Abstract: This essay aims to join the conversation of animal theory of mind. I lament that until we better understand the role of emotion in animal perceptions and memory storage, we appear incapable of finding sufficient evidence to suggest we can attribute to animals anything more than submentalising processes. I base this on how we cannot rule out that animals may store away their own emotion toward some stimuli rather than ascribe to that stimuli any agency or status as an agent. I examine a case study involving horses and argue that there is room for emotion-specific cognitive mechanisms to simulate beliefs. However, before arriving at such conclusions, I first lay out what is theory of mind, where this discussion is academically situated, and what allows Andrews to take on the title of nonpropositional mentalising Mengzian.
Descartes’ famous cogito ergo sum is supposed to give us the relief that because we actively think and hold thoughts, we must exist. However, this relief is doomed to be short-lived; once we understand we exist and contain thoughts, it is only natural to wonder about the nature of these thought-containers; whether other conscious things possess them to the same degree we feel we possess them. Thus was born the study of brains, consciousness, and theory of mind – the conception of the ability to recognise as well as attribute to others mental states including desires, emotions, intentions, and beliefs.
When we know we have a mind, and others may also potentially have minds, it follows that we would want to know what we can learn about other minds and maybe see what sorts of things we have in common. Test a few things and we quickly see that living things are capable of reacting and learning a lot, even if they do not always have access to why things are a certain way. This is the heart of what is referred to as the logical problem. In general, this is why when we test other minds, we tend to test for false-belief-understanding rather than take on belief-understanding and its impenetrable quality.
A simple false-belief test usually consists of actors One and Two. One takes an action to demonstrate to a witness he holds some belief. One leaves with the belief intact. Two intervenes and changes the outcome of One’s action without One’s knowledge. In the moment Two intervenes, the belief One had left the scene with is rendered false. Often, One returns to the scene to cement to the witness the reality that One now carries a false belief. The witness is then tested in their ability to differentiate between what is true and what One falsely believes.
In the matter of our minds and the minds of others, Andrews is rightfully critical of these false-belief tests. She does not believe that passing the test requires a theory of mind. She stands defending a social intelligence hypothesis which entails that our smarts are responsible for helping us keep track of our relations with others, ultimately guaranteeing our success as social creatures. Within the sphere of those who ascribe to this theory exist two clashing camps: that of the Machiavellian and that of the Mengzian.
We are informed briefly by Andrews that the Machiavellian social intelligence hypothesis is one advocating us as having belief/desire-oriented cognitive preprogramming. Here, only humans came up with the ‘social technology of belief attribution’. Actions are caused by a combination of both beliefs and desires, and so anyone wishing to manipulate the behaviours of others can choose either variable to affect. A Machiavellian can then extract special advantages by predicting the behaviours of others.
Rather than endorse this, Andrews supports the view: in which humans developed unique and cooperative social intelligence so as to maintain more positive relationships with others within the group. She likes this theory too because it is “consistent with the human practice of telling stories and offering explanations in terms of beliefs in order to explain, and better yet, to justify behaviour, as a means of solidifying communities when individuals act outside of the group norms”. Thus, with a Mengzian conception, Andrews argues for the primary function of belief/desire-oriented cognitive preprogramming to be that of explaining behaviour rather than predicting it. This is supported by reasoning that because any prediction is based off an explanation, explanations must always exist prior to predictions. Andrews also enjoys a special cleverness of history: that the Mengzian literally existed prior to the Machiavellian.
This reverence for explanation places Andrews staunchly in between two opposing viewpoints: valuing chimp and ape behaviour not as highly as the the propositional mentalist but certainly more than the submentalist.
Beginning with the propositional mentalist hypothesis: this theory suggest that apes and chimps are able to pass the false-belief test specifically because they “have a concept of belief, and are able to ascribe propositional attitudes such as false beliefs to others in order to predict behaviour”. However, this falls short for Andrews rather quickly. She is not sympathetic to prediction-prior-to-explanation viewpoints, and is uncomfortable with how many things are enailed by this hypothesis. She correctly points out that we do not know if apes and chimps have a concept of true and false, nor can we confirm any sort of understanding on their part of the general rule that what others see in the world can directly affect their behaviour. Though she does not particularly mention cross-species difficulties, she may be interested in questioning how we can account for an animal’s response to feeling tested/experimented upon. Though eerie to admit, we cannot truly know if animals are failing or passing our tests according to their own larger intention, putting us through grander tests of their own – and this renders useless a lot of evidence gathered through experimentation.
In opposition to high-flying assumptions, Celia Heyes’ ideas inform us of the submentalising hypothesis: that apes do not actually recognise individual agents or impose narrative structures, nor do they attach reason behind actions. A false-belief test can be passed by introducing a concept as simple as a colourblock; no belief-attribution required. Suppose: a creature looks off into the horizon and encounters normal visual impairments – distance, poor lighting, momentary lapses in periphery awareness. The creature would be unable to tell the difference between an individual with agency or a simple colourblock; and so an ‘Object’( in this case, a man wearing denim) might not appear to the creature as a living being at all. The creature would be unaware there is one object (denim) being worn by another object (man).
When the Object disappears from the creature’s line of sight and the creature turns to look for it, it may seem as though the creature is aware the Object has moved; however, it is unclear whether the creature has noticed the Object as a thing that moves itself or a thing that was moved by life. Moreover, the creature’s brain may store memories in a way not known by humans. Thus, without formulating thoughts of beliefs, the creature may face the location where it last formed a conscious memory of the Object and so give the appearance of mentalizing. Andrews is critical of this line of thought as she believes chimps do demonstrate a lot of complexity. She points out that they provide social support for each other – yet this could be easily expected in creatures of emotional complexity and does not especially help Andrews’ case as it is compatible with Machiavelian as well as Mengzian theory. Furthermore, as a Mengzian, Andrews may have to do a little more work explaining why the pressure to cooperate within social groups of a species would result in cooperation and emotional across species.
Emotion clearly plays a huge role in animal functioning. I recall a case study in which horses were shown photos of a visibly angry or happy individual. We can fairly assume the horses chosen for the experiment must have had enough human contact throughout their lives to be at least familiar with human emotions and expressions. Additionally, prior research informs us that animals tend to look at things that are negative or threatening with their left eye, as this connects directly to the right brain hemisphere. A few hours after being shown the large photo of the emotionally-affected individual, the horses met the depicted individual in real life – this time the individual maintaining a neutral countenance. The horses demonstrated the predicted pattern: peering at prior-happy individuals for more time with the right eye than the left, even going so far as to exhibit stress behaviours including chewing, sniffing the floor, licking themselves.
While it could be seen that horses store a memory of an agent from the last encounter, it could also be the case that horses store emotional reactions based on associations the horse manufactures between non-agents around him. Some tall redhead with a furious countenance might cause the horse to respond to his environment in such a way that anytime he sees a similar wispiness, colour tone, or red box-shape the size of a handbag, he switches his ‘viewing around me’ mode to ‘fearful’, and always peers with his left eye. In the case of apes and chimps, it is reasonable to believe the circumstances of being a subject of an experiment could lead the animals to start developing/storing new associations that function as beliefs they summon when dealing with humans.
In short, Andrews does a remarkable job summarising the discussion around social intelligence hypotheses. Her critical thinking skills are demonstrated well here: much of what she questions deserves an answer. Unfortunately, I am left still with the belief that an impossible amount of work remains to be done in understanding how we can approach the issue of animal minds and how they would engage with ours.