Abstract
This study, integrating linguistics, computer science, statistics, musicology, and cultural study, explores how the lexical tones of Mandarin Chinese interact with rap music. We develop an algorithm to compute the Tonal Congruence Index (ITG), measuring the extent to which tones are preserved in vocal music. Analysis of a dataset with over 8000 syllables from Mandarin Chinese rap songs shows that the tonal congruence is related to subgenres that have evolved over time: while tones are largely preserved in the earlier boom-bap style, tonal distortion occurs more frequently in recent trap music. Moreover, we find that the falling fourth tone in standard Mandarin accounts for 38% of line-ending syllables in the lyrics of 140 rap songs and further suggests that the pre-eminence of the fourth tone may be a generic feature of Mandarin Chinese rap, particularly in boom bap. In addition, Chinese rappers have increasingly utilised English words over time to achieve musical flow. Integrating analytic findings with qualitative interviews of five national award-winning Chinese rappers, we conclude that the restriction of linguistic tones in Mandarin Chinese rap has gradually weakened over time and that there is a decreasing correlation between tone and rap with the emergence of new rap styles such as mumble rap and melodic rap.
Acknowledgements
The authors are grateful for the two anonymous reviewers’ extremely detailed and helpful comments that significantly informed the revisions of this article. This research is supported by Ivan Allen College of Liberal Arts ‘Small Grants for Research’ funding and the Vertically Integrated Projects (VIP) Program at Georgia Tech. We thank Kyle Liang for developing an open-source tool called Tone Analyzer (see section 3.2.4) and for doing most of the calculations. We would very much like to express our gratitude to Mr. Liu Xiaofu (aka. Xiaoqiang shushu) and Mr. Yu Shaohua for introducing Chinese top rappers and for facilitating the interviews. We thank the rappers, Li Daben, Aire, Knowknow, Huang Xu, and Ice Paper for taking time to conduct the interviews, and their music labels, ModernSky (MDSK), Dream Music Group (D.M.G), and Free-out for their generous support and assistance to our research.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Correction Statement
This article was originally published with errors, which have now been corrected in the online version. Please see Correction (http://dx.doi.org/10.1080/09614524.2024.2374098).
Notes
1 The controversy over the relationship between dialect and language is a global and often politicized problem. Here we use the traditional term ‘Chinese dialects’ to translate the Chinese term fangyan (lit. ‘regional speech’).
2 Chou made this remark during a performance of his ‘The Invincible’ world tour in Taipei on September 30, 2017.
3 Duinker (Citation2021) warns that Genius is open-source, and its reliability is not infallible. In order to validate the use of lyrics from the Web, we compare two calculations, one based on the line structure from Genius.com, and the other from manual identification of lines of lyrics. Although the total numbers of end rhymes are different, the percentage of tonal distribution is very similar, which is what this study is concerned with.
4 Condit-Schultz (Citation2016b, p. 140) defines musical phrases ‘by the appearance of either a /3/ or a /4/ in the prosodic **break spine,’ but he also recognises that his approach is ‘very simplistic, doing little justice to the complexity of phrasing in rap flow.’
5 In this study we arbitrarily set the cut-off for the number of pitch samples as five. Admittedly, this is a working assumption. Since we retrieve a Hz value every 1 millisecond (the default time step in Praat), a very small number indicates the possibility of a sampling error.
6 Our code can be found on Github at https://github.com/amandarshe/pitch_analysis.
7 See Bradley (Citation2009, pp. 48–56), Edwards (Citation2009, pp. 99–104), Condit-Schultz (Citation2016a, pp. 51–71) and Duinker (Citation2020, pp. 105–111) for more comprehensive taxonomies of rhyme types in hip-hop music.
8 This tool can be found on Github at: https://github.com/crown523/tone_analyzer.
9 The periodization of old-school and new-school rap is different in the United States, where the dividing line is ‘typically placed in the mid to late 1980s’ (Condit-Schultz, Citation2016b, p. 124).
10 This reminds us of the term ‘pitch polarity,’ by which Carter-Enyi (Citation2021, p. 17) refers to the juxtaposition between phrase endings of high and low pitch. He suggests that pitch polarity is not limited to tonal language vocal arts, such as the Yorùbá culture that he studies, but pervasive in American rap as well.
11 Gai renders the song in Chongqing Mandarin. According to Qing (Citation2014), the tonal value of T1 in Chongqing Mandarin is controversial. While traditional fieldwork investigations regard T1 as high level 55, recent acoustic results yield a high rising 35, particularly among the younger generation. Here it is better to interpret the effect of tian in Wei Ran’s example as that of the traditional even tone pingsheng (T1 and T2 in Mandarin).