How to test for a ritual: The case of whispering

J. Cirillo and D. Todt

Institute of Biology, Free University, Berlin, Germany

Recent research on the social use of whispered words suggested to explain this vocalisation as a ritualised form of normal speech. Here we summarise an approach designed to test this explanation. Our methodological framework was given by a combination of classical and modern concepts (criteria of ritualisation; theory of honest signalling). The cited theory predicts that ritualised signals should be more costly than their originals. Referring to such prediction, properties of whispering (unvoiced speech) were compared to properties of normal speaking (phonated speech).

In a first line of study we investigated the production efforts of verbal utterances and measured the number of syllables which adult subjects were able to produce by the air of one single breath. Comparisons showed that the amounts of whispered syllables were clearly smaller than the amounts of syllables spoken normally. Our findings documented that whispering is more arduous than normal speaking, and thus also more ‘costly’. In a further approach, we investigated the early ontogeny of speaking abilities. Here, analyses of data collected in different kindergartens documented that the development of whispering requires a prior training and competence in speaking with a phonated voice. Although a few children could produce some of their words in whispered versions already at an age of about two years, spontaneous interactions by whispering were observed only in children of about four to five years of age; i.e. when children have developed their own concept of secrecy and privacy.

Taken together our results confirm that whispering can indeed be regarded as a ritualised form of normal speaking. Parallely they allow to conclude that the procedure applied here makes an expedient methodological tool of testing signals for specific ritual properties.

By combining classical procedures with modern methods of signal analyses, this paper advertises a ‘comeback’ of studies on ritualisation. Such studies dominated the ethological research for long, but are rare currently.

Paper presented at Measuring Behavior 2005 , 5^th International Conference on Methods and Techniques in Behavioral Research, 30 August - 2 September 2005, Wageningen, The Netherlands.

© 2005 Noldus Information Technology bv