Big data is big news. In Spring
2015 two reports about popular music gained worldwide attention. The first came
on 22 April, when Ajay Kalia posted his blog entry, ‘Music Was Better Back Then’. It was
followed on 6 May by a report in the journal Royal Society Open
Science. In their article, ‘The Evolution of Popular Music:
USA1960-2010’, a team of academics from Queen Mary and Imperial College
London used data to investigate ‘the evolution of popular taste’ and determine
periods in which there had been ‘rapid change’. I have written about Kalia’s
research in a previous blog entry. This second
article has much in common with it. The public was drawn to the reports for
similar reasons: their eye-catching and over confident use of data; the way
this data can be contested; the focus upon ‘new music’.
The
academic researchers believe that their data forms ‘the basis for the
scientific study of musical change’. They have analysed 17,000 American chart
hits from 1960 to 2010, classifying their ‘harmonic and timbral qualities’. The
resulting data has then been employed to chart the rise and fall of these
qualities through the time span of their study. Their conclusion is that
‘musical evolution is punctuated by revolutions’. There are three years in
particular in which they posit rapid change: 1964, which saw musical
developments in rock and soul, 1983, which had advances in new wave disco and
hard rock, and 1991, which witnessed the break-through of hip-hop.
Although
the researchers believe that ‘Those who wish to make claims about how and when
popular music changed can no longer appeal to anecdote, connoisseurship and
theory unadorned by data’, there are several grounds upon which their
quantitative analysis of musical qualities can be challenged. The first is that
it does not include enough musical data. Where, for example are lyrics within
their analysis? Moreover, where is the music that lies outside of the Billboard charts? Many would suggest that
musical revolutions first occur within the underground. Secondly, although the
researchers have taken genre into account, they have not made any allowance for
different rates of progression. While change is the hallmark of some genres,
others are relatively static. In the former case, wild diversions can be the
mark of stability rather than change; in the latter, mild alterations to the
form can be of great significance. Thirdly,
their research can be criticised for not including enough non-musical data. Genres are not about
music alone, but also about the ways that music is articulated and presented.
Finally, the research can be criticized for not considering a wide enough range
of statistical data. Two of their peak years – 1964 and 1991 – can be
explained, in part, by changes to chart rules. Billboard did not have a separate chart for
black music between November and January
1965. As a consequence, there was an influx of soul music into the Hot 100 in 1964. Similarly, it was in 1991
that Billboard first used the sales information
from barcodes to determine its chart positions. Hip-hop consequently gained a
greater chart presence, as it was selling more records than had been previously
been quantified.
The
academic researchers are media savvy. They have pointedly come up with three
revolutions, thus tapping into the ‘rule of three’ beloved by storytellers,
politicians and joke tellers alike: if you want to make a list stand out, then
give it three items. They have also come up with three curious years. 1964, the
year of the Beatles invasion of America, might be the most obvious of their
dates for musical upheaval, but the researchers excitedly report that the
Beatles were the result, rather than the cause, of this revolution. They also
stress that 1991, the year of hip-hop, represented the most revolutionary phase
of all. This revelation has prompted headlines, such as CNN’s ‘Hip-Hop is More Important than the Beatles’.
The musical revolutions are not, in fact, the main emphasis of the scientists’
paper. They are instead more concerned with publicizing their data methods as a
whole. The team has nevertheless latched on to the fuss they have generated and
have re-branded their material for more popular media. In The Conversation they boast: ‘How We Discovered the Three Revolutions
of American Pop’.
The two
big data reports are both concerned with age and new music. They come at their
target from different angles, however. Kalia analyses new music in purely
quantitative terms. For him it represents the latest releases by the latest
artists. He seeks to determine the age at which we lose interest in these new
forms. ‘The Evolution of Popular Music’ adds a qualitative dimension. The
academics want to know the eras in which music was at its newest: were there
times when it was more revolutionary than others? In doing so, they address a
widespread belief, particularly amongst the old, that there is ‘a relentless
decline in cultural diversity of new music’. The two surveys could be said to
answer each other: one reason why older people are not interested in ‘new
music’ is because it is not, in fact, new.
Things
aren’t quite that simple, however. The team from Queen Mary and Imperial
College retain a faith in newness. For them, ‘musical diversity has not
declined’. I agree with them. As I have argued before, the music of the modern era
has a distinct timbral quality and it features particular ways of singing. Does
this mean, then, that Kalia is right: older people no longer have the appetite
for newness?
Well,
it all comes back to which meaning we want to wrestle from that word ‘new’. The
word can have qualitative meanings: it can point towards things that are ‘unfamiliar or strange’. It obviously has quantitative meanings as
well, but these are complex. The OED
defines ‘new’ as being ‘produced, introduced, or discovered recently or now for
the first time; not existing before’. It also states that the new can be
‘already existing but seen, experienced, or acquired recently or now for the
first time’. In addition, new can be ‘superseding and more advanced than
another or others of the same kind’. It is therefore quite possible for an old
person to lose faith in the new because they feel that is already exists. On
the contrary, they may turn away because they find the new too new. It is
strange. They think that it’s revolting.