Rainforest Asylum. Sara Ashencaen Crabtree

floral epithet for a ward exclusively for male offenders, attracted my interest as did, although to a lesser extent, the sick ward.

      The acute wards, Female Ward 1 and Male Ward 1, and the long-stay wards, Female Ward 2 and Male Ward 2, were chosen on the basis of a combination of factors. The acute wards provided an interesting basis for comparison with the long-stay wards, in view of the fact that the selected long-stay wards were not exclusively but generally more likely to be the eventual destination of acute patients later perceived to be chronically afflicted. The daily routines, recreation and patient and staff interactions were livelier on the acute wards with the exception of the forensic ward. On the wards selected most of the patients enjoyed comparative youth compared with the remainder of the long-stay wards; and therefore, to generalise, were more likely to be able to communicate with me as opposed to wards where there were a greater number of elderly and mentally infirm patients.

      Overt observation techniques provided the major part of the data I gathered in which I made no pretence to on-lookers to be there for any other purpose than that of observation. This strategy conformed closely with that described by Tim May (1999: 140) in which the ‘participant as observer’ role is a public one. This involves not only observation but also the development of working relationships with participants as informants for the study. Information duly gathered in this way proved sufficient to obtain good insights into particularly interesting phenomena, such as methods of control or the use of patient labour on the wards. This also enabled me to make an informed decision regarding which wards should eventually be selected for closer scrutiny, as well as drawing my attention to those individuals whom I felt I could approach and those who might represent a threat to the study, or more prosaically, to myself.

      In the early days, however, my method of observation was closer to that of a ‘shotgun’ approach in which interesting people, events and activities were noted down with little discrimination and less understanding, in a small, handy notebook on site. As the study progressed my comprehension of events taking place around me increased and allowed me to target certain phenomena on the ward. Patient mealtimes, medication routines, bedtimes and awakenings were just some of the events I sought to observe at certain times of the day and night. I therefore made myself present for early morning breakfast rounds on the wards, and mid-morning snacks; present for soporific afternoons and patient siesta time, and occasionally kept a night-time vigil with the staff night shifts. These latter shifts proved to be the most sociable and companionable, with staff most amused by my persistence and supportive of my endurance. On these occasions I could rely on coffee and mee goring (fried noodles) to be liberally supplied to keep tired eyes open, including mine.

      At first, I had felt grotesquely conspicuous on the wards and felt that staff in varying degrees were self-conscious when going about their everyday business under these artificial circumstances. After a considerable amount of time and personal discomfort had lapsed I eventually manage to achieve a certain level of invisibility where everyone, staff and patients alike, apparently ignored my presence albeit on brief occasions. These occasions were punctuated by activities in which individuals would regularly engage me in conversations. Over time my explanations that I wanted to see what it was like on the wards began to be accepted by participants with less suspicion as to my exact motives.

      In this way, therefore my observations narrowed down over time from a broad sweep of noting everything and anything that caught my attention to a narrow, and hopefully, more acute focus (Bannister, 1999). Through the use of observation techniques employed in a comparative exercise, I found that data from observations both informed and synthesised my developing hypotheses in a rigorous and synergistic relationship (Burgess, 1995).

      Observation strategies on the wards allowed me freedom to adjust to situations taking place and consequently I would often engage or be approached by patient and staff informants. In common with Shaffir’s (1991) experiences, most of these were informal conversations on a particular topic that I wanted to explore further. These interviews being ‘unstructured’ and ‘flexible’, informants often initiated the conversation from the outset (Lee, 1993). Here my interviewing strategy tended towards a deconstructive manoeuvre of attempting to uncover hierarchal distinctions through an appearance that was casual, with informal language and mannerisms, and generally trying to avoid with varying degrees of success the attitude and appearance of an orang puteh (White) lady visitor. Conversations with patients were fluid and spontaneous with participants joining in and departing from the discussion at hand as they pleased. This less formalised approach meant that patients chose the location to discuss matters and involved various settings. Occasionally we sat on stools under trees, or on the open veranda that most wards had, sometimes in the canteen or otherwise just sitting on beds inside the ward or in the rather bare recreation room. Some conversations took place in the occupational therapy department with patients chatting to me while they worked. Sometimes patients, usually men, would approach me to ask for a cigarette, which I did not have, or money, which I concealed, and then following this overture a discussion might be struck up. Similarly casual conversations with staff took place at the nursing station on wards, in private offices during tea breaks or while carrying out duties.

      At other times, interviews were more formal when I wanted to discuss a range of issues based on a semi-structured interview guide that I had prepared earlier. The only criteria used for these interviews with patients were that they were willing to talk to me and fit enough to be interviewed; and here I relied on advice from the ward staff on the patient’s state of health and lucidity. Semi-structured interviews with patients, as opposed to informal discussions, took place in the treatment room at the end of the wards. This room was separated from the main ward by a grill gate and was about the only private place that could be allocated to me. Nonetheless, interviews were often inadvertently interrupted by the nursing staff, cleaners or other patients who wandered in. Interviews would then be momentarily suspended if possible, before continuing. Semi-structured interviews were conducted at various intervals with selected members of staff, including medical officers, nursing staff and allied personnel, such as occupational therapists and the two social workers, as well as former members of staff. Normally these interviews required careful planning due to medical schedules and outpatient appointments, therefore they were usually tape-recorded and supplemented by extensive note taking during the interview process itself.

      Semi-structured interviews with patients were usually taped with their consent. The open use of a tape recorder in informal situations was eventually seen to be too intrusive for general conversations after I detected that, in particular, members of staff felt uncomfortable and inhibited by the idea. Furthermore, I had the impression that the tape recorder was distracting for patients, as well as intrusive, and tended to curtail spontaneous disclosures. Reliance on an increasingly elastic memory for informal conversations meant the flow of conversation was not interrupted; and a more relaxed and confiding atmosphere could be created. Hastily but discreetly written up notes in shorthand on small notebooks usually took place in secluded corners of the ward following these valuable sessions, as like the patients I was not able to enjoy freedom of movement due to ward ‘lock-up’ procedures.

      Although I had initially hoped to engage a wide range of respondents, in reality some were considerably more responsive than others. Opportunities to talk to both patients and staff were seized more on the basis of luck than design, commensurate with Burgess’ definition of ‘opportunistic sampling’ (1995: 55).

      Amongst the patient population my key informants were nearly all women; male patients tended to shy away from contact or at any rate often seemed less likely to respond to my questions with relevant information. On the face of it this is in keeping with the rapport Ann Oakley discovers in her research activities through the democratisation of the interviewing process, premised on the notion of shared commonalities, of which she writes:

      The women were reacting to my own evident wish for a relatively intimate and non-hierarchical relationship (Oakley, 1984: 47).

      However, I lacked the basic common grounds that Oakley held; she was a British mother, interviewing British mothers. Whereas I was a foreign woman who had never been admitted to a psychiatric hospital and was attempting to develop a rapport with women and men, many of whom had spent years of their lives being processed by the Malaysian psychiatric services. Yet, despite Daphne Patai’s