Thursday, June 11, 2015

Aesthetic obsolescence VI: Limitations of Youtube data


A good deal of the discussion relating to my recent studies revolved around the reliability and propriety of using Youtube viewership data. The prefatory remarks to my very first essay on the subject had admitted that the data source is far from ideal for systematic research, and that I was using it in full awareness of its limitations.

This is why I submitted the data to only a basic level of mathematical analysis. The measure I used was a simple “average views per month of availability” with respect to the aggregate of views over a period. Throughout the reporting of the results, there was only minimal reference to the mathematical ratio. I preferred, instead, to call it an “audience involvement measure”. I did not report any of the statistical trend analysis I had done on the data for my own understanding. Even the graphic treatment of my computations was historical and generational rather than musician-specific.

I understand that, despite its known limitations, Youtube viewership data is now being used by researchers even in the more advanced research environments. Presumably because it is better than nothing, its use will not stop, And, because of its known limitations, the controversy over its value, too, will not end.

For those who feel interested in this issue, I decided to write out this report on the subject.

ONE: Youtube reports “views” for each recording. The first question is: what is a view? For a 30 minute video, what minimum duration of viewing qualifies as “view”? Further, is a 10 minute viewing the same viewership value as a full 30-minute viewing? We have no way of knowing this and, therefore, of using the notion of “viewership” more intelligently.

TWO: In the early days of Youtube, there was a duration limit on the uploads. This affected Hindustani music very significantly, as most performances exceeded the limit, and had to be split into 2, 3, and sometimes 4 parts for upload. After the duration limit was lifted, mostly complete performances were uploaded. Looking at that data today, can we consider the viewership data for a concert split into 3 parts comparable with that of a complete performance as a measure of viewership involvement? There is no logical way of building this known discrepancy into our measurement of audience involvement.

THREE: Youtube viewership data is cumulative from the date of uploading. By using that data, I am implicitly accepting one view of 2006 on par with one view in 2015.  During this period, Youtube viewership has grown exponentially, and its audience profile for every kind of content has almost certainly changed radically. Intuitively, we know the assumption is flawed. But, we have no way of adjusting for this flaw in our analysis and interpretation of the data.

FOUR: Another problem with cumulative data is that it obliges us, for instance, to equate 20,000 views accumulated over 20 months with 40,000 views accumulated over 40 months. Intuitively, this equation does not look reasonable. One of the two has to be more valuable as a measure of audience involvement. If the propensity of a recording to accumulate viewers is important, 40,000 over 4 years is more valuable. And, if the speed of audience accumulation is considered important, 20,000 over 20 months is more valuable. Youtube data, as available, does not permit us even to ask such a question, leave alone answer it.

FIVE: The Youtube audience is global, and so is the audience for Hindustani music. But, we have no data on the nation-wise mix of viewership for each recording. In our analysis, we are obliged to treat the viewership number as a homogenous mass -- which it certainly is not. By implication, we are assuming that foreign audiences of Hindustani music —across all nationalities and cultures -- have the same relationship with the music, and the same profile, as Indians have. There is enough evidence to show that this is not so. Our analysis and interpretation of the numbers can therefore lead to unjustified and even misleading inferences.

SIX: Although Youtube is a video medium, a large amount of Hindustani music on it constitutes either only audio content supported by a still visual, or audio supported by a slide-show. The nature of the content is itself not uniform. In fact, even the notion of “viewership” may be irrelevant to a lot of the content. Do 100 people listening to an audio recording with just a photograph of the musician on the screen represent the same level of audience involvement as 100 people watching him or another musician in action on film? If not, how do we devise an “equalization factor” between the three formats?

SEVEN: This listing merely covers data source features which impinge directly upon the focus of my studies. Researchers with different perspectives could enumerate a different, and perhaps larger set of limitations.

It appears that, with specific reference to Hindustani music, Youtube neither offers a uniform media experience to its audience, nor publicly provides a rigorous measurement of audience engagement. What, then, is Youtube data good for?

I believe it is better than nothing. It cannot be said to represent “statistics”, but can represent “orders of magnitude”. It can be used only as indicative and never as conclusive. Inferences should be drawn most judiciously from its analysis, with every inference reflecting the analyst’s awareness of data limitations.

A close scrutiny of Youtube user-interface suggests that a lot of very sophisticated analytics are being generated on viewership/ audience engagement by its managers for internal consumption. And, these analytics are being used for strengthening the relationship between Youtube and its users and advertisers. It is safe to assume that the viewership information Youtube currently offers publicly is also aimed at strengthening those relationships. It was not intended to be helpful to researchers and may, in fact, have been purposefully kept unusable for such purposes.

As Youtube grows into a major cultural force, it will find it necessary to understand itself better in order to keep growing.  For this, it will need to engage constructively with social scientists and media researchers in all the geographies and cultures where it has a significant presence. This could launch an era of greater transparency in Youtube analytics.

Until then, the Indian musicologist should be content with “orders of magnitude” and indicative inferences. Is this better than consulting 10 veteran Rasika-s and observers of the music scene? I believe so, because Rasika-s can acquire "personal" preferences (biases, prejudices); impersonally generated numbers cannot. 

© Deepak S. Raja 2015

COMMENT
Tejaswinee Kelkar, Musicologist, IIIT Hyderabad, writes: 

How the data from hundreds and thousands of videos are even displayed to users for search queries is not an uncomplicated task. It is well known that in search algorithms, popular musicians / videos that are viewed more are marked as relevant far more times than other ones. This skews the access to videos that aren't by the few of the most popular artists. It is known that search queries make views and results quite a bit unequal, by algorithmically computing relevance based on things like 'views', which can be very misleading many a times. 

This means that the search and view data that we have is already skew and a little less indicative of actual preferences, as opposed to bottlenecked by search results options that are shown. This is worthy of enquiry because we are talking about a system in which the available choices are more salient for views than ideally 'all possible' choices. 

On google trends, https://www.google.co.in/trends/explore#q=raga they do allow us to peek at search query data on various google products including youtube, and including country of origin of query etc. On this site it is possible to find query results, which i think could be a good way to substantiate what we are finding already from the data you have collected. 

Computational ontologies are quite responsible for the structure of music query results. The IEEE ontology, which is used for music databasing in most software such as iTunes, etc - is completely unrelated to the actual structuring features that we would like in classical music, as it has categories like 'genre', 'artist', 'album', which are less relevant to us than, say 'raga', 'gharana', 'form (khyal / dhrupad)' etc. We are trying to work on this by developing new ontologies that are relevant to the categories we want to search, and trying to build a system which will search open data (like youtube) through ontological annotations.

Friday, May 29, 2015

Aesthetic obsolescence: V


Earlier essays in this series have looked at this subject from several angles. I now attempt to tie up the various strands of thought by including data on contemporary Khayal musicians, using the same measure of audience involvement as in the earlier studies.

Unlike earlier studies, however, I shall not report the audience involvement measures for each musician individually. I avoid this for two reasons: Firstly, such measures are irrelevant to the purpose of this study. Secondly, there is a danger that individual measures for contemporary musicians might be interpreted in a manner unwarranted by the limitations of the data and the nature of the study. 

The contemporary vocalists covered by this study are: MaliniRajurkar, Veena Sahasrabudhe, Kishori Amonkar, Jasraj, Shruti Sadolikar, Ashwini Bhide-Deshpande, Shruti Sadolikar-Katkar, Ulhas Kashalkar, Rajan+Sajan Mishra, Venkatesh Kumar, Ajoy Chakravarty, Gauri Pathare, Arati Anklikar-Tikekar, Manjusha Kulkarni-Patil, Rashid Khan, Kaushiki Chakravarty, Jaiteerth Mevundi, Dhananjay Hegde, Manjiri Asnare-Kelkar, Sanjeev Abhyankar.


With respect to contemporary musicians, a small change had to be made in the selection of sample recordings. For this sub-sample, I have only considered recordings uploaded after January 1, 2010. This is because Youtube viewership data is cumulative from the day of upload of the recordings. This would have made data for the senior and the very young musicians non-comparable. By limiting the duration of the upload dates between January 2010 and May 1, 2015, I have made sure that the data across the contemporary sample remains broadly comparable. This of course assumes that even the youngest of the considered musicians has been on the horizon of significant musicianship for at least five years. This was the only refinement possible to the methodology of a study made using data of known and admittedly limited value. 

The focus

We are looking at the involvement of the present YouTube audience in the Khayal vocalism specimens of musicians born between 1872 and 1980, a period of 108 years. Assuming that the earliest uploaded recordings were made after the musician had reached at least 30 years of age we are looking at music of the period 1902-2015. At the audience end, we are looking at listeners who are between 30 and 75 years of age in 2015.

By the Jose Ortega generational model of 30 years constituting a distinct cultural generation, we are looking at music made by 3.8 generations of musicians, and its hold on the minds of 2.5 generations of audiences. However, Ortega also suggests that the face of the world changes in some ways every 15 years – which is half a generation. We are therefore probably looking at the music of a period within which the world has changed, in some ways, as many as seven times.

GRAPH 1
Graph 1 plots the average monthly Youtube views reported for the Khayal music of 33 Khayal vocalists, born between 1872 and 1980. Significant patterns are as follows:

ONE: The long-term trend-line confirms our view that, the older is the music on offer, the less likely it is to be heard by listeners of Hindustani music. The obsolescence hypothesis is adequately supported.

TWO: The wavelike pattern reflects the simultaneous effect of multiple generations of listeners relating to the same music. In any given population, at any given time, there will be emerging generations, and receding generations of listeners as well as musicians.  The musical values and preferences of the two will diverge partially while also being partially convergent. This partial overlap creates a wavelike formation.  The graph shows seven or eight waves of varying durations over a period of 105 years of music, perhaps suggesting that the musical culture has thrown up significant representative musicians each time the world has changed a little.

THREE:  The suggestion of a long-wave postulated by the Kondratiev Model discussed earlier in this series of essays is also evident in this graph. After Abdul Kareem Khan (born 1872), the graph shows the next big upsurge in culturally relevant musical value with Kishori Amonkar (born 1931) – a period of 59 years.Though it may be a bit early to speculate on the emerging scenario, there are signs that the Amonkar peak may be bettered by vocalists born around 1980 – 50 years after her.

FOUR: If the world changes in some ways every 15 years, it is proper to look at groups of musicians born within the same 15-year period (rather than individual musicians) as collectively expressing the “zeitgeist” of the era. The complex pattern of Graph 1 can be simplified, and made historically more relevant, in Graph 2, derived from the same set of computations as Graph 1.

This graph once again confirms the obsolescence hypothesis, and does so perhaps more sharply than the previous one. In addition, it suggests that the world has perhaps changed significantly six times (rather than seven or eight) in the last 105 years. The peak of this step-graph represents the era of DV Paluskar, Bhimsen Joshi, Kumar Gandharva, Jasraj, and KishoriAmonkar (birth: 1920-1935).

GRAPH 2
Two successive half-generations after them do not appear to claim a comparable share-of-mind amongst audiences. This appears to validate my contention that the first post-independence generation of Hindustani vocalists remained under the stylistic shadows of the pre-independence giants, failed to address their own generation effectively, and thereby caused a substantial loss of audiences to Hindustani vocalism.

It is with the third half-generation after the peak (born 1965-1980) that audience involvement shows a brisk rise with the possibility of matching – and perhaps surpassing – the highest peak evident so far. If a long wave of 50-60 years heralds a paradigm shift, this indication suggests that Hindustani vocalism could be on the threshold of one.

Non-Khayal repertoire

Unlike Dhrupad vocalists, Khayal vocalists have always maintained a non-Khayal repertoire as an integral part of their musicianship. In an earlier essay, we have observed that this non-Khayal repertoire of departed musicians dominates the interest of contemporary Youtube audiences. This is, of course, a commentary on the musical values of contemporary audiences more than the musical tendencies or temperaments of the musicians of the past.

We may now consider whether the pattern holds true when contemporary musicians are considered along with departed musicians. This relationship is reflected in Graph 3, which superimposes audience involvement measure with respect to non-Khayal repertoire upon the measure for the Khayal repertoire. In order to preserve the historical focus of this study, the comparison is, once again, presented by 15-year periods, rather than for individual musicians.

GRAPH 3
The inferences from the emerging pattern are as follows:

ONE: The long-term trend-line for Non-Khayal repertoire exhibits a much sharper obsolescence factor than the trend-line for Khayal repertoire. This confirms our earlier observation that Non-Khayal repertoire responds more efficiently to changing aesthetic values and musical needs of society because of the freedom it enjoys from the relatively more durable musical values characteristic of the Khayal genre.

TWO: It is also significant that the upswings in the viewership of Khayal recordings have consistently been accompanied by much sharper upswings in the viewership of non-Khayal recordings. Likewise, a stagnation or depression in Khayal viewership is accompanied by a stagnation or depression in the viewership of Non-Khayal recordings. This would imply that the appeal of a musician’s non-Khayal repertoire is more crucial to his/ her professional success than that of his/her Khayal repertoire.

THREE: The long-term trend line of the Non-Khayal repertoire runs at double the audience-involvement level of the Khayal repertoire trend line.  This pattern confirms our view that contemporary audiences are far more involved with Non-Khayal repertoire of any era within their frame of consideration than with the Khayal repertoire.  This may be interpreted to imply that The Khayal genre, as hitherto understood, is fast ceasing to satisfy the musical needs of a vast majority of contemporary audiences.

FOUR: Genres may become obsolete; but musicians do not – because they have to survive. The Khayal genre must therefore now take a back-seat in the total musical persona of a classical vocalist, and/or undergo a major re-calibration of its aesthetic assumptions in order to remain relevant to contemporary society.

FIVE: From the contemporary perspective, the last paradigm shift in Hindustani vocalism is traced to the 15-year period that felt the impact of DV Paluskar, Bhimsen Joshi, Kumar Gandharva, Jasraj and Kishori Amonkar. For that period (See Graph 3), the audience-involvement for Non-Khayal music runs at about 900 viewers per month above that for Khayal music. For the first time since then, the latest period under review shows a gap of over 600 views per month between the two graphs.

When such a large gap emerges between durable musical values and ephemeral musical values, the community of musicians can be expected to narrow it in the interest of their own economic security. How does this happen? For an answer to this question, we draw on Prof. Ranade’s observations on how a mainstream genre builds, maintains and protects its supremacy. Amongst other processes, it does so by (a) adopting the features of the ascendant or rival genres and (b) by attracting talent from all kinds of sources.

Firstly, the proportion of Non-Khayal repertoire to a musician’s total public presence will shoot up. Instead of being the “dessert” at the end of a meal, it will progressively become the “Main Course”. Secondly, the stylistic distinction between Khayal and Non-Khayal repertoires will narrow. Khayal renditions will have to start reflecting the musical values of Non-Khayal genres, if they wish to command an audience at all.

The same process may also be viewed a little differently. The predominance of Non-khayal repertoire as a success factor for professional musicians will tend to attract vocalists who have the training and temperament more suited to Non-khayal music than to Khayal music. Their presence in the Khayal segment – such as they may be able to achieve – will give them the respectability of the “Classical Music” platform, without putting their competence to test. Under such conditions, their Khayals can only remain unconvincing by the traditional yardstick of Khayal vocalism. Both these processes would be logical manifestations of a paradigm shift -- either imminent or already under way.

What is Khayal or Non-Khayal?

The present author has surveyed the Youtube portfolios of 33 departed and contemporary Hindustani vocalists for this study, recorded over a century. With particular reference to contemporary musicians, he has had to resolve interesting conceptual issues while classifying individual recordings as belonging either to the “Khayal” or “Non-Khayal” categories.

The researcher in Hindustani music today encounters a bewildering variety of “genres”, many of which neither have appropriate names, nor a well-defined character. An attempt is made here to list some of the names that have suggested themselves.

Khayalised Thumree, Thumriised Khayal, Khayalised Sargam, Sargamised Khayal, Bhajanised Thumri, Bhajanised Khayal, Thumri-ised Bhajan, Khayalised Bhajan, Kriti-isedKhayal, KhayalisedKriti,  Bhajanised scriptures, Scripturised bhajans, Scripturised Khayal, Khayalised scriptures, Vocal-instrumental jugalbandi, Hindustani-Carnatic jugalbandi, Hindustani-Carnatic fusion, Hindustani-Carnatic-Western Pop/ Jazz  fusion, Hindustani-Opera fusion.

All these “genres” respond to a notion of Raga-ness, even if only tangentially or remotely. But, the relationship is incidental more than purposive. Even where the Khayal is encountered in its original architecture, the manifestations of Raga-ness in the rendition rarely cross a set of identifying phrases.

This suggests that Hindustani music audiences are now so hungry for novelty, that almost any well-trained vocalist can create a niche for himself/ herself by creating a new "genre", as long as he/ she can execute it with competence and confidence. And, if his/her non-Khayal repertoire clicks with the audiences, he/ she need not fear being astutely evaluated by orthodox audiences for his/ her Khayal competence.  

A highly fragmented aesthetic environment does not, of course, push the orthodox Khayal featuring a deliberate Raga exploration entirely out of circulation, or place it under an immediate threat of extinction. As we have observed earlier, today’s audience consists of several generations of listeners and several generations of musicians. The orthodox Khayal can still remain (like Dhrupad) in marginalized circulation for another couple of generations; but perhaps not longer. And, even during its residual life, it will change beyond recognition. This is because the very essence of a paradigm shift is that it changes society’s perception of itself, never to return to its original state.

Cultural projections are difficult at this stage because, tangentially, the age-profile of the Indian population has to be a determining factor. Long term trends suggest that the youngest age group is growing fast as a proportion of the total population, while longevity is also improving steadily. This will soon leave the nation with the very young and the very old constituting the vast majority, with a very lean middle-age group.

According to an authoritative study reported on April 17, 2013 by The Hindu, every third person in an Indian city is a youth. In about seven years the median age in India will be 29 years, very likely a city-dweller, making India the youngest country in the world.

GRAPH 4 
It is impossible to foresee how these schizoid realities will affect the future of the Khayal. This is relevant because the Khayal (along with Dhrupad) represents the more durable aesthetic values in the musical culture valued presumably more by the receding generations, while the non-Khayal (and non-Dhrupad) genres represent the more ephemeral values, supposedly valued more by the emerging generations.

The totality of the contemporary Hindustani vocal experience suggests to me the features of Post-modernism.

Post-modernism

As a cultural movement, postmodernism has its epicenter in Europe. Its original triggers obviously have nothing in common with the Indian socio-economic reality. It is therefore difficult to define postmodernism in the Indian context. It is, however, possible to relate some of the well-documented features of postmodernism to the manifestations of their worldview in the contemporary Hindustani vocalism.

Charles Jencks, a reputed interpreter of postmodernism, describes a postmodern building as – “part modern, and part something else: vernacular, revivalist, local, commercial, metaphorical, or contextual”.[Jencks, Charles: What is Postmodernism, quoted in Key Ideas in Human Thought. Ed: Kenneth McLeish, 1993. Facts on File Inc., New York, Pg. 584-585]. McLeish observes that “There are no boundaries [in postmodernist art] save our individual competence: creator and spectator are locked in a conspiracy against history, against geography, and against specificity, which may be seen as liberating or destructive (the lunatics taking over the asylum!}, but which is entirely without precedent in the story of the arts”.

Foster points out with reference to postmodern films [Foster, Hal. Postmodernism: A Preface, Ed. Post Modern Culture, Pluto, London, 1985], that postmodernism is “a combination of lots of different genres, signs, and cultural elements. The features of postmodern films are hedonism, and decadence – things that are often seen towards the end of a culture’s life. … A feature of postmodernism is its ability to shock, without linking this device to a message”. In a similar vein, Jean Luc Goddard described his films as having “a beginning, a middle and an end; but not in that order.”

The postmodernist worldview rejects the existence of anything called “human nature” [Chagani, Fayaz. www.geocities.com/Athens/Agora/ 9095/ postmodernism. html] . It argues that there is nothing universal  – whether in a global context or in a culture-specific context, or even across time  – about the manner in which people convert sensory stimuli into meaning. This "weltanschauung" does not therefore accept any established assumptions regarding the proper, or even the most efficient, means of communicating in any language – whether verbal or otherwise.  In the postmodernist world, there is no fundamental purpose for something to be what it is, except that it is what it is.

Note: The computations supporting the above graphs can be provided on request to scholars and researchers at the author’s discretion.

© Deepak S. Raja 2015

Friday, May 15, 2015

Aesthetic obsolescence IV



In the earlier essay, we found that the non-Khayal repertoire of Khayal vocalists achieves a much higher level of audience involvement than their Khayal repertoire.  The musician lives with this reality every day and responds according to his temperament and inclinations. He does not need to figure out the extent to which this is attributable to his Khayal-related status, to the independent appeal of his non-Khayal repertoire, or to the uniqueness of the combination. But, the researcher does need to resolve this issue.  

If we are to judge the strength of the non-Khayal segment of the musical culture as a possible challenge/ distraction to the fundamental character of the Khayal as we know it, we need to isolate the non-khayal segment for measurement. Circumstantial evidence of this can be obtained from the analysis of audience involvement in the music of the specialist performers of the non-Khayal genres. In this analysis, the absolute numbers delivered are as important as the trend of obsolescence. This essay attempts to look at these patterns.

Research design

The methodology followed is identical to the earlier studies conducted on samples of Khayal vocalists.

The sample of available Youtube recordings was drawn from the leading specialist performers of non-Khayal genres in the segments of the Thumri and allied genres, Ghazals, and Bhajans. For want of the author’s knowledge, specialist performers of the regional devotional genres were not included.  The musicians were chosen for the significance of their presence on the concert platform over a long enough period.

In the Thumree segment, two of them – Girija Devi and Chhannulal Mishra – are still alive, and are included.  In the Ghazal segment, the sample selection was restricted to the singers of the classical ghazal – Begum Akhtar and Mehdi Hassan, while the more recent singers of the modern ghazal have been excluded.

This approach has given us a sample of 11 musicians in whose case, a minimum of 15 and a maximum of 35 recordings were available for the measurement of audience involvement.

Indications


The graphic plot shows a long-term growth trend of audience involvement. But, it also exhibits two waves within the period considered. The two peaks are represented by Begum Akhtar and Mehdi Hassan. Two waves, within a long-term growth trend is a pattern very similar to the pattern noticed in the earlier study of the non-Khayal repertoire of Khayal vocalists. The generational interpretation of this graph could be similar to that of the graph seen in the earlier study. 
 It would appear that at this particular juncture in history, two generations of audiences are involved with the music of the specialist performers of the non-Khayal genres. However, the two titans who dominate the two visible peaks – Begum Akhtar and Mehdi Hassan are born at an interval of only 13 years. This might mean that, in effect, only one generation of musicians appears to dominate the non-Khayal music-scape at the moment. This compares with three, perhaps four, generations of audiences involved with the Khayal music of the past generations of vocalists.

While we may consider these two giants as representing the same generation and appealing to the same generation, a certain obsolescence factor is evident even in their relative scores.  It is probably the younger end of the audience generation which gives Mehdi Hassan a score almost four times the score obtained by Begum Akhtar -- ostensibly emanating from the older end of the same generation. 

The power of the specialist non-Khayal indicator as an influence on Khayal music is more sharply reflected in its absolute numbers than in the obsolescence trend. The bar-graph tells the story.

The Khayal music of Khayal vocalists begets only 203 viewers per month of recording availability. Compared to this, their own non-Khayal music notches up 914 views, and the non-Khayal music of the specialist performers of non-Khayal music crosses that 1000 views per month. 

Admittedly, this average for specialist performers of the non-Khayal genres is greatly bolstered by Begum Akhtar and Mehdi Hassan, and not every musician achieves their level of audience involvement. However, the average of over 1000 views per month cannot be ignored as a perceptional reality of the musical culture and of relevance to professional musicians.

By pure arithmetical coincidence, if we add the audience involvement ratings for Khayal and non-Khayal repertoires of Khayal vocalists, we get a figure almost exactly equal to the audience involvement rating of the specialist non-Khayal performers. We may then surmise that the primary motivation for Khayal vocalists to perform non-Khayal music is reaching out to a much larger audience.

The interaction of genres

With audiences exhibiting a marked preference for non-Khayal repertoire, the Khayal will tend to shrink in terms of its durational presence on the concert platform. As this trend gathers momentum, Khayal vocalists will begin to lose their aesthetic grip over the Khayal as a genre. Their approach to the Khayal will increasingly reflect the aesthetic values of the non-Khayal repertoire in vogue.

This process may be better understood with a historical analogy from the print medium in the country, which is familiar to most of us.

In the 1950s and 1960s, the mainline dailies published a sports page only twice a week.  In those days, several sports magazines (weeklies/ fortnightlies/ monthlies) showed a healthy growth in sales and revenue. The mainline dailies saw this opportunity for an enlargement of their share-of-mind and share-of-market, and stepped up sports coverage to every day. By the 1980’s all the specialist sports magazines had died for want of readers and advertisers.

The same thing happened to the film-magazine boom of the 1960s and 1970s. By the year 2000, mainline dailies which earlier had only weekly/ bi-weekly cinema supplements, began cinema coverage on a daily basis. Almost all film magazines have since been gasping for breath. This history repeated itself with the arrival of specialist society/celebrity magazines in the 1980s.  By 2010, society and celebrity coverage has become regular Page 3 material in the mainline dailies. The specialist magazines in this segment of journalism are now virtually extinct.

As a result of these changes, India's mainline English dailies have changed almost beyond recognition in the last 30 years.

In the present context, there is no reason to believe that a transformation of the Khayal will drive the Ghazal or Thumree totally out of circulation. But, as the Khayal starts resembling the relatively Raga-neutral genres in some respects, it will shrink the aesthetic space available to them for remaining in independent circulation. And, some would argue that this tendency is visible already.

It has been argued, for instance, that Begum Akhtar and Mehdi Hassan brought so much melodic sophistication to the Ghazal, that they left the original Banaras Thumree struggling for survival. It has also been noted that once the Khayal genre annexed and enriched the Bandish-ki-Thumree as the Chhota Khayal, the Bandish-ki-Thumree lost its footing in the musical culture.  Further, it is observed that with the advent of the romanticist brigade in Khayal vocalism – Kumar Gandharva, Jasraj and Kishori Amonkar – the Bol-banav Thumree and some other semi-classical genres lost a good deal of their aesthetic  territory.  These are purely aesthetic observations, which are impossible to either prove or disprove. But, they are not  rendered invalid by this alleged infirmity.

Here we go back to Prof. Ranade’s argument stated in the earlier essay. The dominant mainstream genre tends to protect its dominance – amongst other tendencies – by adopting the musical features of ascendant rivals.  By this reasoning, Khayal, the dominant mainstream genre, is likely to respond to the threat/ opportunity presented by the non-Khayal genres by adopting their musical values. This possibility does suggest a paradigm shift because the Khayal is founded on the durable musical value of Raga exploration, while the aesthetic assumptions of the non-Khayal genres are fundamentally Raga-neutral, even though Raga-s may in practice, inform their melodic content.

Is this happening already?

The Youtube data studied so far pertains to vocalists who are either already departed, or in their advanced years. We need to examine several additional pieces of contemporary and/or  trend-related evidence to support the hypothesis emerging here. These may constitute the objects of continued enquiry.

© Deepak S. Raja 2015

Monday, May 11, 2015

Aesthetic obsolescence III


In the earlier essay, an attempt was made to ascertain the patterns of aesthetic obsolescence in Khayal vocalism across the last century over the period 2010-2015. Appropriately for the focus of the earlier study, the non-Khayal repertoire of the considered vocalists was left out of the measurement. However, as a parallel reality, and a component of the share-of-mind a musician enjoys with audiences of his own and future generations, it cannot be ignored.

Unlike practitioners of the medieval Dhrupad-Dhamar genre, Khayal vocalists have, for long, retained an involvement with the semi-classical genres. This facet of Hindustani vocalism acquires greater importance because of the major changes that have taken place in the music-scape, particularly after independence. The specialist singers of the semi-classical Thumree and its allied genres have virtually disappeared from the scene. Simultaneously, vocalists trained in the Khayal genre have adopted these genres in performance, and also added the Bhajan to their musical persona. Several leading Khayal vocalists of the 20th century are acknowledged to have contributed significantly to the evolution of these two genres.

It is neither possible, nor necessary, for us to ascertain whether their non-Khayal repertoire has contributed to their share-of-mind as Hindustani vocalists, or the other way round.  There exists a synchronous relationship between them, which needs to be examined from the point of aesthetic obsolescence. This essay attempts to explore this relationship.

Research design

The research design is identical to the design of the study of aesthetic obsolescence of Khayal repertoire. The sample consists of the same vocalists, truncated by the non-availability of data for some of the vocalists.  Of the 21 vocalists considered in the earlier study, 8 had to be dropped on account of inadequate or missing data. Expectedly, the number of Non-Khayal recordings available for logging was consistently much smaller than Khayal recordings. The minimum readings considered acceptable for considering any vocalist were 3, and the maximum readings available for any musician was 34. Average views per month of exposure were computed precisely in the same manner for comparability of results.

The repertoire considered in this study covers Thumree, Tappa, Dadra etc., Ghazal, Natya Sangeet, Bhajans, patriotic songs, regional devotional music, and songs performed in Hindi or regional films.

Indications

The first table and graph plots average views per month of recording availability against the year of birth of the musician. The picture differs substantially from the pattern seen in the earlier study of Khayal recordings. While the Khayal viewership graph showed a long-term linear trend accompanied by a series of wave patterns, this graph is predominantly an exponential graph, with the hint of a wave within it. This means that the more recent the musician, the more disproportionately he gains viewership amongst present-day audiences for his non-khayal repertoire. Worded differently, non-khayal repertoire music tends to become obsolete faster, and with greater certainty, than Khayal music.

The two-wave pattern evident in this graph also has a defensible generational interpretation. It would suggest that, currently, the non-Khayal musical values of two distinct generations of musicians engage the attention of Youtube listeners. The first is represented by the musical values of Bade Gulam Ali Khan and the second predominantly by Bhimsen Joshi, but also Kumar Gandharva and DV Paluskar. It is interesting that the same set of musicians showed a three-wave pattern with four peaks for Khayal music. This further establishes the difference between Khayal music and the non-Khayal repertoire of the same musicians. Their Khayal repertoire has an audience of three or four generations, while their non-Khayal repertoire repertoire commands the attention of only two. 

These patterns support what is known about the semi-classical and allied genres of vocal music. Relative to Khayal vocalism, they are designed to appeal to a much larger audience, which is more likely to respond to ephemeral musical values the the more durable musical values. 

There are no significant cycles apparent in this graph, and the possibility of a paradigm shift does not appear to arise from available data.

The second graph below provide further insights.

On an average, vocalists considered here have logged almost double the viewership for Non-Khayal recordings compared to his/ her Khayal repertoire. Even within the overall pattern of faster obsolescence of non-Khayal recordings, the absolute level of durability of appeal is substantially higher for non-Khayal music than Khayal music. But, the pattern deserves a closer look.

DV Paluskar’s non-Khayal repertoire logs twice the rating of his Khayal repertoire. After him, the multiplier shows a smart rise. Bhimsen Joshi’s non-Khayal repertoire is almost 4 times more popular than his Khayal repertoire. And, finally, with Kumar Gandharva, the non-Khayal repertoire almost touches 8 times the Khayal level in terms of audience involvement.  The sharpness of this trend would suggest a backdrop of fairly radical changes in the social and economic environment of which musicians and their audiences are concurrent products.

It appears that the non-Khayal repertoire, which has a faster rate of obsolescence, is claiming a progressively higher share-of-mind amongst audiences. This has several implications for the musical culture, all of which are already on the horizon.

Implications of the patterns

The inflow of fresh talent into classical vocalism could shrink, with more and more trained singers opting for a career in the non-Khayal genres. Because of this, the non-Khayal genres could witness the emergence of high quality specialist performers. Khayal vocalists  -- once they have established their credibility as trained classical singers -- will be drawn towards increasing their involvement with non-Khayal repertoire to protect their share-of-mind and share-of-market.  

The most significant implication of the patterns evident here is that the Khayal appears to be a receding genre in terms of the society’s share-of-mind, and the non-Khayal genres of vocal music appear to be the ascendant genres. This indication may require further evidence to be stated with greater certainty. But, it is not insignificant even with the evidence available here.

The eminent musicologist, Prof. Ashok Ranade made some interesting observations about the behavior of the dominant genre in times of challenged dominance.

“.... we should look at how any musical form achieves and sustains its dominant position. First, it attracts all kinds of performers towards it. Second, it tries to assimilate the musical tendencies of other forms. Third, it allows individuals enough freedom to express themselves, especially in the initial stages. Fourth, it makes allowances for a distinction between the larger disciplinary model of a gharana, and the style of an individual musician.” (Journal of the Indian Musicological Society, 1999)

This summarises the history of Khayal vocalism fairly accurately. For the present, we may focus  sharply at the prospect of the Khayal as the threatened genre adopting musical values of its challengers as a protection against its own extinction. 

The Khayal is three-dimensional genre – it is a contemplative art, an expressive art, and a communicative art. Its contemplative facet involves an improvised exploration of the Raga’s melodic and emotional personality. It is this facet that distinguishes it from the non-Khayal genres of vocalism. If the Khayal is tending towards imitating the non-Khayal genres, what it is abandoning is either the Raga discipline or the deliberate exploration of the Raga’s melodic-emotional personality, or both. It is in this context that the issue of an impending paradigm shift may be relevant.

At this point, we may recall what a paradigm shift means in the present context.  When a musical value, considered fundamental for long, is abandoned by a majority of  performing musicians -- without necessarily identifying an emergent alternative -- a paradigm shift can be said to have taken place.

The contemplative facet of Khayal vocalism, and the centrality of Raga elaboration to it, is a musical value considered fundamental for at least more than a century.  It faces the prospect of being abandoned in favour of the more ephemeral musical values of the non-Khayal genres.  The architecture of Raga rendition  characteristic of the Khayal may remain in circulation even with a shrinking presence. But, its distinguishing feature, as has been understood so far, could be headed for extinction.

This study began with looking at aesthetic obsolescence of the music of eminent Khayal vocalists, and incidentally of the major Khayal gharanas. Now, having considered the divergent patterns in the non-Khayal music of the major Hindustani vocalists, it has raised issues relating to the aesthetic obsolescence of the Khayal genre itself. There is insufficient evidence yet to permit a categorical view on this possibility. It may therefore remain, for now, a hypothesis based on astute observation, and theoretical speculation.

© Deepak S. Raja 2015


Thursday, May 7, 2015

Aesthetic obsolescence in Hindustani music II


In an earlier paper, I have suggested (1) that aesthetic obsolescence is a reality in classical music, and (2) that a paradigm shift in aesthetic values can be expected at intervals of 50-60 years.  Those propositions were based on anthropological, and macro-economic constructs evolved in the American/ European/ global context. Despite their intuitive appeal, they are not sufficient to support a definitive view with respect to Hindustani music.  Indian evidence with respect to such phenomena needs to be considered.

Aesthetic obsolescence and paradigm shifts in Hindustani music I

The Hindustani music audience accesses and consumes music through an increasing variety of sources, and their mix itself is changing constantly. Many of these sources are either informal, or generate quantitative information on consumption patterns, which is available only to its providers. The Hindustani music market is just not large enough to spawn independent sources of reliable quantitative information. The glimmer of hope in this regard comes from YouTube, which now features an substantial and increasing volume of Hindustani music, and is emerging as a valuable source of archival music for music lovers.

YouTube as data source

Every video recording featured on YouTube now provides information on the date of uploading, and the number of views against each recording. By dividing the total viewership of a recording by the number of months/ weeks since the date of uploading, it is possible to get a measure (views per month/week) which enables a comparison of individual recordings and musicians on the level of audience of involvement and interest logged up to a given date. This simple measure can be submitted to statistical analysis for extracting valuable insights.

YouTube is, admittedly, not a perfect solution to our problem. Even amongst internet-based music repositories, it is one of the many, though perhaps the largest and the most popular. Net-based repositories are themselves one of the many media through which Hindustani music is being accessed. It is impossible to account for the various determinants of availability of any recording on YouTube or alternative internet music portals/ archives.  It is also impossible – without access to YouTube’s internal analytics – to appreciate how YouTube defines viewership, thus making a comparison between recordings and musicians an uncertain exercise.  

Another major issue concerns YouTube itself. The portal itself is growing constantly in terms of content diversity, content volume and usership base. Inevitably then, its global audience profile is also changing constantly. So, it becomes conceptually tricky to compare, say, 50 views per month for a recording Ustad Fayyaz in 2010 with 50 views per month of the same recording in 2014. Do the two represent the same level of public interest and involvement? 

The methodological issues raised by this data source are innumerable.  With all its limitations, however, YouTube is the most substantial repository of Hindustani music, which also reports quantitative information on the level of audience interest and involvement in archival music starting from the early years of the 20th century. It therefore makes sense to examine YouTube data for possible indications of aesthetic obsolescence and paradigm shifts with respect to Hindustani music.

Research design

This study is based on recordings of deceased Khayal vocalists because, in Hindustani music, only the Khayal genre enjoys considerable structural stability across the century we are looking at, and features a continuous flow of significant musicians to monitor. Admittedly, several Khayal vocalists have achieved additional popularity and stature because of their non-Khayal repertoire such as Bhajans, Thumrees, and Tappas. There is clearly no way of isolating the effect of this facet of their musical personality on their mainstream persona. To keep all musicians and recordings comparable, the study chose to ignore the non-Khayal recordings of the considered musicians available on YouTube.

The selection of significant musicians to monitor was based on the author’s knowledge of Khayal vocalism, sometimes constrained by the availability of sample data. An attempt was made to ensure that significant vocalists from every major Gharana of Khayal vocalism were included in the study. Even an otherwise significant musician was not considered if his/ her YouTube presence did not cross 5 different ragas. This gave a total listing of 21 vocalists, whose YouTube presence was even indicatively measurable.

For every significant musician considered, at least one recording of every raga available on YouTube was included. Wherever more than one recording of the same raga, or even the same performance was available, all were considered. Because of the varying pattern of availability, it could not be helped that, the study considers as many as 58 recordings of one musician at the upper end, and as few as 6 of another at the lower end.

In a vast majority of the cases, the considered recordings had been uploaded between 2010 and 2015. In very few cases, upload dates go back into 2008 or 2009. The number of months for which each recording had been available was computed using May 1, 2015 as the cut-off date. The data was recorded between May 2 and May 4.

The table below shows the computations of audience involvement/ interest/ viewership of the considered musicians. The graph below presents average viewers per month of the musicians plotted against their year of birth. Admittedly, the year of birth is not the ideal landmark for comparability because different musicians have acquired stature and influence at different stages in their lives. No stage-of-life-cycle alternative would have been satisfactory because that would have left out a musician like DV Paluskar, who achieved considerable stature in his short life of 34 years.

Indications

The graphic representation of the analysis should be considered first for its conceptual and theoretical implications.

The graph supports the proposition that aesthetic obsolescence is a reality in Hindustani music. It exhibits a clear trend of viewership favoring musicians born more recently as against musicians born in earlier years. A long-term trend-line would suggests that, on an average, a musician born in 1930 would have 5 more YouTube viewers per month than a musician born 60 years ago, in 1870.


In addition to a linear trend, the graph also exhibits a cyclical trend. Every once in a while there appears to emerge a breakthrough vocalist whose music achieves a higher level of audience involvement than expected. And, this booster evidently provides the momentum for the subsequent breakthrough to seek an even higher intensity of audience involvement. Here, of course, we do not know how much of this booster effect is attributable to the music, and how much to the dynamism of the media environment – most notably YouTube itself.  But, then, media grow on the strength of their content. Therefore, which is the cause and which the effect, is an intractable issue.

This three-wave pattern has a defensible generational interpretation. From this, it would appear that currently, there are a three -- probably four -- sets of musical values which command the involvement of Youtube audiences. The first wave, represented by Abdul Kareem Khan, the second by Bade Ghulam Ali Khan, the third by Mallikarjun Mansur, and the fourth by Bhimsen Joshi, Kumar Gandharva, and DV Paluskar. This would suggest that three -- perhaps four -- distinct generations of listeners are currently involved in the recorded Khayal music of the 20th century.

In the history we have plotted, Abdul Kareem Khan (born: 1872) appears as the first breakthrough Khayal vocalist. After him, Bade Gulam Ali Khan (born: 1903), represents the next peak.  Ameer Khan (born: 1912) defines the following peak. The last peak covered by this study is defined by Bhimsen Joshi (born 1922). It will be noticed that the time-gaps between the successive peaks are not even approximately similar. This is not surprising because art is not obliged to provide the econometrician with convenient solutions to his problems.

The peaks are, undoubtedly, defined by those we consider “landmark” musicians. But, do we have any indications here of a paradigm shift or a long cycle of 50-60 years? Possible, we do. On the graph presented here, we are looking at three smaller waves with the chronological distance between the first and the fourth peak being exactly 50 years. We could well be looking at an Indian version of a large Kodratiev wave of 50-60 years, which subsumes three shorter Kuznets waves, as described in my earlier paper on this subject. If this speculation has any merit, Hindustani music could today be on the threshold of a paradigm shift – this being the hypothesis I have articulated during my speech in Bangalore in December, 2014, which triggered off this inquiry.

Of course, we need to be circumspect about such speculation because this study does not permit a view on this aspect of the obsolescence theory. This is so mainly because the data source itself is constrained by the history of electronic amplification and recording technologies. It is known that developments in recording and storage-media technologies made a significant contribution to making Abdul Kareem Khan, Bade Gulam Ali Khan, Ameer Khan and Bhimsen Joshi “landmark” musicians. By the same logic, the recordings of “landmark” musicians from the early days of sound engineering – such as Kesarbai Kerkar and Fayyaz Khan – could have suffered truncated aesthetic lives on account of a comparatively primitive acoustic environment.   

With the recording industry itself being less than 120 years old, it would be impossible to observe even two full long-cycles of 50-60 years, even if they existed, because reliable quantitative data to support such a theory would be impossible to compile. Therefore, the notions of paradigm shifts or long cycles will perhaps remain in the region of scholarly opinion and public debate.

Gharana perspectives

While the individual viewership ratings of the various musicians are of wider popular interest, the Gharana affiliations and stylistic legacies reflected in this study are of greater importance in understanding the cultural process. The groupings, as attempted, reflect my understanding of stylistic tendencies.

Those who described the demise of Bhimsen Joshi as the “End of an Era” were probably more prophetic than they realized. If there is a 50-year cycle culminating with the musicianship of Bhimsen Joshi, the event might have been acceptably described as the “End of the Kairana Era”,which began with Abdul Kareem Khan born exactly 50 years before Bhimsen Joshi, and spawned a veritable galaxy in between. 

It is easy to see the Kairana group heading the list with an average of 313 viewers per month. It does so, on the strength of Bhimsen Joshi (652), Ameer Khan (486) and Abdul Kareem Khan (410). Patiala stands tall alongside (301) with just one vocalist, Bade Gulam Ali Khan holding the fort.

The Gwalior group (156) stands almost on par with the Jaipur-Atrauli group (125). But, they present a picture in contrast. With the exception of DV Paluskar, the Gwalior group claims its share-of-mind entirely on the strength of highly original, individualistic, or reformist musicians – starting from Ramakrishna Vaze and ending with Yashwant Buwa Joshi. In sharp contrast, the Jaipur-Atrauli group consists entirely of orthodox vocalists, trained by the founding family.

This confirms Prof. Bonnie Wade’s view (Khayala: Creativity within India’s Classical Music Traditions, Cambridge University Press, 1984) that in the second quarter of the 20th century, the Gwalior gharana suffered a loss of identity, and was obliged to reinvent itself. This crisis resulted in several Gwalior trained vocalists drifting towards the dominant style such as Agra (e.g. Yashwant Buwa Joshi),  the ascendant style such as Jaipur-Atrauli (Mallikarjun Mansoor), or to emerge as highly individualistic originals (e.g. Omkarnath Thakur, Kumar Gandharva).

This contrasts apparently with Jaipur-Atrauli, which holds its share-of-mind alongside Gwalior so far on the strength of its orthodox musicianship. It is not, however, insignificant, that the Gharana group rating is being held up substantially by Mallikarjun Mansur, who performed orthodox Jaipur-Atrauli music, but was in fact a mature migrant from Gwalior. It is also debatable whether Mansur's music ever lost traces of Gwalior vocalism. Also, what is not reflected in this study is the immense influence of Kishori Amonkar of the same lineage, starting from the mid-1970s, which has left the orthodox stream of Jaipur-Atrauli gasping for breath.

Therefore, in the Gwalior group, as well as the Jaipur-Atrauli group, we are looking at a reformist phase struggling against the forces of aesthetic obsolescence.

The Agra group (45), headed by Fayyaz Khan (110) is pulled down to the fifth position by his successors.

Individual highlights

In the Kirana group, Roshanara Begum (104) is a surprise with a viewership rating on par with Kesarbai Kerkar (99) and Fayyaz Khan (110).  This is all the more interesting since she migrated to Pakistan at the time of independence, and visited India only rarely thereafter. She is not known to have performed in India after 1947. Her recordings have been her sole contact with Indian audiences. Merely by belonging to the next generation, and by virtue of her musicianship, she stands today on par with titans of just a generation before her.


The Gwalior group (156) is held up in the third position by DV Paluskar (243) who died in 1955 at the age of 34. His present-day rating is almost twice that of Omkarnath Thakur (127) who acquired immense stature and popularity a generation before him. In addition to the perennial youthfulness of his musical legacy, Paluskar enjoyed the advantages of a buoyant recording industry and media environment just a quarter of a century after Omkarnath Thakur.

The Jaipur-Atrauli group (125) is currently placed in the fourth position by Mallikarjun Mansur (208) much more than by Kesarbai Kerkar (99), the empress of the concert platform a generation before him. In this, we once again observe the effect of a generational shift in musical values, supported by the dynamism of the electronic media.

Conclusion

Despite the limitations of the data source, this study supports the notion of aesthetic obsolescence as a reality in Hindustani music. This constitutes valuable confirmation of reports to the same effect received from significant players in the commercial recording industry as well as the barter market for archival recordings.

This study does not provide clear support to the supposition of a paradigm shift in musical values which, I have observed, is currently taking place. However, considering that the study shows wave patterns resembling three short components of the Kondratiev wave of 50 years, the possibility of an imminent paradigm shift cannot be ruled out either. The argument in favour of such a possibility is that if musical values are changing constantly, the change cannot remain imperceptible indefinitely. At some stage, the accumulation of imperceptible changes will become perceptible in the form of a paradigm shift. Whether such a watershed can be expected every 50-60 years, or at shorter or longer intervals-- or even at irregular intervals -- in the context of Hindustani music is, as yet, inestimable. 

At this point, one may devote a thought to how the accumulation of imperceptible changes over a period becomes perceptible as a watershed or a paradigm shift. When a musical, considered fundamental to the acceptability of sound music, is abandoned by a majority of performing musicians -- without necessarily identifying an emergent alternative -- a paradigm shift can be said to have taken place. If such a shift is impending at this juncture, I expect much will be written on this subject in the next few years by astute observers of the Hindustani music culture. 

There is, of course, no reason why the change and periodicity patterns evident in Hindustani music should fit neatly into anthropological or econometric models developed in the developed world. We need, therefore, to be cautious in drawing upon such theoretical constructs for interpreting the Indian reality, with all its probable uniqueness.

However, with sufficient indications favoring the essence of our argument, there could be justification in pursuing an examination of social and economic forces that might help refine our understanding of the musical culture.

© Deepak S. Raja, May 2015