First, the acoustics give an indication of how sound traveled at the site. That is relevant information: Archaeologists have uncovered evidence of the hippodrome dating back to approximately the fifth century BCE and the lower sanctuary from the seventh century BCE, which indicates 100–300 years of ritual use before the visible stone buildings were permanently erected. Details such as where announcements could best be made, how well a hypothetical procession to the altar could be heard, or where to find the best natural seats for spectating can provide information about daily use on par with microscopy of sedimentary levels. The acoustics work also locates potential new areas for further study—archaeologists may be able to find items that ancient spectators dropped in the seating area, for example.
Acoustic research also might provide insight into the logic of building placements. Is it simple happenstance that the administrative building, which was not necessarily open to the public, has few acoustic relationships to other locations around the site, while the central and public seats, which once held life-size statues and names of previous athletic victors, appear very well connected acoustically? Could the seats in fact have played a prominent role in ritual practice? If only a few weeks of study provided me with notions of where certain activities might have taken place, one can imagine what ancient architects could have done knowing the usage practices of the area for hundreds of years.
Next steps
Much work remains to compare and understand all of the recordings I’ve gathered. The most popular acoustic tools for analyzing a space, such as the Speech Intelligibility Index, either are designed for interior applications where the geometries and materials are known, are insufficient for complex outdoor environments, or do not consider human perception specifically. That’s why I’m turning to the Relative Approach analysis, a powerful tool that analyzes a sound by integrating the complex components to human hearing.
Human hearing is not absolute, but rather context dependent in space, time, and experience. For instance, your reactions to a child screaming next to you in a playground versus on a crowded bus will likely be different. Music played at a concert may seem disturbingly loud to a person who has never attended such an event before, but not to a seasoned concertgoer.
Human hearing also depends heavily on pattern recognition in perceiving and differentiating foreground and background sounds. That is why we can still hear a friend speak while walking down a busy street at rush hour; we are able to tune out the traffic and focus on the pattern of language instead. This effect can vary somewhat depending on the frequency of the sound.
The Relative Approach tool incorporates all those special features of the human auditory experience and produces visualizations that accurately depict human experience rather than absolute acoustic conditions. For instance, a typical spectrogram of a field recording will show a great deal of activity at low frequencies due to wind. It might appear as if the wind accounts for all the sound in the area. However, a Relative Approach analysis of the same recording integrates the natural wind filter present in human hearing. Its spectrograms depict the relative intensity of sounds we might be more likely to hear, such as speech. And since human hearing physiology has not changed significantly since antiquity, what is heard today can be measured as a representation of the past as well.
Sound can provide an entirely new way to look at a place, to experience the present as a direct reflection of the past. It is important to note that with my research I am not attempting to answer how the ancient soundscape was interpreted or what it meant to the original users. That continues to be the purview of archaeologists and historians. Rather, I am striving to expand our ideas of what constitutes operative heritage and to provide a new way to experience that history. Researchers and the public may soon be able to step into the past by listening to historic remnants rather than simply looking at them.
Pamela Jordan (RA, LEED AP) is an Amsterdam-based architect, heritage specialist, and HEAD Genuit Foundation research fellow investigating acoustics in historic preservation. Her research was made possible by funding from the HEAD Genuit Foundation, the Alexander von Humboldt Foundation, and field research assistance from Simon Villegas. She extends additional thanks to the village of Ano Karyes, the Mt. Lykaion Excavation and Survey Project, and the Greek Archaeological Service Arcadian Ephoreia of the Ministry of Culture.
Approximate translation of the German: “We listen to the splashing stream. He wins six games in a row. A little boy was the winner. The burden was too high. Now, the foundations are being laid. The house does not have any garden. They give standing ovations. The day after tomorrow we drive away. I am looking forward to the meal. Nevertheless nobody got full." (Original German text excerpted from the Göttinger Satztest/Sentence Test)
