The Influence of Hindustani Classical Music on Bollywood Songs. A Statistical Outlook

Table of Content

PREFACE

List of Tables and Figures

CHAPTER 1: INTRODUCTION TO HINDUSTANI CLASSICAL MUSIC
1.1 Introduction
1.2 Principles of Hindustani music
1.2.1 Terminologies Used In Hindustani Classical Music
1.2.2 Notation Used in Describing Ragas
1.3 Raga Yaman (Evening Raga)
1.4 A Statistical Outlook
Notation
1.5 Objectives

CHAPTER 2 LITERATURE REVIEW

CHAPTER 3 METHODOLOGY
3.1 Research Methodology
3.1.1 Bandish
3.1.2 Mann Re

CHAPTER 4 RESULTS & DISCUSSION
4.1 Results
4.1.1 Statistical Analysis of a Bandish
4.1.2 Statistical Analysis of the Song Mann Re
4.2 Interpretation from the figures
4.3 Comparison:
4.4 Note Duration and IOI
4.4.1 Graphs (note duration and IOI)
4.4.2 Discussion

CHAPTER 5:SOME FAMOUS BOLLYWOOD COMPOSERS AND SINGERS AND AN ILLUSTRATIVE LIST OF RAGA BASED SONGS COMPOSED OR SUNG BY THEM

CHAPTER 6 CONCLUSION

REFERENCES

PREFACE

Any music originates in the society and develops with the changing realities of it. It accepts new and modifies the existing cultural norms in different periods of time. This process of acceptance and rejection makes any form of art exist for long. Inspite of all this, in various phases, Hindustani classical music, being the base of many popular Bollywood songs has helped in their popularity and lifelong existence because of the strong focus on melody. A raga, which is the nucleus of Indian classical music-be it Hindustani or Carnatic-is a melodic structure with fixed notes and a set of rules which characterize a certain mood conveyed by erformance. Hindustani ragas have embraced the elements of several Bollywood songs, which has given these songs a strong impact despite the strong influence of western art music in Bollywood music industry. The present work attempts to study this impact in a legitimate manner using a statistical approach emphasizing on statistical modeling of musical structure and performance and other statistical features such as note duration and inter onset interval with a case study in raga Yaman. It turns out that the same statistical model for both the raga bandishand a song based on the same raga, i.e., Yaman, an evening raga of the Kalyan thaat.

Amrita Singh

Soubhik Chakraborty

List of Tables and Figures

Figure 1 Double Exponential Smoothing Plot for C1

Figure 2 Residual Plots for C1

Figure 3Double Exponential Smoothing Plot for C5

Figure 4 Residual Plots for C5

Figure 5 Note Duration and IOI of the song Mann Re

Table 1 Middle Octave

Table 2Lower Octave

Table 3Higher Octave

Table 4 Note sequence of bandish based on Yaman followed by pitch value

Table 5Data of fundamental frequency in Hz (Pitch value) for song Mann re

CHAPTER 1: INTRODUCTION TO HINDUSTANI CLASSICAL MUSIC

This chapter deals with exploring and analyzing the phenomenal incorporated in Hindustani classical music followed by its principles and statistical outlook. It also elaborates the aim of the project.

1.1 Introduction

HINDUSTANI CLASSICAL MUSIC: Hindustani or North Indian classical music (also known as ‘Shastriya Sangeet’ ) is the traditional music of the Indian subcontinent [1]. It was originated from the 12th century CE, when it diverged from Carnatic music, i.e., the classical tradition of southern regions of the Indian subcontinent. Hindustani classical music mainly evolved in North India around the thirteenth and fourteenth centuries A.D. It owes its development to the religious music, as well as popular and folk music, of the time. It has strongly influenced by Bollywood music especially in instrumentation, melody and beat. Indian classical music from its inception to the present has gone through various phases of transformations and transitions. In the process it has embraced and rejected many features while maintaining an absolute balance with the old age tradition. Hindustani classical music is a heritage that has evolved through the centuries. It is a blend of ritualistic, folk and cultural expression of the sub-continent and represents music of different genres. At one extreme, it is classical music whilst at the other extreme; it is a mixture of musical genres of different regions that reflect the diversity of India. Hindustani classical music is an Indian classical music tradition that took shape in northern India in the 13th and 14th centuries A. D. Its origins lie in existing religious, folk and theatrical performance practices. The origins of Hindustani classical music can be found in the Samaveda (wherein Sāman means "melody" and Veda means "knowledge"). The Samaveda comes second in the usual order of the four Vedas. There is division of classical music, first is melody and second is rhythm. There are seven basic notes with five interspersed half-notes, resulting in a 12-note scale. Most of the old melodies are based on ragas. Music which was very much of India has easily crossed the frontiers and reached almost every corner in the world. Indian music got appreciation and recognition in the world musical platforms. Simultaneously Indian music accepted many of the western components and Indian music also became a part of the western musical circles. Ragas and many other form of Indian classical music began to influence many rock groups. Indian music is traditionally practice-oriented and taught by teachers through an oral tradition. Until the 20th century, it did not employ notations as the primary media of instruction, understanding or transmission. The rules of Indian music and compositions themselves are taught from a Guru to a shishya under the guru-shishya parampara or the teacher-student tradition. An important landmark in Hindustani music was the establishment of gharanas (style and content of singing) under the patronage of princely states. A gharana is more a school of thought rather than an institution. Each gharana developed distinct facets and styles of presentation and performance. Indian classical music has one of the most complex and complete musical systems ever developed in the history of mankind. It divides the Saptak (octave) into 12 swaras or semitones (5 shudha + 4 komal+1 tivra + 2 sthira) out of which the seven basic notes are Sa, Re, Ga, Ma, Pa, Dha and Ni, in that order.

The melodic foundations for improvisation akin to a melodic mode in Hindustani classical music are called ragas [3]. One possible classification of ragas is into “melodic modes” or “parent scales”, are known as thaats, under which most ragas can be classified based on the notes they use. Ragas are particular ascending and descending of notes. There is a condition for raga that it must have at least five notes (of which Sa has to be there being the tonic and one of Ma and Pa has to be there). The origin of the raga may be from any source including religious hymns, bhajans, folklore, folk tunes and music from outside the Indian subcontinent. We can compose a poem or story, colours for a nice painting using words; like that raga is a composition of musical notes. A raga may be defined as a melodic structure with fixed notes and a set of rules that characterize a certain mood conveyed by performance. For the performance of the ragas, they are claimed to have specific timings of the day and night such as ragas for morning, ragas of the noon, afternoon, evening and ragas of the night. Raga is the nucleus of Indian classical music, be it North Indian (Hindustani) or South Indian (Carnatic). The “certain mood” refers to the emotional content that is typical of the raga. The rules (like how a particular note or note combination should be used in the raga, the note sequence allowed in ascent or descent, etc.) help in building the raga mood and are not meant to handicap the artist. In fact, the artist has infinite freedom to express himself/herself despite these rules! Although it may be possible to associate a part of the emotional content of the raga with an identifiable human emotion such as sadness (“karuna rasa”), it hardly tells the full story. Furthermore, the raga is not just a collection of the notes that are allowed to be played in a piece of music. There are also rules governing how the notes may be used; for example, the notes used in an ascending (arohi) scale may be different from the notes in a descending (awarohi) scale. Some notes may be considered “main pitches” in the raga, while others are used in a more ornamental way typical of the raga. The raga may even affect the tuning of the piece.

1.2 Principles of Hindustani music

TheGandharva vedais a Sanskrit scripture describing the theory of music and its applications in not just musical form and systems but also in physics, medicine and magic[4]. It is said that there are two types of sound:āhata(struck/audible) andanāhata(unstruck/inaudible). The inaudible sound is said to be the principle of all manifestation, the basis of all substance. There are three main octaves: low (mandra), medium (madhya)and high (tāra).Each octave resonates with a certain part of the body, low octave in the heart, medium octave in the throat and high octave in the head.

1.2.1 Terminologies Used In Hindustani Classical Music

A) Raga: The heart of Hindustani classical music is the raga. As mentioned earlier, a raga may be defined as a melodic structure with fixed notes and a set of rules characterizing a certain mood conveyed by performance. The following are the characteristics of the raga. A raga consists of a fixed set of five or more musical notes (so this is one of the rules!). Ragas (in Sanskrit it is known as color or passion) are supposed to evoke various moods in the listener. In Hindustani music, there are certain ragas which are specific to different season or time of the day. Monsoon ragas belong to the Malhaar group, though they are mainly performed during the rains, while morning ragas, such as Bibhas and Bhairavi, and night ragas, such as Kedar and Malkauns, are suitable for rendition in morning or night, respectively. We also have afternoon ragas like Bhimpalashree and evening ragas like Yaman. Also, Hindustani classical music classifies ragas into ten parent raga groups called thaats, as organized by Vishnu Narayan Bhatkhande in the early 1900s.

1.2.2 Notation Used in Describing Ragas

Notation is the art of describing musical ideas in written characters or symbols. Indian classical music has seven basic notes and is called shudh (natural or pure) swara. They are shadja, rishabha, gandhara, madhyama, panchama, dhaivata, and nishada. In short form, they are known as Sa, Re, Ga, Ma, Pa, Dha, and Ni. This group of Indian notes is called a saptak (seven notes of diatonic scale). There are three types of saptak:

– Mandra/mandar (lower octave)
– Madhya (middle octave)
– Tara (higher octave)

In addition, there are four komal (soft or flat) notes (Re, Ga, Dha, and Ni) and one teevra/teebra (sharp) note (Ma), thus making a total of 12 notes in the chromatic scale. These five notes are called modified (vikrita) notes. In this book the notes (swara) used in representing the ragas are: Small letter indicates a komal swara as komal Re is r, etc., and capital letter for sudh swara as sudh Re is R, etc. (see footnote 1), with only exception of m as tivra Ma (as M represents sudh Ma). Further, a note in italics, normal, and bold font stands for the notes in lower octave, middle octave, and higher octave,respectively.

Further the ragas are described with following terminologies:

– Thaat (parent scale)
– Jati (class)
– Aroha (ascent) and avaroha (descent)
– Vadi (most important note) and samvadi (second most important note)
– Peshkash (rendition)
– Rasa (aesthetic joy or emotion)
– Pakad (catch or grip of the raga)

Let us take a look at each of these musical terms.

– Thaat :This is a method of grouping of ragas according to the specific notes used. Two ragas using the same notes will be placed in the same thaat even if their melodic structures, mood, and emotions are different. The ten thaats in Hindustani classical music are Kafi, Bilaval, Purvi, Asavari, Todi, Khamaj, Kalyan, Bhairav, Marwa, and Bhairavi. Some ragas like Ahir Bhairav and Charukeshi do not fall into any of these thaats.
– Jati: Jati literally means “caste.” Just as there are castes in any community in India, there are three castes or classes of raga. There is Arava/Audava/Oudava, pentatonic (five notes); Sharva/Shadava, hexatonic (six notes); and Sampoorna, perfect heptatonic (seven notes). Thus an Aurabh-Sampoorna raga means five distinct notes are allowed in ascent, seven in descent.
– Aroha and avaroha :They depict the sequence of permissible notes in ascent and descent, respectively. They help in characterizing the mood of the raga.
– Vadi swara: This is the most important or dominating note in a raga, which is a sort of key to the unfolding of its characteristics. As it is the pivotal note, it is played very prominently or repeatedly. In it lies the particular rasa of that raga. It also determines the time for the singing of the raga. If the vadi swara falls in the first half (Sa to Pa), then the raga is called Poorvang Pradhan. If it falls in the second half (Ma to Sa), then the raga is called Uttaranga Pradhan. Ma and Pa are common to both the halves. If the vadi swara is Ma or Pa, expert’s guidance is needed to decide whether the raga is Poorvang or Uttaranga Pradhan.
– Samvadi swara: This is the second important note in the raga, after the vadi note. Its position is at an interval of fourth or fifth from the vadi note. It has the same status as a minister in relation to a king represented by the vadi note.
– Peshkash: Classical musicologists have assigned a specific time to the performance of a raga. This has been based on the types of swara (notes) used in a particular raga.

The main architect of the present system of Hindustani music is Pandit V N Bhatkhande, who was responsible for the classification of the Ragas into the 10 'thaats'. There are total 83 parent ragas in Indian classical music. Out of 83 ragas, we have chosen Raga Yaman (an evening Raga ).

1.3 Raga Yaman (Evening Raga)

It is a sampurna Indian classical raga of Kalyan thaat. It emerged from the parent musical scale of Kalyan. The first author adds: “During my vocal training classes, raga Yaman was the first raga taught to me because it is one of the most fundamental and grandest ragas in Hindustani tradition” [1].It is performed during the 1st quarter of the night. Taking its origin from Kalyan thaat, Raag Yaman demonstrates the various emotions or rasas such as happiness, devotion (bhakti) and peace (shaant). The raga has all shuddha swars except for the tivra Ma. It is this note which gives the raga its distinctive quality of peace, and tranquility. Some modernists consider Yaman, Yaman-Kalyan, and Kalyan to be one raga and only traditionalists consider these as three distinct ragas.

Bandish based on raga Yaman is taken from book (Bhatkhande, V. N.: Hindustani sangeet paddhati kramik pustak mallika) and the bandish is (SADA SHIV BHAJ MANAA….)[1] in which the

Vadi swar for Raga Yaman : G

Samvadi swar : N

Vikrit swar: m tibra

Aaroh: N R G, m P, D, N S

Avroh: S N D, P m G, R S

Prahar: Late evening

Pakad (catch phrase): N R G, R m, m P D, D N S

Similarly, the Bollywood song based on raga Yaman which is to be compared with bandish is Mann Re is in the form of audio signal recorded by the second author for research purpose (originally sung by Mohammad Rafi in the Bollywood movie Chitralekha (1964)). Sahir ludhianvi wrote the the lyrics of this song while its tune was composed by Roshan.

To have a clear idea about the influence of Hindustani classical music on Bollywood music, we have chosen one Bollywood song (Mann re) based on raga Yaman and one bandish (Sada shiv bhaj manaa…) from the book (Bhatkhande:Hindustani sangeet paddhati kramik pustak mallika).

1.4 A Statistical Outlook

One of the strengths of statistics lies in modeling. Modeling a musical structure or a musical performance has been a coveted research area in computational musicology. There are three fundamental steps in statistical modeling: deciding which model to fit, estimating the parameters of the chosen model, and verifying the goodness of fit of this model. We all know that statistics can be broadly divided into two categories: descriptive and inferential. In statistical modeling, both are involved—as we first describe a pattern (through modeling) and then infer about its validity. Two types of models are used in statistics: probability models and stochastic models. Through a probability model, we can tell the probability of a note or a note combination but cannot predict the next note. Through a stochastic model, we can predict (make an intelligent guess of) the next note, given the previous.

The above mentioned bandish and song based on raga Yaman are both compared on the basis of plottings obtained by applying double exponential smoothing in Minitab Statistical package .

Double exponential smoothing: A major strength of statistics lies in modeling. Modeling a musical structure or performance is both an interesting and challenging endeavour given that the true model is not only complex but unknown even to the composer. It may also be referred as second order exponential smoothing, This is being termed double exponential smoothing because it is the recursive application of an exponential filter twice[6]. It employs a level component and a trend component at each period. It uses two weights, (also called smoothing parameters), to update the components at each period. Basic idea behind double exponential is to introduce a term to take into account the possibility of a series exhibiting some form of trend. It is a general smoothing method and in order to provide short term forecasts when our data have a trend and do not have seasonal component.

The double exponential smoothing equations are as follows:

Abbildung in dieser Leseprobe nicht enthalten

Where Yt is the observed phenomenon (here the pitch characterizing the musical note) at time t. The cap (^) is used to indicate the predicted value from the model.

Double exponential smoothing employs a level component Lt and a trend component Tt at each time period t. It uses two weights, or smoothing parameters α and γ, to update the components at each period. If the first observation is numbered one, then level and trend estimates at time zero must be initialized in order to proceed. The initialization method used to determine how the smoothed values are obtained in one of two ways: with optimal weights or with specified weights.

Notation

Abbildung in dieser Leseprobe nicht enthalten

Therefore, here we have applied double exponential smoothing in Minitab software to obtain the graphs of bandish and Bollywood song (which was first put into Praat software to get their pitch values).

Minitab is a complete statistical software package developed at the Pennsylvania State University by researchers Barbara F. Ryan, Thomas A. Ryan, Jr., and Brian L. Joiner in 1972. which provides convenient features that streamline our workflow, a comprehensive set of statistics for exploring our data, and graphs for communicating our success. Minitab provides the tools you need to analyze data and find meaningful solutions to our toughest problems. Statistical analysis software such as Minitab automates calculations and the creation of graphs, allowing the user to focus more on the analysis of data and the interpretation of results.

Similarly another software used for the analysis called ‘Praat’ which is a free computer softwarepackage for the scientific analysis of speech in phonetics. Mann re song used in the project is in the vocal form and that is why first it was analyzed in Praat for having the pitch values, after that those pitch values are used in Minitab software for the fitting of double exponential smoothing.

1.5 Objectives

Everyone is fond of music. Some people like traditional music and some like western songs. But only few people know the origin and history of these genres of music. So the basic aim behind making this project is to show that inspite of creation of new songs, why the old songs retain their popularity and why they sustain for a long time. Any music originates in the society and develops with the changing realities of it and are acceptable also but the reality is Hindustani classical music is the base of many of the Bollywood songs as also of other genres like ghazals, Rabindra Sangeet (Tagore songs) etc. It is Indian music which is acclaimed and accepted in many of the western and asian countries. Ragas and many other form of Indian classical music like Tagore songs influence Bollywood music strongly.

So the purpose is to show Indian classical music is a heritage of all that has influenced other songs and has evolved through the centuries in the form of melody. Therefore by using statistical tools such as modeling musical structure and purpose e.g. with double exponential smoothing, one can show the strength of lifelong existence of Hindustani classical music as compared to Bollywood music which is obviously not older.

CHAPTER 2 LITERATURE REVIEW

Chakraborty et al. [2]-[3], reported that a major strength of statistics lies in modeling. Modeling a musical structure or performance is both an interesting and challenging endeavour, given that the true model is not only complex but unknown even to the composer. On the other hand, although statistical models are neither perfect nor unbiased, it should be understood that i. we can at least make the data objective or nearly so ii. the true model may have multiple parameters and we usually do not have explicit knowledge about them nor we know how or in what functional way they enter the model and iii. doing a stochastic realisation of this deterministic true model (the decision process of the composer is deterministic as any musical sequence of notes is planned and not random) is within the scope of statistics including controlling the errors in the model. For example in Single Exponential Smoothing, α is the smoothing factor. This is the only parameter in the model that needs to be determined from the data. The term smoothing factor applied to α here is something of a misnomer, as larger values of α actually reduce the level of smoothing. It showed that smoothing technique is useful when there is a trend, there is no seasonal component, there is no missing value, we want short term forecast. The model equations for Single Exponential Smoothing are Ft+1 = αYt + (1-α)Ft, 0<α<1; F0=Y0 where Ft is the predicted value corresponding to the observed Yt at time t.

Chakraborty et al. [2]-[3] investigated the significant relations between the analytical structure of classical Western compositions and the tempo curves of human performances. It immensely helped music enthusiasts who have knowledge of Western classical music but are new to Indian music. The authors talks about the role of statistics in computational musicology which was described by RUBATO, the music software for statistical analysis. The authors teach us how to analyze a musical structure using a statistical approach. The book [3] was written with the sole objective of promoting computational musicology in Indian music, and is primarily aimed at teaching how to do music analysis in Indian music. It assumes that the musical data is already available. A recent book which deals with how to acquire the musical data using signal processing in the context of Hindustani classical music is by Datta et. al. [25].

Kalekar’s [6] paper concentrates on the analysis of seasonal time series data using Double exponential smoothing methods. This method is used when the data shows a trend. Exponential smoothing with a trend works much like simple smoothing except that two components must be updated at each period - level and trend. The level is a smoothed estimate of the value of the data at the end of each period. The trend is a smoothed estimate of average growth at the end of each period. The motivation behind using the adaptive technique, as opposed to the non-adaptive technique is that, the time series may change its behavior and the model parameters should adapt to this change in behavior. Tests carried out on some standard time series data corroborated this assumption.

Vedabala[4] reported that any music origins in the society and develops with the changing realities of it. It accepts new and modified the existing in different periods of time. This process of acceptance and rejection makes any form of art exist for long. India is known for its rich musical heritage around the globe. There are numerous forms and genres of music. Among which the most respected is Indian classical music, be it Hindustani or Carnatic. The music represents an exemplary standard and long established principle or style based on methods developed over a long period of time. Some technological innovation that have influenced the classical music are upgradation of recording/archiving technology, virtual music classes and online availability of music. Music, being one of the indivisible aspects of society, it cannot devoid itself from the changing realities of time. It has accepted the undesirables and rejected desirables in various phases of its evolution. The contradiction can be dealt with maintaining balance between the trend and the tradition. Technology mixed with traditional values can be an absolute advantage to the antiquated institution of music.

Hampiholi [7] wrote that in the past decade in the field of audio content analysis for extracting various information from audio signal, a lot of research has been done. One such significant information is the ”perceived mood” or the ”emotions” related to a music or audio clip. This information is extremely useful in applications like creating or adapting the play-list based on the mood of the listener. This information could also be helpful in better classification of the music database. The author has presented a method to detect mood in Indian Bollywood music based on Thayer’s mood model comprising four mood types, Exuberance, Anxiety, Serene, and Depression. Audio features related to Intensity (Energy), Timbre (which depends on the spectrum and envelope of the music signal), and Rhythm (Tempo) are extracted from the musical data. Music classification model based on mood is arrived at using machine learning technique.

Rodriguez [10] proposes a method for analysing the differences between the “consumption” (acceptance and acclaim) of popular and classical music. By using the information contained in the Survey of Structure, Conscience and Biography of Class, the authors estimate a bivariate probit model to characterize the audience for each kind of music; the authors then quantify the influence exerted by various socioeconomic features on the demand for these goods and we describe the average profile of consumer. To achieve this task the authors use a bivariate probit model. This model has an important advantage: it allows them to estimate simultaneously two equations that represent two decisions and to discover if there is a significant correlation between their random disturbances. In our case, therefore, they can identify the principal features of classical and popular music listeners, find if there are common characteristics and discover how similar they are. The data used to estimate this model comes from the structure. This survey combines information about individual consumption of cultural goods – including classical and popular music – with various social, demographic and economic variables of the interviews. The authors analyse the relationship between these two types of music listeners, using a bivariate probit model. They have tested the presence of a positive and statistically significant correlation between classical music and popular music listeners. So the authors can identify the presence of a common background between both groups that can be associated with the presence of an “innate” taste for music.

Cross [11] argued that approaches to music that employ evolutionary theory must seek to define ‘music’ as explicitly as possible and specifically music cognition, from an evolutionary perspective. This is a strand of the cognitive and musical literature that has grown in volume and significance only in the last decade. The author suggests that ‘music’ can best be explored in terms of a tripartite model that embraces music as sound (what might conventionally be thought of as constituting music from a Western perspective), as behavior (which embraces the musical—and ‘nonmusical’—acts of musicians, and the activities in which the production of music is embedded) and as concept (how people think about music in terms of its powers and its relations to other domains of human life). Music appears in these papers as sound and (in part) as behavior (little attention is devoted to the range of activities in which music may be embedded), but the concept is missing; there is no consideration here of how is music constructed as functioning by those who are engaged in it or by the societies within which it manifests itself. Consideration of music as a generic behavioral capacity does require that music be characterized as rigorously and fully as possible. Only then does it seem feasible to attempt both to understand how music relates to other aspects of human life and to formulate proposals about the evolutionary roots of human musicality.

Baumgartner’s [12] studies evoked emotions by presenting visual stimuli. Models of the emotion circuits in the brain have for the most part ignored emotions arising from musical stimuli. To our knowledge, this is the first emotion brain study which examined the influence of visual and musical stimuli on brain processing. Results showed that the experienced quality of the presented emotions was most accurate in the combined conditions, intermediate in the picture conditions and lowest in the sound conditions. These findings demonstrate that music can markedly enhance the emotional experience evoked by affective pictures.

Blacking[13] showed the properties of musical intersubjectivity: how music works as a medium of communication between people and how it brings them together. The author saw music as a kind of language that is culturally rooted and socially enacted. It induces and invokes the participation of the whole person, body and soul. The human capacity to send and receive messages through tone, melody, and rhythm is, then, a biological phenomenon as well as a cognitive one. The paper is rich with symbolic connections between music and politics. The analysis of tone and melody give further confirmation of this finding. The rhythms and melodies are indeed influenced by patterns of words and speech-tone, but the music of the songs is more than a mere embellishment of their words.

[...]