I rarely read a piece of scientific journalism and think “what absolute tosh”, in part because I tend not to use the word ‘tosh’ and in part because I know that science journalism involves digesting and reconfirming a complex idea. It’s not easy.

But this article had me gnashing my teeth. It’s a summary of a paper by Ganea et al 2014 [in press pdf download – only link I can find]. The essence of the paper is this: children which grow up in urban environments (in this case pre-school age children from Boston and Toronto) are not exposed to animals. When they’re given anthropomorphic stories about unfamiliar animals (cavys, handfish and oxpeckers) they will agree with statements that attribute complex emotions to those animals, but not statements which attribute human physical capabilities, e.g. talking, to the animals. The conclusion is that anthropomorphic animal stories inhibit a child’s ability to learn animal facts.

The science I think is interesting – it is the conclusion and the bandying about of the word ‘anthropomorphism’ that get my goat. Let rant at you.

The article’s author says:

Setting aside the shades of grey as to whether non-human animals have analogues for things like friends, the findings suggest that for young kids, “exposure to anthropomorphized language may encourage them to attribute more human-like characteristics to other animals than exposure to factual language.”

 

 

This anthropomorphism spectre infuriates me at times. Let me put it this way, one of the questions asked of the children was “do oxpeckers have friends?” I’m asked relatively frequently if cows have friends, and if I want to answer that question accurately, I have to dance around terminology and use baffling scientific language to answer it in a way that means ‘yes but I can’t really say that because I’m a scientist’.

Cows have preferential associations within their herd. Being with these other individuals makes them more capable of physiologically coping with stressful events (Boissy & Le Neindre, 1997) such as being reintroduced to the milking herd (Neisen et al, 2009), being milked (Hasegawa et al, 1997), or feed competition (Patison et al, 2010a). They will preferentially engage in social interactions with these preferred associations, and these associations go on for longer than with other animals (Faerevik et al 2005, Patison et al, 2010b).

How do you explain this to a 2-5 year old child from Boston without using the word ‘friend’ or any synonym of it? Is it any wonder a child might reasonably assume that animals can have friends? Is it wrong to say that an animal can have a friend?

My irritation here lies with the writer of the article saying children believed ‘falsehoods’ about animals, based on anthropomorphism. We get one link, to a website I can’t access being based in the UK, to research which might suggest animals are similar to us in some ways. Then we move on to a paper I’ve referenced before talking about how dogs’ guilty looks are based on our behaviour (Hecht et al, 2012). The underlying assumption is still that animals are so different from us that children are wrong to believe that animals have the capacity for friendship and caring.

Now I’m fascinated by dogs for precisely this reason. They are so excellent at communicating with us, and reading us, that they are almost in-animal as much as they are in-human. They’re a possible model for human-child behaviour they’re so adept at this. I wouldn’t necessarily use dogs as an example for how the rest of the animal kingdom thinks if I was very worried about making cross species comparisons.

Anthropomorphism is either the attribution of human characteristics to animals. In which case it cannot be used pejoratively. For example, to say “This cow has eyes” would be anthropomorphic.

Or anthropomorphism is the inappropriate attribution of human characteristics to animals, in which case you must carefully consider why the characteristic is inappropriate when given to animals. It is not anthropomorphic in this case to say “This cow feels fear”, because fear, as we understand it, is an evolutionary mechanism to increase your chances of survival, it has physiological and behavioural components and the cow meets all of these. Ergo, this cow feels fear, and that is not an inappropriate characteristic.

Much as I lament the fact urban children have very little contact with the natural world, and I think this is a major issue for animal welfare, food sustainability, and the mental health of the children, I don’t fully agree with the paper’s conclusions, or the writing up in the Scientific American blog.

Firstly, the study found that all children learned new facts regardless of whether they read the anthropomorphic story or the non-anthropomorphic story. The results appear to indicate to me there was less fact-retention in the anthropromorphic story (and while I’m not a psychologist, I have worked with children and I do now work in education, I wonder if the anthropomorphic story, being similar to entertainment, indicated ‘you do not need to pay attention here’ to the kids. This does not appear to be discussed in the paper.).

Secondly, the study found that the children who had anthropoorphic stories told to them were more likely to describe animals in anthropomorphic terms immediately afterwards. Now again I’m no psychologist, but after I went to see Captain America I was partially convinced I was a superhero. It faded after the walk home. I’d like to know more about the extent of this effect over time before I declared anthropomorphic stories as damaging to children’s learning.

Thirdly, the Scientific American article presents some ‘realistic’ and ‘anthropomorphised’ images of the animals side by side. This is not what happened in the paper. In the first experiment the children were shown ‘realistic images and factual language books’ or ‘realistic images and anthropomoprhic language books’. The second study used ‘anthropomorphic images and factual language’ and ‘anthropomorphic images and anthropomorphic language’. The upshot of this is that the realistic image condition was not directly compared to the anthropormphic image condition, regardless of how it seems when you read the Scientific American article.

The paper says at one point:

This reveals that, like adults, young children seem to have a less clear conception of differences between humans and other animals in regard to mental characteristics, as opposed to behaviors. However, exposure to anthropomorphized language may encourage them to attribute more human-like characteristics to other animals than exposure to factual language.

 

 

Well there’s little wonder about that because even we scientists don’t have a particularly clear conception of the mental differences between humans and other animals. The paper itself is interesting and well worth a read, but it falls into the trap of thinking about anthropomorphism as a wholly negative thing. If I was a reviewer I’d suggest Serpell (2002) as an excellent starting point for a more balanced view of the phenomenon.

And I’d also suggest they watch this video before assuming that kids are daft for thinking animals feel emotions.