Should we recognise
humanity in another species?
In the 21st century, a
growing campaign in the West has argued that orangutans and other great apes
should be given qualified ‘non-human’ rights. The campaign focuses
on protecting great apes from being held in captivity (in zoos and circuses)
and from being used in medical experiments. This call is for rights going
beyond humane treatment for all animals and beyond the protection of great apes in the name of conservation.
Instead,
it argues that the close neurological similarities between humans and great
apes (and sometimes a few other animal groups such as dolphins, elephants and
parrots) mean that these animals need to be treated with special consideration.
In the case of orangutans, the argument rests on impressive evidence that they
have complex social relations and are even able to feel compassion for other animals.
Fascination
with the similarity between humans and orangutans goes back more than three
centuries. Scientists, philosophers, theologians and creative writers have
juggled with just how much importance to attach to anatomy, intelligence and
culture in deciding where to place the orangutan in the human family.
In early
travellers’ tales, it can be difficult to distinguish between stories of
orangutans and stories of exotic human communities. Scattered through the
travel literature are accounts of orangutans who hunted elephants, made fire,
cooked meals, built simple huts and buried their dead.
Most of
the orangutans to reach the West during the 18th and early 19th centuries,
however, were infants, taken from their mothers in Sumatra and Borneo and
shipped to Europe as pets. The Dutch Stadholder Willem had one; Napoleon’s
wife, the Empress Josephine had another.
Few of
them survived even a single European winter, but while they lived, they
captivated their owners and broader publics with cute behaviour. They would
wear clothes, eat in a genteel way in imitation of humans, and show a
child-like affection for their carers.
These
observations, along with the travellers’ tales inspired the French philosopher
Rousseau to conclude that orangutans were indeed humans whose evident lack of
development was due to their isolation in remote jungles. He and other
philosophers of the day were fascinated by so-called ‘wild children’ who had
grown up in forests away from human company and who had the physical from of
humans but none of their social characteristics. The name given to such people,
homme sauvage (‘man of the forest’) was an exact equivalent to orang
utan (Malay, ‘man of the forest’) and the Dutch ‘bosjesman’, which was
applied to the indigenous peoples of southern Africa.
This
early recognition of orangutans as humans crumpled in the face of three
developments. First, specialists in the new discipline of anatomy began to
demonstrate that the physiological differences between orangutans and humans
were significant. Second, growing Western knowledge of other parts of the world
pushed the early stories of ‘advanced’ orangutans into the realm of myth. And
there was a sharp attack on orangutan status in the name of human rights.
Although
Rousseau is hardly to blame, his idea of orangutan human-ness were taken up by
the proponents of slavery. Recognising orangutans as human made it easier to
argue that there was a hierarchy of humankind, stretching from orangutans at
the bottom through indigenous Australian and southern African peoples, then
Africans, then Asians, all the way up to Europeans.
The
advocates of slavery even recalled a story that orangutans could speak but
chose not to do so in order to avoid being enslaved. They argued that hard work
was a characteristic of civilisation and that in this way slavery worked to
ennoble primitive peoples who would otherwise be trapped in ape-like indolence.
In these
circumstances, the advocates of human equality argued vehemently that
orangutans should not be considered human. They emphasised the wretchedness of
orangutan life in the jungle and stressed the unbounded savagery of great apes.
The culmination of this portrayal was King Kong. Despite his filmic resemblance
to a gorilla, his origin from an island off Sumatra makes him biogeographically
an orangutan.
Ironically
a similar concern with human rights underpins some of the modern-day resistance
to recognising special rights for great apes. The argument is sharp: if we take
the cognitive and emotional capacity of orangutans as definitive in giving them
rights, there is a risk that we devalue
the human rights of profoundly disabled humans.
The
resemblance between humans and orangutans is discomforting and intriguing. The
idea that we could reach across the species boundary and recognise humanity in
another creature is both inspiring and repelling. Perhaps, however, the lack of
respect that we show for Neanderthals, who are much closer to us than
orangutans, suggests we should not expect too much.
Robert Cribb is Professor of Asian History at
the Australian National University and, with Helen Gilbert and Helen Tiffin,
author of ‘Wild man from
Borneo: a cultural history of the orangutan‘
No comments:
Post a Comment