How animals feel and think

Text: Marieke Enter | Illustrations: Rhonald Blommestijn

When the Wageningen animal scientist Karel de Greef was doing his degree, ‘quite a while ago’ as he puts it, nobody had any doubt that fish could not feel pain. Yet nowadays the catfish that are fitted with a small transmitter in their abdominal cavity for a Wageningen Marine Research monitoring project are first placed in a tank containing 2-phenoxyethanol to anaesthetize them before the researchers make an incision in their abdominal wall. That is because these days the precautionary principle is applied for fish: it has to be assumed they can feel pain. There is sufficient consensus among scientists about that at any rate, although it has yet to be proven incontrovertibly whether pain stimuli penetrate their consciousness. ‘That same precautionary principle will be applied to insects too one day,’ predicts De Greef, referring to society’s changing views about animals.

It is a bit of a chicken-and-egg question: is our relationship with animals changing because we know more about them thanks to research, or is that changing relationship in itself the reason why we want to do research to learn more about them? Regardless, there has indubitably been a shift. Take relatively recent initiatives such as the Dutch Party for the Animals, the Kalverliefde brand of milk whose unique selling point is that the calves are kept with their mothers, or the dating service to help find rabbits – social animals by nature – a suitable mate. ‘The point of view that humans are entitled to rule over animals is slowly making way for an attitude that is typified by concern for animals and acceptance of their unique nature. We are heading towards a new relationship between humans and animals,’ concluded the Council on Animal Affairs back in 2019.

These words come from the first State of the Animal report, on a major survey of the Dutch general public’s views on animals. In the second survey four years later, it turned out people attached even more importance to animal welfare. The respondents gave a score of 8.4 out of 10 for how much animal welfare matters to them. Moreover, over a quarter of the respondents thought humans and animals are equal, while the proportion who thought humans are superior had fallen to 36 per cent.

Animals’ language and culture

There are solid scientific foundations for that paradigm shift, says the animal ethics specialist Bernice Bovenkerk, an associate professor in the Philosophy chair group. ‘Increasingly, research on behaviour and language is showing that many animals live far more complex lives than we assumed in the past.’ It turns out, for example, that animals can have cultures. ‘It has been known for a while that primates use tools and pass on knowledge about that to each other – which is a form of culture. But now the same turns out to apply for bumblebees and honeybees.’

That has been revealed by the work of the German behavioural ecologist Lars Chittka. He used classical conditioning – the Pavlov method, if you like – to train bumblebees to solve a puzzle together that required a combination of actions a single bumblebee would never be able to perform. He carried out the same experiment with pairs of trained and untrained bumblebees. Although the untrained bees didn’t know what the intention was, they were soon doing what was needed to solve the puzzle. That shows bumblebees transmit knowledge amongst themselves.

Animals have also been found to use much more complex language systems than was previously assumed. Bovenkerk: ‘You were never allowed to call the sounds animals make “language” because language was seen as something characteristic of humans that had to include at the very least a sentence structure – grammar and syntax. The American biologist Slobodchikoff made short shrift of that idea. After years studying the calls of a group of prairie dogs, he demonstrated that the meaning of a call is partly determined by the order in which the sounds are made. Which is syntax.’

More humility

Bovenkerk says the recent findings suggest humans ought to have more humility. ‘We often reckon we know what is good for an animal, but we can’t really assume that is the case. Animals can decide better for themselves. Methods have been developed for some animals, mainly domesticated ones, to find out what they want. For example, horses are well capable of indicating whether or not they want a blanket. Cats can also make perfectly clear what they want.’

That has to do with agency, an important concept in animal ethics. According to Bovenkerk, you can define agency as the ability to act. ‘The crucial thing for animal ethicists is that it’s a relational concept: the context can increase the animal’s agency or restrict it. In a very sterile environment with hardly any room for choice, an individual animal may find it hard to develop any agency at all.’ She spent the past four years investigating the concept in a research project funded by a Vidi grant. She will be publishing a book on the topic for a general readership this autumn.

In practice, agency is often measured by the extent to which an animal has self-determination, freedom of choice and control over its own life. Those are key factors in determining the welfare of animals, says Bovenkerk. ‘Animal scientists are now agreed that animals have a perspective on their own lives. They have personal preferences and make their own choices on how they want to lead their lives.’

In dairy farming practice, such insights are reflected in the modern barns with freedom for the cows to roam, in contrast to the barns with cubicle stalls. Milking robots also give dairy cows more agency than the traditional milking setup, as the cow rather than the farmer decides when it is milked. Agency doesn’t mean the animal should always be able to do exactly what it wants, emphasizes Bovenkerk, if only because one animal may exercise its agency at the expense of another animal. That is a relevant issue not just in the savannahs of Africa, for instance, but also in Dutch gardens. Think of the agency of pet and stray cats versus that of the birds they catch in excessively large numbers. Should ‘nature’ – if you can call it that – be allowed to take its course or do humans have a moral duty to intervene? And who decides, on what basis, whose agency takes priority?

‘You see the same dilemma in the polarization around the wolf. Not everyone is willing to accept wolves killing sheep and other farm animals – over 2400 in the Netherlands in 2024. Some people would rather the wolf’s agency was restricted, for example by giving them a limited area in which to roam. But these same people don’t see a problem in the fact that humans kill sheep, tens of thousands of them a month. Modern livestock farming is quite problematic anyway from an agency point of view. It’s always the animals who come off worse. That shouldn’t be the case. When animals live alongside humans, that should not be disproportionately at the animals’ expense.’

Weighing up interests

‘The ideal might be to aim at all times for what is best for the animal,’ says De Greef, a researcher in animal husbandry & society at Wageningen Livestock Research, in response. ‘But when keeping animals, you always have to weigh up the interests of the animals against other interests. That applies to the family dog that has to be kept on a leash, the rat or mouse you don’t want in your cellar or the sow on a pig farm who usually doesn’t have the option of taking a mud bath. They all face big constraints on their agency. But the animal’s interests aren’t the only ones at stake.’

And you will always be weighing up interests, he stresses, whatever the category – pets, farm animals, hobby animals, circus animals, lab animals or even wild animals. ‘In natural or semi-natural environments, humans intervene to control the number of animals or restrict them to a certain area because we don’t want them being run over.’ He therefore doesn’t think you should consider animal welfare purely from the perspective of the animals. ‘That turns you into a kind of advocate on behalf of the animal and you lose sight of the why society weighs up interests as it does – including economic aspects.’

An important question when weighing up those interests is what society understands by animal welfare. ‘In the past, it was mainly about preventing adverse effects: no suffering, pain, stress, wounds and other negative things,’ explains Fleur Hoorweg, an animal health and welfare researcher at Wageningen Livestock Research. ‘Now the bar is set higher, both in politics and in society at large. We want to be sure animals can have positive experiences. But what does that mean? That raises quite a few scientific questions.’

Animals Act

Those questions are particularly pressing in Dutch livestock farming after the Netherlands stipulated last year via the Animals Act that the sector must become ‘more humane to animals’ by 2040. The Ministry of Agriculture, Fisheries, Food and Nature is working on specifying what ‘animal-friendly livestock farming’ entails. The ministry will do that in the form of sector plans for the four biggest groups of farm animals, namely pigs, poultry, cows and calves. The plans will be set out in an implementation decree.

Wageningen University & Research is one of the organizations being consulted on the issues that the plans will cover. One example is air quality, an important factor for animal welfare and mainly an issue in barns because of the ammonia levels.

‘That is quite a precarious role for WUR,’ says De Greef. ‘Personally, I don’t think we researchers should be setting the norms. Our task is to do research and then, based on that, explain the situation for the animals, how they perceive things, how the farms can deal with that and how to compensate for any deficiencies. We can indicate bandwidths: when do we see signs of positive welfare, what values do we find associated with discomfort and at what point could there be damage to welfare or health? But ultimately it is up to politicians to set the norms.’

Knowledge agenda

Of course, it is better if such norms are based on sound scientific research. Wageningen has a long tradition in this area. Decades ago, investigations at Wageningen revealed the associations between farm conditions and stereotypical behaviour such as feather picking in chickens, tongue rolling in pigs and pacing endlessly back and forth in calves, all signs of frustration and discomfort. Such insights from Wageningen have been used to develop better farm systems over time for various species.

But although the research output of Wageningen’s animal scientists has increased substantially – De Greef estimates that it has doubled over the past ten years – there are still gaps in the knowledge or unanswered questions regarding some aspects. Hoorweg heads a project commissioned by the Ministry of Agriculture to assess those knowledge gaps. De Greef is also working on this project. He conducted dozens of interviews with stakeholders including livestock farmers, animal welfare organizations, retailers, supply chain players and government officials. ‘A large and complex task,’ he says. ‘Not just the broad assessment; ordering and classifying the knowledge gaps is also tricky and labour-intensive.’ The final result is a knowledge agenda that will be sent to the Dutch parliament along with the ‘Covenant on a Livestock Industry that is More Humane to Animals’.

There are plenty of questions still, in both applied science and more fundamental science. One area is positive animal welfare, Hoorweg’s research field. ‘Until recently, a lot of animal research focused on gauging negative animal welfare: how can we measure stress or pain, or rate discomfort?’ she says. ‘But animal welfare is much more than merely the absence of pain and stress. Is an animal’s happiness quantifiable, and if so how? What measures can we use to determine whether animals get the optimum degree of mental challenge in their environment?’ Research on this is starting to take off now (see inset).

Technology

The hope is that new technology can play a key role in investigations into positive animal welfare. Sensors, movement detectors and machine learning have been used for some time to identify welfare problems. Now artificial intelligence (AI) is being added to that list. Bovenkerk was awarded a Vici grant this spring, worth 1.5 million euros, for a research project on modern technology in the context of human-animal relationships. Among other things, she wants to draw up guidelines for the ethical deployment of technology.

She cites an example where that is in doubt: Norwegian and Chinese plans for very large-scale aquaculture facilities with over a million fish per pool, controlled entirely by artificial intelligence, from the supply of food and regulation of the water quality to the slaughter process. Bovenkerk: ‘That is not necessarily a bad thing for the fish; they might even be better off than with a particularly heartless human fish farmer. But if the AI makes a mistake or uses parameters that are based on an excessively narrow definition of welfare, then a million fish pay the penalty.’

In her new research project, the animal ethicist is also looking at examples of how technology can make a positive contribution to human-animal relations, and under what conditions. ‘For instance, AI could help us understand the communication or language of animals such as whales, or recognize individual animals’ preferences. I am convinced modern technology can show people that there is much more to animals than they always thought’.