BlaiseLaPsy on Twitter, in response to what I said about the Freudian concept of social instinct, raised an important point about the work of John Bowlby and Harry Harlow that appeared to provide evidence for the existence of a social instinct. I have reservations about the conclusions drawn from Bowlby’s and Harlow’s findings, and about how the term ‘instinct’ is used. First, a brief round-up of Bowlby and Harlow’s research.
Bowlby, who graduated in medicine and qualified as a psychoanalyst in the UK in the 1930s, was interested in the development of children with behavioural problems and those who had been separated from their parents due to being orphaned or hospitalized. Influenced by René Spitz’s work on orphans, Bowlby became an authority on the effects of maternal deprivation and developed Attachment theory. He concluded that for normal social development, children need a secure relationship with a primary caregiver (usually the mother). Mary Ainsworth later found in her ‘strange situation’ experiments that children showed one of four patterns of attachment to their primary carer.
Harlow qualified at the same time as Bowlby but had a very different academic background. He was an American psychologist; his PhD supervisor was Lewis Terman, who developed the Stanford-Binet IQ test. Prompted by Bowlby’s work, Harlow studied maternal deprivation in rhesus monkeys and macaques. His most famous experiment showed that rhesus monkey infants raised with substitute ‘mothers’ consisting of a wire frame covered/not covered with a terry cloth, preferred the cloth ‘mother’ and clung to it when frightened, even if it was only the wire mother that provided milk. The baby macaques were raised for varying lengths of time in isolation; Harlow looked at the effect on their development, which was invariably abnormal.
What the work by Spitz, Bowlby, Ainsworth and Harlow appears to show is that human and/or primate infants have a social instinct and that instinct triggers a typical pattern of social development. If an infant’s relationship with their primary carer is disrupted by lengthy separation, abnormal social development results. My main problem with these conclusions is the concept of ‘instinct’. Instinct is one of those constructs like ‘love’ or ‘education’ that everybody thinks they understand until they try to find out how it works, or until they discover that their concept of it is different to someone else’s.
We all know what we mean by ‘instinct’ – an automatic, unconscious behaviour. We know what Bowlby, Harlow and others mean by ‘social instinct’ – it’s an automatic, unconscious, typical pattern of social behaviour that appears to develop in the same way in everyone unless something stops it. But Bowlby, Harlow and their contemporaries faced three problems when it came to instinct.
First, as Blaise points out, conceptual models are influenced by cultural worldviews. At the time Spitz, Bowlby, Ainsworth, Harlow and Kanner were researching, human social behaviour was generally assumed to be governed by instincts. Darwin’s work on natural selection implied that many characteristics peculiar to a given species – physical features, physiology and behaviour patterns – were inherited. All male blackbirds have similar songs. Bowerbirds build and decorate complex bowers to attract mates. Ants live in complex colonies, dogs run in packs, cats tend to be solitary. Because human beings tend to behave socially in similar ways across cultures, there was no reason to suppose human social behaviour wasn’t as instinctive as that of blackbirds, bowerbirds, ants, cats or dogs. The main alternative to the psychodynamic framework at the time was Watson and Skinner’s behaviourism, which proposed that complex behaviours such as social interaction were learned. But behaviourism was widely treated with suspicion because it was seen as reductionist. (How can you reduce something as nuanced and complex as social interaction to something as basic as a rat’s tendency to run through tunnels or a pigeon’s tendency to peck?)
Secondly, no one working in child development prior to the 1960s knew much about how the brain worked. They were all guessing. Their guesses were often extremely well informed, but they were guesses nonetheless. Spitz, Bowlby, Ainsworth, Harlow and Kanner all came down on the ‘instinct’ side; Watson and Skinner on the ‘learned’ side, but none of them knew about the biochemical mechanism of learning in the brain.
Thirdly, none of the child development researchers needed to figure out how social instinct worked because the idea of ‘instinct’ itself explained their findings. It was a ‘black box’ concept. They didn’t know what was inside it and didn’t need to know; what they were interested in was what happened when the social instinct was disrupted.
People who did need to figure out how instinct worked and what was inside the black box, were ethologists studying the development of animal behaviour. In the 1950s, researchers such as Lorenz and Tinbergen began to look more closely at the difference between instinctive and learned behaviour. Most people are familiar with the famous pictures of Lorenz being followed by a column of baby geese. Because goslings and ducklings follow their mother from the moment they hatch, it was assumed that this was an instinctive behaviour. What Lorenz discovered was that the tendency to follow something was instinctive, but that what the goslings followed was learned – they followed the first moving object they saw after hatching. It might be their mother, a chicken foster-mother, the farm dog, a pair of boots (didn’t matter who was wearing them) or Konrad Lorenz.
There was a debate about instinct amongst ethologists in the 1960s because it had become clear that different researchers were using the term in different ways and so definitions got tightened up. Unfortunately, apart from Bowlby, many practitioners working in medicine or psychiatry wouldn’t have read the ethology literature – they weren’t (and often still aren’t) interested in the behaviour of goslings or wild macaques even if they had time to keep up to date with it. That’s a pity, or a tragedy depending on how you look at it, as far as instinct is concerned because what has emerged from animal behaviour research is a picture of instinct as an umbrella term that can be applied to a range of different concepts. Essentially, instinct refers to behaviours that are genetically determined, biologically controlled, automatic and unconscious. But it isn’t quite as simple as that.
Levels of instinctive behaviour
Starting at the lowest level, human physiology is genetically determined, biologically controlled, automatic and unconscious; circulation, respiration, digestion, growth and sexual development occur without any awareness or intervention on our part although we are aware of what happens as a result of them. We know a lot about how these autonomic functions work and that they are very similar in everybody. But we wouldn’t usually call autonomic functions ‘instinctive’ because instinct is about how organisms behave rather than how they function.
The most simple form of instinctive behaviour is the reflex – a simple, automatic motor response to specific stimulus. Reflexes – such as the rooting, palmar grasp, startle, swimming and stepping reflexes – are present from birth. Some have obvious survival value; others, like the stepping reflex, form the foundation for behaviours that emerge later – in this case, walking. We know a lot about how reflexes work and that they are very similar in everybody. Most people would classify reflexes as instinctive, I think.
More complex species-specific behaviours, like birds learning songs or building nests, often vary between individuals. Songbirds develop their own unique songs, bowerbirds make their bowers out of whatever materials are available. Although all human beings, regardless of culture, show similarities in social behaviour, the evidence to support the existence of a social instinct is pretty flimsy. We’d expect organisms with similar autonomic functions and similar reflexes to behave in similar ways, but that’s about as far as the evidence takes us. How people interact and communicate with each other and how frequently they do so varies much more than their autonomic functions or reflexes. Some people choose to live in tightly-knit highly interdependent groups, others to live in isolation. Some are highly gregarious, others prefer the company of cats, dogs, horses, the landscape or machines.
I’m a fan of Monkey Life, the TV documentary series about Monkey World, the primate sanctuary in Dorset, UK. A few years ago, Monkey World took in 88 capuchin monkeys from a lab in Chile. Some of them had been born in captivity, others had been captured from the wild. During their rehabilitation, Alison Cronin the sanctuary director commented that the wild-born capuchins instinctively knew how to eat their natural food but the cage-born capuchins didn’t – they had to learn to do that. Alison’s comment introduces a slightly different use of the word ‘instinct’, meaning a behaviour that happens automatically and unconsciously, but isn’t genetically determined and biologically controlled. The documentary also showed that chimps and orang-utans born in captivity tend to be poor mothers. You could argue, as the psychodynamic theorists would have done, that the capuchins’ feeding instinct and the chimps’ and orang-utans’ maternal instincts had been disrupted by their captivity and so hadn’t been allowed to develop normally. That’s one theory. What’s also possible is that the cage-born primates, or those captured in infancy, simply hadn’t had the opportunity to learn how to forage, peel fruit or rear babies.
Human social behaviour varies widely. That variation could be because the normal social instinct is disrupted by events in childhood. But because we don’t know exactly what ‘normal’ human behaviour looks like, and we have no idea how the social instinct works (in contrast to what we know about autonomic functions and reflexes), a more likely explanation is that some aspects of human social interaction are instinctive and others aren’t. Social behaviour is hugely complex, so the question is which bits of social behaviour are instinctive and which aren’t?
As I pointed out in the post about the social instinct and Kanner’s syndrome, the areas of the brain dealing with social behaviour handle complex information from many areas of the brain. From an information-processing perspective, social behaviour, far from being instinctive, results from an interaction between the way the body works, environmental factors such as nutrition, and experience. Most researchers in all areas of child development are aware of the importance of interaction between factors in development, but by necessity, they are usually focusing on one factor only, and tend to overlook anything outside their field of expertise.
Social or sensory deprivation?
A final observation about Harlow’s work. There’s no question that Harlow’s baby primates were socially deprived. But they also suffered sensory deprivation as well, and I don’t know if Harlow controlled for that. Given his conclusions I don’t get the impression he did. Some of the infant macaques, for example, were kept in total darkness for months. Light is essential for entraining circadian rhythms, so absence of light alone would have seriously messed up their physiological functions. Coincidentally, I was listening to Crossing Continents yesterday on BBC Radio 4. A former inmate of Louisiana State Penitentiary, who’d been in solitary confinement for 30 years, was describing his experiences. He highlighted, not so much the social isolation, as the sensory deprivation. He’d had a long time to think about what it was he missed; it would be all too easy to assume in the same situation we’d miss other human beings, when what we might actually miss is the complex sensory input we get from interactions with other people. I’m not trying to reduce social contact to sensory stimuli – social contact is clearly more than the sum of its parts – I’m just saying that it’s very difficult to make a distinction between social interaction and the sensory input that comes with it.
The benefits of hindsight
I’ve been quite critical of psychodynamic theorists, but I’m very aware that they were working with the knowledge that was available at the time. I think what we need to be wary of is assuming that knowledge develops in a straight line; that Freud, Bowlby, Harlow and Kanner were basically right but we now know more than they did. With the benefit of hindsight we can see which aspects of earlier theories are supported by later research and which aren’t. I don’t think there’s much evidence to support the idea of social instinct. What the evidence does suggest is that although instinctive behaviours are quite likely involved in social interaction they are only part of the story.