|
Why are they so popular?
(Filed: 16/06/2004)
Random copying explains why some
cars, dogs and pop singers are fashionable – and why Apple could become
a popular name. Roger Highfield reports
Some moan that the work of contemporary artists has no redeeming merit.
They carp that haute couture is an empty charade. They are equally
scornful about the never-ending procession of celebrities who seem to
be famous for being famous.  | | The
re-release of 101 Dalmations led to an increase in demand for the
breed, but Britney's success seems to have had a detrimental effect on
the name |
They are
not culture-blind philistines, however. According to research by an
international team on the mathematics of cultural transmission, they
are right to suspect that fad and fashion are all show and no substance. Among
academics in the social sciences, there has been a debate about why
certain ideas become fashionable, while others are outmoded. Some
argue that cultural ideas - whether the cut of clothes, designs on
pots, or choice of first names - spread simply by random copying (that
is, they are popular because, er, they are popular). Others
believe that cultural ideas become fashionable because they have some
symbolic meaning, merit or function that helps them to thrive in a
particular society. Over the past year or two, a
team has been funded by the Arts and Humanities Research Board to test
why cultural ideas take hold in the population by looking at the rise
and fall in the popularity of dog breeds, baby names, patterns on
ancient clay pots, copying of internet links, sales of pop albums and
subjects tackled by US patents. Their findings
provide powerful support to those who believe that beige has as much
merit as any other colour, who cheered when the works of Tracey Emin,
Damien Hirst and the Chapman brothers recently went up in smoke, and
who still puzzle over why no one likes to be called Cyril or Gertrude
any more. What's in a name? Not much, according to
the researchers. Is one brand of art better than another? No. When
bell-bottom jeans of the 1970s gave way to tight-fitting bleached jeans
in the 1980s, did it say anything meaningful about culture? Nope. Is
green the new black? No more than red or blue or orange is, according
to research by Dr Matt Hahn of the University of California, Davis,
Prof Hal Herzog of Western Carolina University and Dr Alex Bentley and
Prof Stephen Shennan of the Centre for the Evolutionary Analysis of
Cultural Behaviour, University College London. The
simple process of imitation is crucial for human society and has
allowed ideas to be passed from generation to generation. The same
process is at the heart of how fads and fashions have fared over the
past century: in essence, if they like it, I like it too. This model
can closely reproduce trends. Take the puppy
registrations to the American Kennel Club over the past few decades.
The top three breeds in 1950 - cocker spaniels, followed by beagles and
boxers - were replaced by Labrador retrievers, followed by golden
retrievers and German shepherds in 2001. The
breeds that have shown the fastest rise in popularity over the past
five years are the Havanese, cavalier King Charles spaniels, Brussels
griffons and French bulldogs. "At some point dogs
went from being a custom - something that people kept because they were
important in their lives and had utility - to a fashion," said Prof
Herzog. "This is an incredibly nice model for looking at rapid changes
in culture.'' Something, anything, whether a
mullet hairstyle (argh!), love of progressive rock (never again!) or
flared trousers (heaven forbid!) forms a distribution over time which
resembles the mathematical pattern that would be produced by random
copying. The resulting distributions of the popularity of a name,
hairstyle or whatever follow an elegant mathematical function called a
power law, which sums up how there are many uncommon varieties and a
few very popular ones, thousands of times more popular than the
majority. This comes as no surprise. Power law distributions are everywhere. They are found in the movements of the stock market and the size of earthquakes, where there are lots of small events, a few big ones and an intermediate number of average-sized ones. The
law also rules the links between sites on the internet. If the message
of this work is that popular ideas persist, you may wonder why we are
no longer dressed in furs and playing bone flutes in caves. The
research does not mean that once a clothing design, dog breed, name or
school of art is popular it will remain so. Although
the most common names tend to remain common, random forces occasionally
unseat the cool, hip and chic. The team finds that the comings and
goings of popular ideas can be modelled with "random drift" - people
randomly copy existing fads and fashions from others, with new ideas
emerging from time to time to churn the overall variety. Thus,
though rare, it is inevitable that a few parents who invent an offbeat
name for their baby, such as Apple, will unwittingly determine the
names of thousands of children in as little as a decade. Similarly, a
relatively unusual breed, such as the chow-chow, can occasionally
unseat more popular dogs. "The popularity of some
breeds just take off - much like a social epidemic," said Prof Herzog.
"This is the case for booms that occurred with Doberman pinschers and
Saint Bernards.'' Nor does this research suggest
that random copying of cultural ideas accounts for everything, though
it is the dominant force on popular culture. Non-randomness clearly
exists alongside the basic random pattern. There is a celebrity effect,
though it does not always work the way that you expect. One
might expect the world to be blessed with more Britneys than it used to
be, but Ms Spears seems to have had a negative effect, said Dr Bentley.
The name was ranked three in the United States in 1995. In 1999, when
she released Baby One More Time, it fell to 37th and last year
languished at around 211. The celebrity effect
can be seen more clearly in dog breeds. Dalmatians became popular after
the re-release of a film version of Dodie Smith's novel 101 Dalmatians
(the original release triggered a modest rise).
Registered English sheepdogs soared from 149 in 1960 to more than 16,000
in 1974 after Disney's The Shaggy Dog in 1959. And rottweilers were
sent into decline by the bad publicity associated with an increase in
fatal attacks in the mid 1990s. Prof Herzog,
citing earlier work by Stanley Lieberson at Harvard University on the
rate of changes in baby names, found a similarity in the timescale of
the waxing and waning of dog breeds and first names, again emphasising
how a common mechanism is at work. "For example,
the rise and fall in the popularity of 'Jennifer' are nearly identical
to the rise and fall of Irish setter registrations at about the same
time, in the late 60s and early 70s.'' Celebrity
is transient, media scares eventually subside and familiarity does
indeed breed contempt. Today, the name John - just like the Dalmatian -
is far less popular than it used to be. And although the scientists
cannot say what will become popular in the next decade, they are
certain that some tiny blip on the cultural landscape today will
balloon into a mass phenomenon. The team has now
studied other real-world examples such as archaeological pottery and
applications for technology patents. According to a recent report in
the Royal Society journal Proceedings B (Biological sciences), the same
power law pattern expected of random copying holds sway. Even
7,000 years ago, a community of Europe's first farmers show random
copying patterns in the way they decorated their pots, according to
Prof Shennan. The findings have wider
implications. The bad news is that inequality (among randomly-copied
ideas) is inevitable, because a power law distribution of popularity
always emerges among the available choices.
"However, if you want to reduce the inequality - reducing the disparity
between the most and least popular - you need a larger population,"
said Dr Bentley. "If you hold the population constant, increasing
mutation rate also decreases inequality.'' The
same goes for money, which is essentially 'copied' from one person to
another, he said. "Since wealth in modern economies is power-law
distributed, the model could predict how reducing inheritance
(effectively 'innovation' by refreshing property ownership each
generation, for example, or punitive inheritance tax) would reduce
wealth inequality." This appears to give the
Chancellor a scientific rationale for using tough taxation to reduce
social inequality. But, of course, the vast majority of people hate
giving their money to HM Treasury. The scientists' model would suggest
that this is one fashion that will never catch on.
Next story: Bryson's scientific journey wins award
|
