Acoustic search for ‘The Witches Valley’ Antonella Bevilacqua 1 Department of Industrial Engineering, University of Parma Area delle Scienze, 43100 Parma, Italy Cobi van Tonder 2 Department of Theatre, Film, Television & Interactive Media, University of York York, UK Ilaria Lombardi 3 Department of Industrial Engineering, University of Campania Luigi Vanvitelli Aversa, Italy

ABSTRACT Benevento is famous as the ‘city of witches’ according to folklore dating from the 13 th century. There exists much mystery and variation around the origin of the legend of ‘The Witches Valley’. Various writers, composers, poets and artists have referred to it, including the chief physician of Benevento, Pietro Piperno, who in 1639 in his book entitled ‘De Nuce Maga Beneventana’ traced the legend of the witches back to Lombards invaders in the seventh century, who gathered south of the city to dance and perform rituals under a magical walnut tree somewhere in the valley of the river Sabato, in a long narrow gorge. This paper aims to confirm the place of this legend based on acoustic properties and suggests two possible locations in the valley of the river Sabato: The Strait of Barba and Wolf Rock, that due to their acoustic properties could have been the place of the legend. A campaign of acoustic measurements has been performed using firecrackers recorded with a Brahma microphone. The acoustic analysis confirms that both these places are character- ized by a development of sound reflections, a phenomenon that can explain why a drum roll was perceived as a very reverberant and ‘magical’ sound. 1. INTRODUCTION

History shows a long trail of how sound is connected to the sublime, whether deliberately manipulated by shamans, oracles, priests or witches in their rituals or whether determining the choice of sacred (magical) sites for spiritual activities. In prehistoric times the phenomena of sound propagation and reflection was unknown; as such, our primitive ancestors associated the presence of supernatural di- vinities to the rocks where the sound was reflected. In France and Spain in a few Paleolithic caves the images of hunting with quadrupedal animals have been discovered, detected in the areas where the reverberation is considerable, while scenes with feathered animals were discovered in the areas where the reverberation is not perceivable [1-3]. Shamans selected reverberant sites such as caves or narrow valleys to amplify their sounds, (which would often be the beating of drums) and intensify an

1 antonella.bevilacqua@unipr.it 2 cobi.vantonder@york.ac.uk 3 ilaria.lombardi@unicampania.it

worm 2022

atmosphere of mystery and otherworldly power in order to impress the participants. Additionally, series of echoes and reflections generated the effect of mysterious voices – the voices of gods. Ex- amples can be found ranging from the pre-Columbian societies of Latin America [4], to ancient Greece [5] that show how reverberant sound was associated with supernatural phenomena: the echoes were believed to be the voices of spirits. In literature, there are various articles on the acoustics of caves and on the acoustics of valleys from which one can draw inspiration for the acoustic study of similar places and with modern acoustics techniques it is possible to investigate the propagation of the sound field in places where one believes they may have been used for sacred rites or functions.

This paper deals with the acoustic investigation related to the existence of valleys in the country- side of Benevento where the legend of witches was born. Specific places located in the valley of the river Sabato have been measured. This valley is still a natural amphitheater surrounded by woods and hills.

2. THE LEGEND OF THE WITCHES’ VALLEY

In Piperno’s ‘De Nuce Maga Beneventana’, the legend of the witches is traced back to Lombards invaders in the seventh century [6]. Though formally converted to Catholicism these Lombards car- ried on with their pagan traditions. The Christians of Benevento would have considered their rites as that of witches. According to the legend, witches were considered women with magical knowledge who could cure diseases with the use of medicinal plants. There is a description of witches that gath- ered south of the city to dance and perform rituals under a magical walnut tree, locating the place of their rituals somewhere in the valley of the river Sabato, in a long narrow gorge. Figure 1 shows a historical representation of Benevento as reported in the book by Piperino. In particular, a tree wrapped by a snake with the witches dancing around is indicated on the right side of the river Sabato, identified as the place where the witches were used to have periodical meetings. Supposedly, Saint Barbatus cut down the sacred tree and tore out its roots leaving acoustic evidence as an alternative way to confirm where the rituals took place. The Lombards would have selected a place where the sound could be amplified by the natural conformation of the landscape in order to impress the partic- ipants. Two possible locations in the valley of the river Sabato - the Strait of Barba and Wolf Rock, due to their acoustic properties could have been the place of the legend.

worm 2022

Figure 1: Representation of the Witches’ Valley. Drawings by Piperino in De Nuce Maga

Beneventana.

3. A STRATEGIC PLACE THROUGHOUT THE HISTORY

The countryside of the actual Benevento was occupied by the Samnite tribes before they were domi- nated by Romans. With the realization of the Appia road, Benevento had a strategic position to be in the middle of an important commercial axis linking the Tyrrhenian with the Adriatic Sea. The legend of the witches took place with the invasion of the Lombards (Germans), who migrated to Italy due to famine and wars in northern Europe. They occupied also a part of southern Italy, establishing their center in the city of Benevento around 570 [7]. The Lombards brought their traditions, religion and habits, including their devotion to the gods of nature to whom they performed their rituals in the forests. One of their rituals was to dance around a tree whilst hitting with arrows a goat skin suspended from a branch; this action was completed by eating some parts of the animal in order to take posses- sion of the strength. Dances and songs accompanied by the rhythm of drums were the musical back- ground of the rituals. The show makers must have noticed in this particular place with flat rock walls, the sounds of their rituals were more amplified and otherworldly, caused by the acoustic reverberation of the area. It is possible that they considered this as a divine place due to the acoustics. It certainly amplified the effect of their rites to become more striking, mysterious, and emotionally engaging. The local populations, who remained tied to the Christian tradition, considered these rituals as pagan ceremonies, and for this reason the legend of witches began to spread throughout the city [8]. 4. DESCRIPTION OF TWO PLACES

Two sites have been selected to have high rock walls that can contribute to the formation of a sound tail by amplifying the sound reflections. The natural sites subject to be studied are the Strait of Barba, located on the border line between the lands of Benevento and Avellino, and the Ripa Lupina. These sites are identified as the main locations where the witches gathered for their magical ceremonies, according to the legend.

worm 2022

4.1. The Strait of Barba The Strait of Barba is located in the south countryside of Benevento and is a narrow valley about 2 km long, characterized by the river Sabato. Nowadays the valley is dominated by woods and vine- yards but during the end of the 19 th century the site was dominated by the activity of sulfur mines, which remained active until the 1960s. The sulfur mines were originated due to a type of stone found in this area that when lit, easily set aflame, which caused intrigue and contributed to the mystery of the valley of the witches. The Strait of Barba is characterized by the vertical smooth rock walls ar- ranged on parallel planes, which form acoustically reflective surfaces. Figure 2 shows the sound propagation within an environment characterized by parallel flat walls.

Figure 2: Scheme of flutter echoes created by parallel walls. The existing condition of the Strait of Barba is shown in Figure 3, highlighting the steep slope on both sides of the river.

worm 2022

Figure 3: View of the Strait of Barba

4.2 The Ripa Lupina The name of Ripa Lupina is indicated in the drawing of Piperno's book, as shown in Figure 1. This site is located in the borough of San Leucio del Sannio, a few kilometers from Benevento. Nearby are ruins of a defensive tower built by the Lombards and also in this area a statue of an Apis bull, found in the Middle Ages, suggesting the existence of an Egyptian temple nearby. Figure 4 shows the view of Ripa Lupina in its current state. The cliff is composed of a very steep rock having a height of about 50 m. At the foot of the cliff there is a side forest with a small stream collecting the rainwater.

Figure 4. View of the Ripa Lupina 5. ACOUSTIC MEASUREMENTS

Acoustic measurements have been carried out in order to analyze the characteristics of the identified valleys. Firecrackers have been used as impulsive sound sources while a Brahma microphone rec- orded the impulse responses (IRs). The sound source was placed at 1.4 m from the ground, while the microphone at the height of 1.6 m. The recorded IRs have been processed with Dirac 4.0 software package, obtaining the main acoustic parameters as defined by ISO 3382-1 [9-13]. Two sound source positions have been identified around the walnut tree, while the receiver has been moved in 6 and 5 locations along the Barba and Lupina valleys, respectively. Figures 5 and 6 show the location of the equipment installed during the acoustic measurements, in Strait of Barba and Ripa Lupina, respec- tively.

worm 2022

Figure 5. Location of the equipment during the acoustic survey in the Strait of Barba.

Figure 6. Location of the equipment during the acoustic survey in the Ripa Lupina. 6. RESULTS

The recorded IRs have been processed by using the plugin Aurora suitable for Audition 3.0. The main acoustic parameters considered in this research work include the reverberation time (T 20 ), clarity in- dexes (C 50 and C 80 ), and definition (D 50 ). These acoustic parameters are reported in the octave bands between 63 Hz and 8 kHz. Figures 7 to 12 show the graphs of the acoustic parameters obtained by the measurements, considered as the averaged values of all the measurement positions.

1.4

1.2

1.0

0.8

T20, s

0.6

0.4

Strait Barba

0.2

Ripa Lupina

0.0

10 100 1000 10000

Figure 7. Measured results of reverberation time T 20 . Figure 7 shows that the measured values of Strait of Barba are more reverberant than Ripa Lupina, reaching 1.3 s at 250 Hz, while the maximum value of T 20 for the Ripa Lupina is 1 s, found at 1 kHz. The variable line trends related to the two sites reveal a non-uniformity of sound energy across the spectrum. The lowest T 20 values have been found at 4 kHz in Strait of Barba and at 250 Hz in Ripa

Frequency, Hz

Lupina. This outcome is strictly related to the configuration of the sites that contribute to reach the echoes. This phenomenon can be clearly explained with the IRs related to the two sites, as shown in Figures 8 and 9.

Figure 8. Impulse Response related to the Strait of Barba.

worm 2022

Figure 9. Impulse Response related to the Ripa Lupina. The distribution of the sound energy is differently concentrated in the two sites, although the IR related to Strait of Barba record a clear echo after 15 ms from the direct sound, while this phenomenon is less visible in the graph of Figure 9.

35

30

25

20

C50, dB

15

10

Strait Barba

5

Ripa Lupina

0

10 100 1000 10000

Frequency, Hz

Figure 10. Measured results of speech clarity index C 50 . Figure 10 shows the measured values of the speech clarity index (C 50 ). The results related to Strait of Barba are found to be higher at very low and high octaves, approximately around 16-18 dB, while at 500 Hz they have been found to be the lowest, around 5 dB. This outcome means that only the mid octaves are meeting the optimal range for a good speech comprehension, while the low and high frequency bands are found to be too clear. In relation to Ripa Lupina, the measured values of C 50 are found to have a peak around 28 dB at 125 Hz and to have a response similar to Strait of Barba from 1 kHz onwards. The values of Strait of Barba and Ripa Lupina are found to be the same at 63 Hz, equal to 18 dB. The difference found with respect to the Strait of Barba is due to the morphological configuration of the sites.

40

35

30

25

C80, dB

20

15

10

Strait Barba

5

Ripa Lupina

0

10 100 1000 10000

Figure 11. Measured results of music clarity index C 80 . Figure 11 shows the measured values of the music clarity index (C 80 ). The results are very similar in line trend to the C 50 values; the main difference is that the maximum peak found at 125 Hz in Ripa Lupina is around 38 dB. This means that the low frequency of the drums would be heard as very clear in Ripa Lupina, while the drums of Strait Barba would be heard clear at low and high frequencies.

Frequency, Hz

1.0

0.8

0.6

D50

0.4

Strait Barba

0.2

Ripa Lupina

0.0

10 100 1000 10000

Figure 12. Measured values of definition (D 50 ).

Frequency, Hz

A good speech definition is defined to have values higher than 0.5 (50%). On this basis, the results obtained in both sites, as shown in Figure 12, are found to be above 0.5, specifically around 1.0 in Ripa Lupina across all the frequency bands; in Strait Barba a downward peak equal to 0.7 has been found at 1 kHz. 7. AURALISATION

It is possible to listen to the acoustics of the Witches Valley and of Ripa Lupina from within the Acoustic Atlas which is a real-time auralisation application running in the browser and connected to a growing archive of room impulse responses (RIRs) from natural and cultural heritage sites [14]. The reader can sing or emit sounds into the virtual acoustic Strait of Barba [15] and Ripa Lupina [16]. Headphones are required for listening and the computer or device microphone must be enabled. 8. CONCLUSIONS

Therefore, considering that the rites performed by the Lombards were accompanied by sounds gen- erated by the rhythm of drums, the acoustic reflections from the surrounding environment played a role in sound amplification and evoking a sense of awe and magic. The results highlight that the presence of rocks naturally amplifies the sound. The Strait of Barba, with its natural conformation of

worm 2022

smooth, steep rocks act as acoustically reflecting surfaces to create flutter echoes. Multiple echoes from each drumbeat could be perceived as a series of continuous impulses. Reverberation was used to create more impression, suspense and emotional involvement of the attendees. The analysis of the main acoustic parameters undertaken by this study consists of the identification of the locations where the legend of witches could take place, focusing on the particular acoustic characteristics that drove the Lombards to select Strait of Barba and Ripa Lupina for their ceremonies.

9. ACKNOWLEDGEMENTS

Recordings were conducted by Prof. Gino Iannace and this paper builds on the work he has previously conducted on the acoustics of the Witches Valley. REFERENCES

1. Reznikoff, I., and Dauvois, M. The sound dimension of painted caves (original in French). B.

Soc. Prehist. Fr., 85(8), 238–246 (1988). https://doi.org/10.3406/bspf.1988.9349 2. Waller, S. J. Sound reflection as an explanation for the context and content of rock art. Rock Art

Res. 10 , 91–101 (1993). 3. Reznikoff, I. Sound resonance in prehistoric times: a study of Paleolithic painted caves and rocks.

In: Proceedings Acoustics’08 , 4137–41. Paris, (2008). 4. Declercq, N.F., Degrieck, J., Briers R., Leroy O. A theoretical study of special acoustic effects

caused by the staircase of the EI Castillo pyramid at the Maya ruins of Chichen-Itza in Mexico. J. Acoust. Soc. Am ., 116(6), 3328-3335 (2004). 5. Iannace, G., Marletta, L., Sicurella F., Ianniello E. Acoustic measurements in the Ear of Dionysius

at Syracuse (Italy). Proc. of 39th International Congress on Noise Control Engineering , INTER- NOISE 2010, Lisbon, paper 580 (2010). 6. Piperno, P. De Nuce Maga Beneventana. 1639. 7. Brogiolo, G. P., Marazzi, F., Giostra C. I Longobardi. Un popolo che cambia la storia. Skira

(2017). 8. Jarnut J. Storia dei longobardi. Einaudi. 2002. 9. ISO 3382-1: 2009 “Acoustics – Measurement of the reverberation time of rooms with reference

to other acoustical parameters”. 10. Trematerra, A., Iannace, G. The acoustics of the catacombs of San Callisto in Rome. Proceedings

of Meetings on Acoustics , 20(1), 015001, (2014). DOI: 10.1121/1.4861195 11. Iannace, G., Trematerra, A. The acoustics of the caves. Applied Acoustics , 86, 42–46 (2014). 12. Iannace, G., Berardi, U. The acoustic of Cumaean Sibyl. Proceedings of Meetings on Acoustics ,

30(1), 015010, (2017). DOI: 10.1121/2.0000606 13. Lombardi, I., Iannace, G., Trematerra, A. The acoustic of a courtyard. Pro c. of INTER-NOISE

2018 - 47th International Congress and Exposition on Noise Control Engineering , 2018 14. Van Tonder, C., Acoustic Atlas. [Online]. Available: https://acousticatlas.de . [Accessed: 1-Apr-

2022]. 15. Iannace G., Van Tonder C., Witches Valley: Strait of Barba Real-Time Auralisation. [Online].

Available: https://www.acousticatlas.de/locations/witches_valley . [Accessed: 1-Apr-2022]. 16. Iannace G., Van Tonder C., Witches Valley: Ripa Lupina. [Online]. Available:

https://www.acousticatlas.de/locations/Wolf_Rock_Witches_Valley2. [Accessed: 1-Apr-2022].

worm 2022