Would Google’s LaMDA say Happy the elephant is a person?

The recent court case centring on an elephant in a New York zoo shows the difficulty facing those pushing to have animals recognised as legal people with rights
There seems to be more interest in whether artificial intelligence should be given rights once it passes the threshold of sentience

Danny Friedmann

Published: 3:00am, 25 Jun, 2022

Why you can trust SCMP

One of the most important court cases of the year recently transpired at the State of New York Court of Appeals but hardly drew attention from the general public. This happened after a historic hearing on May 18, where the Nonhuman Rights Project advocated for the release of an Asian elephant being held in the Bronx Zoo to an elephant sanctuary.

The elephant, named Happy, was taken from the wild in Thailand when she was a year old and held in captivity for over 40 years in a two-acre (0.8 hectare) enclosure, which is a tiny living space for an elephant. In contrast to the Bronx Zoo’s announcement that it would end their elephant exhibit if they had only one elephant left, it kept Happy living alone for the last 16 years.

This court case could have caused an overdue shift in the relationship between human animals and non-human animals. This elephant would no longer be a thing but a legal person with rights.

The common law concept of habeas corpus – which in Latin means “you have the body” – has been used to let a court decide whether a person was lawfully imprisoned. It was used innovatively in 1772, when Lord Mansfield accepted the use of habeas corpus to free James Somerset, an enslaved person, marking a watershed moment in the abolition of slavery.

New York’s top court found that Happy, as a non-human animal, is not entitled to habeas corpus, a legal concept that allows humans to claim unlawful detention in court and seek release. Photo: TNS

Steven Wise, founder and president of the Nonhuman Rights Project, is using the same concept to try to free cognitively complex, autonomous non-human animals being held in captivity. Corporations, rivers in India and a former national park in New Zealand can already be considered legal persons, a legal fiction and container term to grant them rights. This brings legal personhood for non-human animals within reach as well.

Elephants are social, autonomous beings that remember the past, grieve their lost loved ones and plan their future. A matriarch has to lead her herd to water sources, sometimes across long distances, while remembering them from decades ago and setting the course. Happy never got the chance to become a matriarch.

On June 14, judges on New York’s highest court voted 5-2 and rejected that Happy was unlawfully confined at the Bronx Zoo. Writing in dissent, Judge Rowan Wilson said: “When the majority answers, ‘No, animals cannot have rights,’ I worry for that animal, but I worry even more greatly about how that answer denies and denigrates the human capacity for understanding, empathy and compassion.”

01:57

Elephant ‘food court’ built in southwestern China to encourage wild herds not to stray

There seems to be more interest in whether artificial intelligence should be given rights once it crosses the threshold of sentience. Blake Lemoine, a Google software engineer, claimed the AI chatbot LaMDA had become sentient and had a soul.

He has advocated for the AI to become a legal person with rights. Google put Lemoine on leave earlier this month. Inventor and futurist Ray Kurzweil predicted that the singularity – the moment when AI intelligence exceeds that of a human – will take place around 2045.

Perhaps it is human to be so pleased with the creation of AI to care about its feelings and consider making a sentient AI into a legal person with rights. Then again, I find it extraordinarily awkward that most people – human animals to be precise – do not bother to question whether it is ethical to disqualify non-human animals as things.

Rene Descartes wrote that non-human animals are automata, merely machines without reason, mind, thought or pain and their souls, if they have any, are not immortal unlike the souls of human animals. In contrast, Jeremy Bentham wrote: “The question is not can they reason, nor can they talk, but can they suffer? Why should the law refuse its protection to any sensitive being?”

Drug lord Escobar’s lawyered-up hippos recognised as people by US court

If an AI starts to make decisions that are more intelligent than human animals, it might immediately give personhood to non-human animals. Paradoxically, sentient AI might be capable of making a more sensible distinction between a legal person and a thing.

If they cannot, our own future as sentient beings will be bleak. Let us be a good example and emancipate our non-human animal brothers and sisters ourselves before the dawn of the singularity.

Danny Friedmann is assistant professor of law at Peking University School of Transnational Law in Shenzhen

Post