Sociology. Anthony Giddens
assigned a role as either a guard or a prisoner. Following a standard induction process, which involved being stripped, de-loused and photographed naked, prisoners stayed in jail for twenty-four hours a day, but the guards worked shifts and went home afterwards. Standardized uniforms were used for both roles (Haney et al. 1973). The aim was to see how playing the different roles would lead to changes in attitude and behaviour. What followed shocked the investigators.
Students who played the part of guards quickly assumed an authoritarian manner, displaying real hostility towards prisoners, ordering them around, verbally abusing and bullying them. The prisoners, by contrast, showed a mixture of apathy and rebelliousness – responses often noted among inmates in studies of real prisons. These effects were so marked and the level of tension so high that the fourteen-day experiment had to be called off after just six days because of the distress exhibited by participants. Even before this, five ‘prisoners’ were released because of extreme anxiety and emotional problems. However, many ‘guards’ were unhappy that the study had ended prematurely, suggesting they enjoyed the power the experiment afforded them.
Reactions to the mock prison regime in the Stanford prison experiment led one inmate to stage a hunger strike just to get out.
On the basis of the findings, Zimbardo concluded that the dispositional hypothesis could not account for the participants’ reactions. Instead, he proposed an alternative ‘situational’ explanation: behaviour in prisons is influenced by the prison situation itself, not by the individual characteristics of those involved. In particular, the expectations attached to the roles being played tended to shape people’s behaviour. Some of the guards’ behaviour had deteriorated – they treated prisoners badly, regularly handing out punishments and appearing to take pleasure in the distress of the prisoners. Zimbardo suggested this was due to the power relationships the jail had established. Their control over prisoners’ lives very quickly became a source of enjoyment for the guards. On the other hand, following a short period of rebelliousness, prisoners exhibited a ‘learned helplessness’ and dependency. The study tells us something important about why social relationships often deteriorate within prisons and, by implication, in other ‘total institutions’ (Goffman 1968 [1961]). This has little to do with individual personalities and much more to do with the social structure of the prison environment and the social roles within it.
Critical points
Critics argue that there were serious ethical problems with this study. Participants were not given full information about the purpose of the research, and it is questionable whether they could really have given ‘informed consent’. Should the study even have been allowed to go ahead? The sample selected was clearly not representative of the population as a whole, as all were students and all were male. Generalizing about the effects of ‘prison life’ is therefore very difficult based on such a small and unrepresentative sample. The constructed nature of the situation may also invalidate the findings for generalizing to real-world prison regimes. For example, participants knew their imprisonment would last only fourteen days, and they were paid $15 a day for their participation. Well-established problems of prisons such as racism, violence and sexual abuse were also absent. Critics say that the experiment is therefore not a meaningful comparison with real prison life.
Contemporary significance
In spite of the somewhat artificial situation – it was an experiment, after all – Zimbardo’s findings have been widely referred to since the 1970s. For example, Zygmunt Bauman’s Modernity and the Holocaust (1989) draws on this study to help explain the behaviour of inmates and guards in Nazi-run concentration camps during the Second World War. In recent years the issue of mistreatment and bullying of older and disabled people in care-home settings in England has been exposed in a series of scandals, resulting in dismissals and the prosecution of staff members. Zimbardo (2008) himself discussed some of the parallels between the experiment and real-world episodes, including the abuse and torture of Iraqi prisoners in Abu Ghraib in 2003–4, suggesting that the focus should be not on finding the ‘bad apple’ but on reforming the ‘bad barrel’. His most general thesis, that institutional settings can shape social relations and behaviour, remains a powerful one.
THINKING CRITICALLY
The fact that some students left early and the experiment was cut short suggests that the experiment’s effects were significant. But which aspects of prison life could experiments of this kind never replicate? Write a 500-word position paper explaining the benefits of allowing sociologists to conduct experiments using human beings. What are the counterarguments?
A historical perspective is also essential in sociological research, as we frequently need a time perspective to make sense of the material we collect about a particular problem. Sociologists commonly want to investigate past events directly. Some periods of history can be studied in a direct way while there are still survivors around, and there have been some insightful studies of the Holocaust in Europe during the Second World War. Research in oral history means interviewing people about events they were part of or witnessed at earlier points in their lives. This kind of direct testimony can be gained, at most, for sixty or seventy years back in time.
For historical research into earlier periods, sociologists draw on documentary research, using written records often contained in the special collections of libraries or other archives. The range of useful documents is extensive, taking in personal sources such as diaries, official sources such as policy documents, records of births and deaths, tax records, and documents from private bodies such as businesses and voluntary organizations, as well as magazines and newspapers. Depending on the research question, historical documents such as these can all constitute primary sources just as much as the data recorded in interviews with war survivors. However, historical sociologists also make use of secondary sources: accounts of historical events written by people after the event. Most documentary studies utilize both primary and secondary sources. However, sociologists face the same issues as historians when they use such sources. How authentic are the documents? Is the information within them reliable? Do they represent only a partial viewpoint? Documentary research requires a patient, systematic approach to sources and their interpretation.
An interesting example of the use of historical documents is Anthony Ashworth’s study of trench warfare during the First World War (Ashworth 1980). Ashworth drew on diverse documentary sources: official histories of the war, official publications of the time, notes and records kept by soldiers, and personal accounts of war experiences. He was able to develop a rich and detailed description of life in the trenches that contained some surprises. For instance, he found that most soldiers formed their own ideas about how often they intended to engage in combat and often ignored the rules and commands of their officers.
Ashworth’s research concentrated on a relatively short time period – 1914 to 1918 – but there have been many studies investigating social change over much longer periods, making use of comparative research in that historical context. A modern classic of comparative historical sociology is Theda Skocpol’s (1979) analysis of social revolutions, which is discussed in ‘Classic studies’ 2.2.
Although anthropology has long made use of visual sources of information such as photographs and film footage, sociology has tended to be a subject focused on written texts (Harper 2010). That is not to say that sociologists do not produce their own visual materials. Representations of numerical and statistical information are turned into easy-to-read pie charts, tables and graphical representations, while ethnographic research is often presented with photographs included. However, these visual elements are almost always ancillary to the main text, which is the more significant part of the articles and books through which the sociologist’s