Bell Shape Embodying Zhongyong: The Pitch Histogram of Traditional Chinese Anhemitonic Pentatonic Folk Songs

Round 1

Reviewer 1 Report

The paper discusses the observation that traditional Chinese Anhemitonic Pentatonic songs (TCAPS) follow a bell shaped distribution of pitches. The idea is sound, the paper is well written, and the contribution is very relevant.

Nevertheless, the paper has a structural problem, probably reflecting a severe methodological issue.

In particular, the first half of the paper just discusses qualitative observations of the histograms, without any quantitative measures of the "bell shape" fitness. If presented in this way, those (reasonable) observations may only be considered as mere speculations.

Quantitative measures are only provided when proposing some machine learning (ML) application, aimed at classifying TCAPS VS other songs.

In other words, the paper completely lacks a rigorous statistical analysis of the data, aimed at verifying the hypothesis (the ML work may be considered just as a very convoluted and not-so-rigorous way to do it).

For this reason, even if I see authors point, I feel uncomfortable with their conclusion that they "discovered that the pitch histogram of the most representative traditional Chinese anhemitonic pentatonic folk songs [...] naturally obeys a bell-like shape"

What I suggest to do is to:

- introduce the "bell shape" fitness features before discussing the histograms

- support the discussion of histograms with quantitative measures

- run appropriate statistical test to verify the hypothesis (TCAPS VS Others).

Other minor issues are:

- all histogram should share the same x axis;

- a more rigorous feature selection strategy in the ML part is called for;

- the use of histograms in education is almost entirely neglected in the paper, I suggest removing it from the title.

Author Response

Dear Reviewer, we sincerely thank you for taking the time to review our article and providing so much insight! We will carefully explain the doubts and revise the article regarding the issues and concerns you have identified. Please refer to the attached document.

Author Response File: Author Response.pdf

Reviewer 2 Report

From the perspective of musicology, there are two major issues in the study design: 1. the note's durations are not taken into account. It will make a significant difference in perceving melodies whether pitches in question appear as longer or shorter notes; 2. if truly traditional folk songs are created by improvisation and handed down orally (I agree to this point), transcriptions of the songs merely are unreliable representations of what the transcriber has heard (or believes he or she has heard). For both reasons it was far more convincing if you apply sound analysis (e.g. using PRAAT) to field recordings.

I am a bit irritated about the debonair use of the term "ethnic" which is avoided in cultural studies since it is hegemonic. Music is culturally defined, not ethnically.

It is highly disputed whether pentatonic scales are present in different cultures by cultural exchange or independently present.

There is a mistake in the introduction. "Do re mi sol la" on a piano's black keys work in F# major, not in C# major.

Modes both in Chinese and in Indian music theory incorporate much more than pitch orders. They are also associated with certain daytimes, performance practices, emotional situations and even wheather conditions. One has to be very careful in comparing modal concepts in different cultures and it has to be clearly stated if one refers to the modal pitch orders only.

Author Response

Dear Reviewer, first of all, let us give our heartfelt thanks for your time and effort in reviewing our articles and providing high-value comments! Please refer to the attached document.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The paper quickly and reportedly improved, now only one last boring but necessary revision is called for:

In the whole Section 2 numerical values of the features should be provided for all discussed histograms. Section 2.3 can then become the closing paragraph of section 2.2, which in turn can be renamed accordingly.
More in general, whenever the "bell-ness" of a histogram is discussed or compared with others, numerical values should be reported.

A minor suggestion: Section 4.3 can be safely moved as final subsection of Section 2, as a validation of what author says previously.

Author Response

Thank you very much for your efficient and informative review! They are beneficial for further improving the article's quality. Please check the revision report in the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Thank you for your detailed report. I am still not entirely convinced by the research design. However, this may be due to the fact that a humanities' scholar is led by other cognitive interest than applied sciences' scholars.

Author Response

Yes, we fully understand your concerns. For computer science researchers like us who are more used to "pattern recognition", it is imperative to listen to experts in the field when we investigate a humanistic topic.

Our article attitudes an early result of "observation-findings-model-preliminary confirmation" that we do not claim ultimate and universal validity.
In order to leave room for confirmation or falsification by scholars in the field, we have weakened the conclusions (many instead of most):

"We discovered that the pitch histogram of MANY representative traditional Chinese anhemitonic pentatonic folk songs known well by Chinese people from all regions nowadays naturally obeys a bell-like shape… Subsequently, we suggested that the merged pitch histogram and the anhemitonic pentatonic pitch histograms can reasonably eliminate the exceptions in the pitch histograms to endow many traditional Chinese anhemitonic pentatonic folk songs with perfect bell shapes…"

Once again, thank you from the bottom of our hearts for your dedication and great respect for your professionalism!