Generational differences in the low tones of Black Lahu

We investigate apparent-time tone variation in the Black Lahu language (Loloish/Ngwi, Tibeto-Burman) of Yunnan, China. These are the supplementary materials for the paper "Generational differences in the low tones of Black Lahu," accepted for publication in Linguistics Vanguard. 

Appendices: 

Appendix A: Wordlist (organized by order of appearance in the story)

Appendix B: Wordless picture book

Appendix C: Cross-tabulation tables

Appendix D: F0 trajectory plots by speaker

Appendix E: LME model results for T45

Script files contained in the analysis: 

F0_estimation.praat is the Praat script used for F0 estimation on the wav + Textgrid pairs

Combine_Speaker_Files.R is used to combine the Praat script output for individual speakers into a single csv file.

Data_processing.R is the main processing script, works on an Excel csv file that contains the output of the Combine_Speaker_Files.R script 

Plotting_Fig1_Lahu_tones.R plots the F0 trajectories of all the tones of Lahu, averaged across all the speakers, works on an Excel csv file that contains the output of the F0 estimation.praat script. 

Plotting_Fig2_age_groups.R plots the F0 trajectories of T2 and T7, in 15-age groups, works on an Excel csv file that contains the output of the Data_processing.R script. 

Plotting_Fig3and6_carryover. R plots the F0 trajectories of T2, T7 and T4 when they occur after silence or Tones 1-7 in three age groups; works on an Excel csv file that contains the output of the Data_processing.R script. 

Plotting_Fig4578_scatterplot.R plots the scatterplots of F0 onset versus Age, works on an Excel csv file that contains the output of the Data_processing.R script. 

Data files contained in this analysis: 

praat_raw_data.csv is the output of the F0_estimation.praat script. Contains the raw data from all speakers.

Speaker_info.csv is the demographic data for each speaker

data_forplotting.csv is the output of the Data_processing.R script. Contains the filtered tokens, normalized for length and converted to speaker-specific semitones. Used to plot Figure 2, Figure 3 and Figure 6

data_forRbrul.csv is the output of the Data_processing.R script, the F0 onset and F0 offset for Tones 2, 4, and 7. Used to do linear mixed effects modeling in the Rbrul interface (Johnson 2009).