The psychosocial effects of organized 
violence and torture: a pilot study 
comparing survivors and their 
neighbours in Zimbabwe 
A. P. REELER AND J. MHETURA 
ABSTRACT 
Studies of survivors of organized violence and torture are 
uncommon in the African setting. Studies of the psychosocial 
effects of organized violence and torture are even less common. 
A Zimbabwean study comparing survivors of organized violence 
and torture with their neighbours was carried out in one 
previously war-affected area of Zimbabwe. The findings 
indicated that survivors were more economically and socially 
deprived than their neighbours in many key areas, especially 
the areas of employment, income, food security and housing. 
In addition, survivors showed indications of lower self-esteem 
and belief that they could change their situation. 
Seen in the context of the increasing real poverty in Zimbabwe, 
the findings suggest that survivors of organized violence and 
torture represent a disabled group that may require targeted 
assistance by the State in order to overcome the social adversity 
they experience. The findings also indicate the need to assess 
more carefully the psychosocial as well as the medical 
consequences of organized violence and torture, especially in 
a region where epidemic levels of violence have been experienced 
in recent decades. 
Background Meithe 
OST WOR 
r o
K
n 
O 
Wester 
N survivor 
n populations 
s of violenc
, especiall 
e and tortur 
y o
e
n 
ha 
wa
s bee 
r veteran 
n done
s 
(Helzer et al. 1987, Kulka et al. 1988), or on survivors who are either 
JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 2000 137
refugees or forced immigrants in host countries (Kinzie et al. 1984, 
Allodi 1980). In Africa and other Third World settings, clinical or 
epidemiological studies of the health consequences of organized 
violence are rare, although there is a large literature making comment 
upon such effects, especially the effects upon children (Dawes 1994). 
There are some clinical studies which generally support the 
conclusions of the Western work (Straker 1993) but few epidemiological 
studies. 
Previous work in Zimbabwe indicated that disorders due to 
organized violence were prevalent (Reeler 1996, Reeler and Mupinda 
1996), but this data was derived from clinic referrals rather than from 
an epidemiological study. Epidemiological studies of Mozambican 
refugees resident in Zimbabwe have also indicated that disorders due 
to organized violence are common (Reeler 1995, Reeler and 
Immerman 1994). These studies did not, however, specifically 
examine for torture. The most recent Zimbabwean study indicated 
that disorders due to organized violence and torture were very 
prevalent (Reeler etal. 1999) .constituting about 10% of all morbidity 
presenting to the primary care setting. However, some caution should 
be expressed since the sample in this last study was mostly composed 
of chronic survivors from the war that ended in 1980. 
Although there are increasing numbers of studies on the effects 
of torture most are concerned with the medical and psychological 
effects and there are no detailed studies on the social and economic 
consequences of torture. This is perhaps surprising when so much of 
the published work has been conducted in refugee populations, a 
prima facie disadvantaged group and when the medical and 
psychological studies themselves indicate persistent disability and 
dysfunction in torture survivors. Furthermore, in situations where 
there has been epidemic violence such as Cambodia or Mozambique, 
the socio-economic consequences of organized violence and torture 
(OVT) must take on an even more important meaning in the process 
of reconstruction and rehabilitation. So questions about the long-term 
impact of torture cannot focus on the individual and the medicopsychological 
consequences alone, but must also focus upon familial 
138 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 No. 2 JULY 2000
Reefer etal. I PSYCHOLOGICAL EFFECTS OF VIOLENCE 
and social effects. This point has been made most forcefully by various 
workers (Kordon et al. 1998, Lagos 1996, Reeler 1994). 
In Africa and Zimbabwe the palpable poverty in which torture 
survivors live is a factor that has led many trauma workers to 
speculate about the efficacy of service delivery when set against this 
feature in people's lives. The AMANI Trust, a Zimbabwean nongovernmental 
organization working with survivors of organized 
violence and torture (OVT) has had consistent reservations about 
how good the prognosis for clients could be when their material 
situation was generally poor and when they are continually subjected 
to persistent stresses arising out of this material deprivation.This 
concern was strengthened by the publication of two well-controlled 
studies on psychological disorders by the researchers from the 
Department of Psychiatry at the University of Zimbabwe (Patel et 
al. 1998, Patel et al. 1997). The first study clearly demonstrates the 
relationship between poverty and psychological disorder, especially 
for women, and indicates that financial stress is one of the strongest 
predictors of the presence of psychological disorder. This replicates 
a long-standing observation in Zimbabwe about the prognostic effects 
of social adversity (Reeler 1986; Reeler and Todd 1994). The 1997 
study indicates that a series of factors are strongly associated with 
poor mental health: 
• female gender 
• older age 
• chronic illness 
• number of presenting complaints 
• economic impoverishment 
• infertility 
• recent unemployment 
• disability 
The second study, which examined the outcome of Common Mental 
Disorders (CMD), indicates that there is a poor prognosis generally 
for these disorders with 41% of the sample remaining with clinically 
significant disorder at 12 months (Patel et al. 1998). Two factors 
were most strongly associated with poor prognosis: disability and 
economic deprivation. 
JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 2000 139
In AMANPs own work with the survivors of torture poverty is the 
single most important presenting complaint and forms the bulk of 
discussions in all counselling sessions. This is reflected, too, in many 
aspects of the data that has been collected on clients of AMANI. An 
examination of the case flies of clients indicated that social adversity 
(family, marital, financial and personal problems) was common, as 
well as high unemployment, low levels of educational attainment and 
observable poverty. Since these are factors that are frequently seen 
in the rural populations of Zimbabwe anyhow, it was decided to 
examine whether survivors of OVT were relatively more disadvantaged. 
Method To
H
f
E 
organize 
MAJOR ai 
d violenc 
m was t
e
o 
an 
determin 
d tortur 
e th 
e (OVT 
e socia 
) o 
l an 
n survivor 
d economi 
s livin 
c effect 
g in
s 
communities that had previously experienced epidemic levels of 
violence during the 1970 war of liberation. Since the survivor group 
has high levels of psychological and physical disability it was necessary 
to determine whether these factors resulted in higher levels of social 
adversity than non-disabled persons living in the same communities. 
In order to determine the psychosocial effects on survivors of OVT 
a comprehensive questionnaire was constructed, based on 
information from the 1993 census (CSO 1993), the Catholic 
Commission for Justice and Peace poverty datum study (Mundy 
1995), the government's own poverty assessment survey (Ministry 
ofPublic Service, Labour and Social Welfare 1997), consultations with 
members of the Department of Rural and Urban Planning at the 
University of Zimbabwe and the writers' own personal experience of 
the context. 
The assessment focused upon a series of key areas, namely: 
• health 
• children's health status 
• activities and employment 
• household income activities 
• housing amenities 
140 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 No. 2 JULY 2000
Subjects 
Reeler etal. I PSYCHOLOGICAL EFFECTS OF VIOLENCE 
• agricultural activities 
• cropping 
• gardening 
• use of natural resources 
• communication and transport 
• household consumption 
• perceptions of poverty 
The questionnaire was given as a structured interview to all subjects. 
TTrust 
HE SUBJECT 
, selecte 
S wer 
d s
e
o 
chose 
as to giv 
n fro 
e a
m
s broa 
amongs 
d a representatio 
t the clients o
n
f 
o 
th
f al 
e AMAN 
l clients
I 
and the various areas in which they live as possible. An attempt was 
made to get at least two persons from each area in Mount Darwin. 
There were two sets of respondents: clients who were registered 
with the AMANI Trust and controls, who were not. 
Clients were compared with a neighbour (from among the 
controls) with the intention of comparing torture survivors with 
people who had not been tortured. In practice this was difficult as 
many neighbours had also experienced organized violence, although 
they had not approached AMANI or been referred to AMANI. The 
clients were recruited from the files by systematic random sampling 
by taking every tenth person on the register. The criterion for 
admission amongst the clients was that they must have a severe 
psychological disorder (a score of more than 10 on the SRQ-20) and 
a positive history of impact torture. This resulted in 28 clients being 
selected and they were finally compared with 27 controls. Several of 
the controls, however, had histories of OVT but had not sought 
assistance from or been referred to AMANI: only about 50% had no 
experience of organized violence. In the end we can thus claim only 
that we have compared those who have sought assistance from 
AMANI with those who have not. 
The purpose of the study was explained to all subjects at the outset 
and consent to participate was inferred from their willingness to 
continue with the interview. No subject declined to be interviewed. 
The instrument was written in English but administered in Shona. It 
JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 2000 141
was piloted on several subjects - not included in the final study - and 
the appropriateness of the questions was evaluated. The interviewers 
were two experienced social workers who had extensive experience 
in working in the rural setting as well as of having participated in 
similar types of study. 
All interviews took place at the homes of the subjects. The 
interviews lasted between 45 minutes to an hour on average, but 
could often take considerably longer since the subjects were 
interested in talking in great detail on many of the topics covered by 
the interview. 
Results THER 
control 
E WER 
s havin 
E demographi 
g a highe
c
r 
difference 
percentag
s
e 
i 
o
n
f 
th 
male 
e tw
s (76% 
o samples 
) tha
, wit 
n th
h
e 
clients (54%). The sample age ranged from 25 years to above 60 
years with no differences between the two groups. There were no 
differences in the marital status of the two groups, with the exception 
that divorce was more common in the control group. Controls tended 
to have smaller families than clients (42% versus 25%), which would 
be expected from an older group in which more children had left 
home. 
If understanding written Shona or English can be regarded as 
evidence of literacy then there are some clear differences between 
the two groups. The data suggests that only 57% of the clients are 
literate as compared to 82% among the control group.Since one of 
the long-term consequences of epidemic violence is argued to be 
reduced integration into the community (Reeler 1998), we decided 
to provide some indication of community affiliation. As can be seen 
from Figure 1 below there were marked differences between the two 
groups in their community affiliation in respect of the formal and 
informal groups to which the sample belonged. 
Health is an area that is extremely sensitive to socio-economic 
factors and, according to the illnesses reported by the respondents 
in the previous month, more controls than clients reported suffering 
from malaria and diarrhoea (see Figure 2). Clients reported suffering 
more from pains, aches and depression than controls. Pains and aches 
accounts for 28.6% among the client group as compared to 0% among 
142 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 No. 2 JULY 2000
Reeler et al. I PSYCHOLOGICAL EFFECTS OF VIOLENCE 
FIOURE1: TYPES OF COMMUNITY AFFILIATION 
25% 
s 
the control group and this should have been expected, given the 
history of impact torture in the client group. 
FIGURE 2: ILLNESS IN PAST MONTH 
50% - 
Time lost at work due to illness is another measure. More clients 
reported having lost a week than the controls (39% versus 24%), 
but more controls reported longer periods of work lost than the clients 
(29% versus 18%). This ties in with the types of illnesses reported 
above, where controls reported serious physical illnesses more 
frequently than the clients. 
Access to health facilities can also determine health status. Here it 
can be seen from Figure 3 that there were no differences in the two 
JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 2000 143
60% 
40% 
20% 
Children 
FIGURE 3: HEALTH CENTRE VISITED 
Karanda Epworth Nyamahobogo None 
groups in the use of the available health facilities. This was expected 
since both groups came from the same geographical area. 
Since most rural people are regarded as poor, many receive free 
treatment under government's goal of providing health services to 
all, but the policy of cost recovery has increasingly lead to this group 
incurring charges for health care. Most report receiving free 
treatment (82% in both groups), but there is slightly higher 
percentage of clients who report paying for health services (4% versus 
0%) and this probably reflects the chronic disorders and injuries seen 
in this group. It also reflects a greater economic burden on clients as 
compared with their neighbours in the control group. 
Tchildre 
HE STUD
n
Y 
i 
als 
n th
o
e 
attempte 
respectiv
d
e 
t 
groups 
o examin 
, sinc 
e th 
e povert 
e healt
y
h 
an 
statu 
d children' 
s of the
s 
health are very directly related. For those who had young children at 
home, we examined whether clients and controls were aware of health 
issues. There were small differences in the views of the two groups 
towards vaccinations: clients were less likely to have had their children 
vaccinated than the controls (64% versus 71%). There are several 
plausible explanations here. Firstly, this difference reflects a general 
apathy amongst the clients and, secondly, it may equally reflect a 
refusal to vaccinate, given that the Apostolic Churches have a strong 
following in the district. Since we did not directly examine the actual 
JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 No. 2 JULY 2000 144
Reeler etal. I PSYCHOLOGICAL EFFECTS OF VIOLENCE 
religious persuasion of the sample, we cannot answer the question, 
but clients were more likely to belong to a religious group. 
Schooling is another area in which the effects of poverty can be 
seen. As can be seen from Figure 4 clients are more likely to have 
young families and this ties in with the reported ages of the 
respondents, where the controls were a slightly older group than 
the clients. As they were therefore more likely to have school-age 
children they faced the consequent economic burden of school fees. 
0% 
FIGURE At NUMBER OF SCHOOL-GOING CHILDREN PER 
HOUSEHOLD 
1-3 
As regards providing for children in school, clear differences were 
found, as can be seen in Figure 5. School-going children must posses 
a minimum of basic clothing such as one pair of shoes, three pairs of 
None 
FIGURE 5: MINIMUM AND ABOVE ABILITY TO PROVIDE SCHOOL 
UNFORMS FOR CHILDREN: CLIENTS AND CONTROLS 
60% 
40% 
20% 
0% 
Boys 
JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 2000 
I • Client 
| | D Control 
Infants Girls 
145
socks, two shirts and two pairs of shorts. Controls reported being 
less able to provide for their male school-going children. 
Although there is generally a similarity between the two groups, 
the controls report being less able to provide for their boy's needs. 
This may reflect a variety of different reasons; for example, the clients 
may be more motivated to send their boys to school than the controls 
and, as it has been recently observed, there has been an increasing 
trend for families to discriminate in favour of boys attending school 
against girls. 
Activities and employment 
TH
o
E
f th 
STUD 
e sampl 
Y ALSO
e 
mad 
in orde 
e an attemp 
r to determin 
t to examin 
e whethe 
e the economi 
r the disabilit 
c activitie 
y o
s
f 
torture survivors affects their livelihood. Most of the respondents 
were small-scale farmers. No differences were observed between 
the two groups in the types of activities reported. It was also of interest 
to see whether the sample had had previous employment in the formal 
sector, since this could determine the capital base of the families 
assessed. Only 14% of the client group had worked after independence, 
whereas 41% of the control had taken up some form of 
employment. This suggests that there was greater unemployment 
in the client group and hence less access to capital. 
FIGURE 6: INCOME RANGES FROM PAST WEEK ACTIVITY 
ii 
80% 
60% 
40% \ 
20% 
0% • «n • Poultry 
|B Client 
I 
146 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 No. 2 JULY 2000 
" j • Control]
FIGURE 7: HOUSEHOLD INCOME EARNING ACTIVITIES 
<$100 $100-$500 >$501 
The general trend is clearly towards controls earning more income 
than clients. As regards the average income earned from nonagricultural 
sources, there was a trend for controls to earn more on 
average (Z$534 as opposed to Z$168) and the controls also earn in a 
greater range than the clients (Z$1114 as opposed to Z$522). 
More clients than controls obtained some form of credit for their 
income earning activities (47% versus 23%). These credits received 
were mainly for agricultural purpose and varied from seed to fertiliser 
and pesticides. Loans can be a mixed blessing and it was unclear 
whether this difference represented the ability of the controls to selffinance 
their activities or merely their inability or reluctance to use 
credit facilities. 
The sources of income of those who did not take up employment 
after 1980 were examined. There were marked differences between 
the two groups in terms of the generation of income, with controls 
more frequently reporting than clients earning some kind of income 
(65% versus 25%). However, the ranges of income earned by the 
two groups was largely similar where they reported earning income, 
as can be seen from Figure 6. 
JOURNAL OF SOCIAL DEVELOPMENT M AFRICA VOL 15 NO. 2 JULY 2000 
Reeter etal. I PSYCHOLOGICAL EFFECTS OF VIOLENCE 
As can be seen from Figure 7 there was also a trend for clients to 
report more frequently no household income earning activities. 
147
Household income activities 
r MHE MAIN HOUSEHOLD activity was market gardening, with clients 
X reporting market gardening activities about half as frequently 
The additional question was whether the sample reported any 
constraints on their agricultural activities. Clients reported drought 
and lack of money as the major constraints. The latter factor was 
also identified by the controls as well as lack of fencing. 
;' a Control 
FIGURE 8: CONSTRAINTS ON ECONOMIC ACTIVITIES 
75% , - - - . - - 
Housing 
None Drought Poor Need Need CHd/sick 
fence medicine money 
Ttraditional 
HE TYPES O
,
F 
moder 
dwellin 
n an 
g wer 
d mixed 
e broke 
. Som 
n int 
e
o 
marke 
three
d 
observabl 
difference 
e types 
s were
; 
observed. Among the clients most people (54%) live in traditional 
homes, while most of the controls (77%) live in mixed homes 
comprising both brick and traditional structures. This probably can 
be taken together with the greater likelihood of the controls having 
had jobs in the formal sector and thus with the capital to build some 
brick structures. 
The type of roof is another factor that determines the durability 
of the dwelling. Most clients (64%) have thatched roofs as compared 
to only 53% of the controls. However, the differences observed 
between the two groups in the types of roofing was not great. The 
floor types were divided into two: those made of mud and those made 
of concrete. Little difference was seen between the two groups. The 
quality of the walls was divided into three categories: unbumed bricks, 
148 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 No. 2 JULY 2000
Reeler ef a/. / PSYCHOLOGICAL EFFECTS OF VIOLENCE 
pole and dagga and burned bricks. This is again another factor that 
determines the durability of the structures. Generally more people 
in both groups have homes made of burnt bricks and it can be 
concluded that the controls generally have better homesteads than 
the clients. 
FIGURE 9: TYPE OF WALLS 
100%, • C l i e n t 
75% • Control 
50% 
25% 
0% 
Amenities 
* 
An attempt was made to estimate the condition of the houses. Here 
some differences were observed. There was a trend for- clients to 
have more "worn out" or "used and rundown" housing than the 
controls (42% versus 24%) but there were no differences in the 
number of "new and well-kept" houses. In general there was a 
consistent trend for clients to live in poorer housing than their 
neighbours in the control group. 
TH
an
E
d 
QUALIT 
the 199 
Y o
2 f 
drinkin 
census
g 
indicate 
water i
d s 
a 
tha 
significan 
t few peopl 
t facto 
e in
r 
Moun 
in goo
t
d 
Darwi 
healt
n
h 
District had access to protected water. As can be seen from Figure 
10 below there is a trend for clients to have better access to protected 
water than the controls. This ties in with the reports by the controls 
of more diarrhoeal disease than the clients. 
Boreholes were not very far away from people's homes, so that 
86% of the clients and 65% of the controls have boreholes within a 
JOURNAL Of SOCIAL DEVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 2000 149
FIGURE 10: SOURCES OF DRINKING WATER 
River 
Unprotected well 
Protected well 
Tap water 
than 2.5 kilometres to collect firewood, but the controls as a whole 
generally reported having to travel greater distances than the clients. 
ncontroll 
L •Client J 
2.5-5 km 5-10 km 
• Control 
• Client 
Borehole 
distance of 0 to 1 km. from their home. However, controls generally 
reported having greater distances to walk for water. 
Toilet facilities are yet another indicator of poverty and access to 
protected toilet facilities has well-known health consequences. Pew 
in either group have access to Blair toilets' but a greater number of 
clients have this form of toilet (29% versus 18%). 
Firewood, used for cooking rather than lighting, is one of the most 
important resources needed by rural people and the district shows 
obvious signs of severe deforestation. As can be seen from Figure 
11, there were some differences found between the two groups, with 
slightly more clients reporting that they had to travel distance greater 
FIGURE 11: DISTANCE TO FIREWOOD COLLECTION POINT 
80% 
60S 
40% 
20% 
0% 
1 Fly-protected pit latrines with soakaways 
150 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 No. 2 JULY 2000
Cropping 
Reeler etal. I PSYCHOLOGICAL EFFECTS OF VIOLENCE 
AGRICULTURA 
activities i
L
n 
ACTIVITIE 
that the
S 
forme 
are distinguishe 
r deals mostl 
d fro 
y wit 
m househol 
h the earnin 
d incom 
g o
e
f 
income from the production of cash crops and livestock. The 
availability of adequate land, both in size and quality is therefore 
significant and here we note that Mount Darwin lies in an area with 
poor soils and low rainfall. Of the clients, 61% as opposed to 35% of 
the controls, report having adequate land. More people (65%) in the 
control group felt that the land was not adequate to the needs of 
their families. Rural agriculture requires such basic equipment such 
as hoes, axes, draught power and ploughs, in addition to adequate 
land. As can be seen from Figure 12, differences were reported by 
the two groups in the availability of various farming implements. 
Neither group reported much access to machines such as shelters 
and there were no great differences in the availability of hoes, axes 
and cultivators. Clients reported less availability of draught power 
for ploughing and owning fewer cattle, but greater access to ploughs, 
so this finding is difficult to interpret. 
FIGURE 12: AGRICULTURAL MACHINERY AND IMPLEMENTS 
AVAILABLE 
C ultivators Ploughs Agricultural Draught 
machines power 
I C I i e n t Y e s • C o n t r o l Y e s 
The ownership of cattle and the confiscations and losses' incurred 
during the liberation war are a frequent topic of conversation amongst 
AMANI's clients. Ownership of cattle ensures the availability of animal 
draught power, a vital component of rural agriculture. Clients 
JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 2000 151
reported owning fewer cattle than the controls: about 39% of the 
clients own less than 5 cattle as compared to 17% of the controls, but 
59% of the controls reported owning more than 5 head of cattle as 
opposed to only 39% of the clients. A significant proportion of both 
groups report owning no cattle. 
Adequacy of pasture is measured by its sufficiency to the whole 
community. Pastures are communal. 56.5% of clients reported having 
sufficient pastures but some clients felt this was inadequate. Since 
maize is the major staple food it was expected that all respondents 
grew it but only 86% of the clients grew maize as compared to 100% 
of the controls. This will have obvious implications for food security. 
This finding is given added depth by the results of the yields from 
maize growth. As can be seen from Figure 13, there was little 
difference between the two groups in the numbers who harvested 
more than 30 bags of maize, but in general controls harvested more 
than clients and a significant number of clients reported harvesting 
nothing. 
FIGURE 13: MAIZE HARVESTED 
50%,- 
25% 
>30 10-30 <10 none 
bags bags bags 
This is further to be added to the results comparing the use of the 
maize harvested. Most clients (95%) use their maize for consumption 
as compared to only 75% of the controls. Controls thus grow more 
and receive greater income than the clients. 
Cotton is a crop that is increasingly grown as an income earner 
and there is little difference between clients and controls when it 
comes to growing cotton but there is a difference in the amount of 
152 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 No. 2 JULY 2000
FIGURE 14: INCOME FROM TOBACCO IN Z$ 
CONTROL CLIENT • 1100 
• 2300 
• 2500 
• 3500 
• 4000 
Most of the money earned from crops grown is controlled by the 
husband (67%) within the client group as compared to only 46% in 
the controls. This often has implications for family welfare since money 
controlled by women is directly available for the family, whilst money 
controlled by men may not be: this can mean families going hungry 
even when there is an income. 
JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 2000 
Reefer et al. I PSYCHOLOGICAL EFFECTS OF VIOLENCE 
cotton harvested. Fewer clients reported large yields (9% versus 
25%), but more clients had average yields than the controls (54% 
versus 25%). Since on average controls grow more cotton this group 
would be expected to make more money than the clients. Significantly 
more controls that clients (67% versus 44%) earned in excess of 
Z$2000.00 from the sale of cotton. Groundnuts can be both a cash 
a*id consumption crops and both groups reported growing these crops 
with similar frequency. Tobacco is the highest income-earning crop 
grown in Zimbabwe. Both groups reported growing tobacco but it 
was grown rather more frequently by the members of the control 
group (100% versus 83%). 
The clients reported making more money from tobacco than the 
controls, especially those with higher levels of income. This suggests 
that clients have moved away from maize into tobacco production, 
which may improve their income but does not strengthen food 
security. Additionally, tobacco is a more capital-intensive crop, for 
which credit was probably required and also more vulnerable to 
drought and pests than either maize or cotton. 
• 200 
• 1000 
• 2000 
153
75% 
50% 
25% 
0% 
Gardens 
T TEGETABLE GARDENS in the rural areas play a pivotal role in the 
V provision of vegetables for both consumption and sale. Clients 
are slightly more likely to have a vegetable garden than the controls: 
86% of clients as compared to 77% of the controls have a member of 
the family with a garden. Successful vegetable gardening is dependent 
on water availability and this was generally similar in both groups. 
However, the nature of the water source must be considered: here 
we see that both groups have access to both rivers and other sources, 
with the clients reporting greater access to sources other than rivers. 
• Clients 
• Controls 
FIGURE 16: PERIOD FOOD STOCKS WILL LAST 
80 
60 
40 
20 
154 
FIGURE 15: WATER SOURCE. 
Control 
CM » 
2 -- £ 
JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 No. 2 JULY 2000
Reeler et al. I PSYCHOLOGICAL EFFECTS OF VIOLENCE 
Another constraint on successful vegetable production is protection 
from grazing by animals, particularly cows and goats. Controls stated 
they had greater access to permanent forms of protection than clients 
(32% versus 8%) but also we should note that they had also reported 
lack of access to fencing as a significant constraint on their activities 
so this requires further explanation. The reasons for not having a 
garden in the two groups are not widely different. The main reasons 
are disability, no land and no water. 
All these activities impact on food security. Food security can be 
obtained in two ways. Firstly, by specifically growing food crops for 
storage and later consumption and supplementing this with 
vegetables and protein from livestock. Secondly, cash crops can be 
grown for income and the income used to purchase food later. A mixed 
strategy may also be adopted where food crops are grown and 
supplemented with food purchased with income earned from the 
production and sale of cash crops. Figure 16 reports the estimates 
by the sample on the food stocks available to them during the year.As 
can be seen there is a trend for controls to report more stocks than 
clients. No controls report the absence of food stocks and more than 
25% report stocks that will last well into the following year. This 
suggests that both groups may experience food shortage and require 
supplementary feeding. However, more clients than controls report 
having received drought relief more than once (8% versus 0%), but 
also more clients have never received any form of drought relief (7% 
versus 0%). 
Use of natural resources 
APAR 
involve 
T FRO 
d in
M 
horticultur 
agricultura 
e o
l activities 
r use loca
,
l 
rura 
indigenou 
l peopl 
s resource 
e may becom 
s from
e 
forests and most people have grown fruit trees or use local indigenous 
trees. However, deforestation is a significant problem in Mount 
Darwin District. This is due in part to population growth, but also due 
to the massive tree cutting that took place during the liberation war. 
Very few clients (29%) have woodlots compared with the controls 
(59%), despite the fact that many more clients than controls cite 
distance as a problem for collecting fuel. 
JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 2000 155
People also use other natural resources from the local environment 
such as grasses for various use, plants for medicine, etc. The utilization 
of natural resources is very high in both groups. Some of these items 
are used for income generation but with very small profits: for 
example, thatching grass is sold and poles for construction are cut 
from indigenous trees. 
Communication and transport Ti
H
s
E 
th 
MOS 
e radi 
T COMMO 
o althoug 
N mediu 
h th
m
e 
fo 
clien 
r receivin 
t group
g 
i 
informatio 
s less likel
n
y 
fo 
to
r 
hav 
both
e 
group 
acces
s
s 
to other forms of information and even less access than the control 
group to the radio. The major mode of transporting field produce is 
animal-drawn (89% and 82% in each group) The general use of other 
modes of transport such as bicycles is minimal. 
Very few members of either group own animals for transport and 
both groups rely on borrowing animals to move produce. For personal 
travel the bus is the most common form of transport. 
Another way in which information may be disseminated in rural 
communities is through organizations working in the communities. 
Figure 17 shows the frequency with which the sample had been visited 
in the past three months by the most common community-based 
agencies in the district. As can be seen, clients were much more likely 
to have received some form of visit than the controls. It is noteworthy, 
however, that the controls were much more likely to have received a 
visit form Agritex, which may be a factor in fostering better 
agricultural production. 
Household consumption Hpurchas 
OUSEHOL
e
D 
o 
CONSUMPTIO 
f items. The
N 
majo 
is clearl 
r consumabl 
y relate 
e item 
d to incom 
s used in
e 
th 
fo
e
r 
pas 
the
t 
month were soap, maize-meal, salt, sugar, paraffin and vegetables. 
We attempted to estimate the use of these. There were few 
differences in the use of soap during the past month in the two groups. 
Maize is the basic food consumed by most Zimbabwean families and 
there were small differences in the two groups, with clients reporting 
higher consumption of maize than controls (79% versus 65% 
consumed more than 5 buckets per month). This may be related to 
156 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 No. 2 JULY 2000
FIGURE 17: FREQUENCY OF VISITS BY ORGANIZATIONS 
7 5 % 
•IC lie nt 
• C o n tr o I 
Reeler etal. I PSYCHOLOGICAL EFFECTS OF VIOLENCE 
157 
1 1 1 • • • 
family size, since the clients had slightly larger families than the 
controls. Similarly, there were small differences in the use of salt with 
the controls reporting using more salt than the clients. This may 
reflect the greater spending power of the control group 
Perceptions of poverty 
ALTHOUG 
it also necessar 
H IT is possibl 
y to
e 
asses 
to determin 
s whethe 
e level 
r thes 
s o
e
f 
objectiv 
poverty objectively 
e measures
, 
are perceived in the same manner by the population under study, 
Accordingly we included some questions designed to assess whether 
the sample perceived themselves as poor, as well as their perceptions 
of their neighbours' poverty. As can be seen from Figure 18 below, 
there were differences between the two groups in what they saw as 
constituting poverty. The controls identified lack of land, food 
shortage and the inability to pay school fees as the dominant 
characteristics of poverty whilst the clients saw poor accommodation 
and the lack of draught power as the most salient features. Clients 
also identified a wider range of factors as being indicative of poverty. 
The sample was also asked what they thought were the main causes 
of poverty among victims of torture and organized violence. As can 
be seen from Figure 19 there were differences between the two 
JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 2000
FIGURE 18: CHARACTERISTICS OF A POOR HOUSEHOLD 
Poor health 
Lack draught powe 
School fees 
Poor housing ( 
Food shortage 
Land shortage 
15% 0% 
groups. Clients saw war destruction as a more frequent cause than 
the controls, whilst the controls saw disability and lack of draught 
power as more significant than the clients. 
45% 30% 
FIGURE 19: OPINION ON MAIN CAUSES OF POVERTY 
!• Draught power j 
| • War destruction 
\ CI Disability ] 
Inner circle = Clients 
Outer circle = Controls 
There were few differences between the two groups in what they 
saw as the best solutions for rehabilitating the victims of torture and 
organized violence. Similarly, there were few differences between 
the two groups in their views about whether their neighbour has 
opportunities to escape from poverty. The clients were more 
JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 No. 2 JULY 2000 158
Reeler etal. I PSYCHOLOGICAL EFFECTS OF VIOLENCE 
100% 
FIGURE 20: ABILITY OF NEIGHBOUR TO IMPROVE 
• Client 
i incontrol! 
75% 
50% 
25% 
0% 
pessimistic than the controls about the prospects of escaping from 
poverty, as can be seen from Figure 20. This suggests that the 
relatively successful experience of the controls made them have more 
self-reliant than the clients. The finding that the greater the 
experience of success, the greater the belief in subsequent success, 
has had empirical validation in recent years (Bandura.1989). 
FIGURE 21: POVERTY ALLEVIATION STRATEGIES 
D Financial assistance D Income generating 
• Material assistance • Resettlement 
Inner circle = Clients 
Outer circle = Controls 
This view is supplemented by the findings on the strategies that the 
two groups see as relevant to escaping poverty. As can be seen from 
Figure 21, clients are more dependent in requiring money to escape 
poverty, whilst the controls, with a greater belief in their own abilities 
and resources, make use of material inputs to create their own 
opportunities. 
JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 2000 159
Clients reported more aches, pains and depression than the controls. 
This suggests that one major difference between the two groups lies 
in the ability of the controls to cope with any disability or disorder 
due to OVT. It may also reflect the fact that controls did not experience 
significant OVT. In addition these are symptoms that would be 
expected from the selection criteria: severe psychological disorder 
and impact torture. 
There were no differences reported between the two groups in 
use of or access to the various health facilities, nor were there 
differences in the amount of time lost due to illness. There was slight 
trend for clients to spend more money on health care, but this was 
small. The controls appeared to have a greater frequency of 
observable disability than the clients but they complained significantly 
less about such disability than the clients. The additional presence of 
psychological disorder in the clients may here be significant, but this 
can only be a tentative conclusion in the absence of screening for 
psychological disorder in the controls. 
There were no real differences between the two groups in access 
to and use of health facilities, but some differences in the kinds of 
illnesses and the days lost due to illness. However, we did not sample 
the frequency of consultation, which would have been a useful 
indicator and has implications for the economic costs suffered due to 
illness. We have one small indication in the amount expended on health 
care indicated by a related Zimbabwean study (Patel et al. 1997). 
This study indicated that persons suffering from Common Mental 
Disorders (CMD), who will include a significant number of survivors, 
spent significantly more on medical treatment than other categories 
of patients. 
In general both groups report high levels of poverty and inability 
to provide adequately for their children, irrespective of whether the 
child is of school-going age or not. Neither group appears to reach 
the minimal standard of provision for school-going children. Against 
these indices of poverty it is thus encouraging to see the high levels 
of vaccination reported in the children. 
Most of the subjects are subsistence farmers and derive their 
income mainly from agricultural activities, especially form the selling 
160 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 No. 2 JULY 3000
Reeler et al. I PSYCHOLOGICAL EFFECTS OF VIOLENCE 
of crops and animals. This is unsurprising. However, the controls were 
more likely to undertake some form of employment and more likely 
to have earnings coming from sources other than agricultural 
activities. The controls reported more frequently that they had 
recently received income and were slightly more likely to have 
earnings in the higher ranges than the clients. These findings suggest 
greater economic security for the controls. 
Although the activities described are regarded as income 
generating, very few in either group derive much income from these 
activities. The main activity described for both groups was market 
gardening, but this activity does not generally yield much income for 
either group. However controls generally earn more than twice as 
much as clients. 
Market gardening is an activity that requires natural resources 
such as land and water as well as a certain degree of both capital and 
labour investment if it is to be successful. The clients received more 
credit than controls. This can be a problem if it increases indebtedness 
and it is worth noting here that clients reported drought as a 
constraint more frequently than controls. 
There are basically three types of housing: traditional, modern and 
mixed. In general traditional houses are synonymous with poverty 
whilst modern houses are more indicative of a degree of economic 
well-being: this is not an immutable law, but the generalization holds 
to a large degree. As was seen above there was a tendency for clients 
to live in poorer quality housing and also for these houses to be in a 
poorer condition than those of the controls. 
As regards the amenities - clean water, toilets and fuel - there 
were some interesting differences between the two groups. Clients 
had more access to clean water which was also generally closer to 
their homes. This ties in with the reports by controls of more 
frequent waterborne illnesses. This also accords with differences 
between the two groups in the type of toilet facilities available, with 
clients more likely to have a Blair toilet than controls. Clients 
reported having to travel greater distances on the whole to fuel wood 
sources than controls and they use fuel wood more frequently than 
cow dung. A range of crops is grown - maize, tobacco, cotton and 
JOURNAL OF SOCIAL DEVELOPVENT IN AFRICA VOL 15 NO. 2 JULY 2000 161
nuts - and some of these (maize and nuts) are grown for both sale 
and consumption, whilst the others (tobacco and cotton) are wholly 
grown for sale. The clients generally have made sturdy attempts to 
generate income but these seem to be at the expense of food security. 
Tobacco apart, clients seem less successful at income generation than 
the controls. Taken together with the earlier finding that clients were 
less likely to be involved - and less successfully involved too - in 
market gardening, this suggests a group that is vulnerable to food 
deficiencies, especially in the years of drought which are not 
uncommon in the region. Certainly the clients report having to make 
application for drought relief more frequently than the controls. 
As regards gardening, there can be very little success without 
access to permanent water such as boreholes. Most rivers, dams and 
wells are seasonal, which makes them unreliable for market gardening 
all year round. The problem of lack of water is compounded by the 
damage done by grazing animals during the dry season and here the 
cost of fencing is significant. Most fencing of gardens is done by tree 
or bush branches cut from local trees, usually acacias. This 
exacerbates deforestation. 
The communal mode of transport is based on two systems: animaldrawn 
and buses. Animal transport is critical for most domestic tasks, 
while the bus is crucial for all travel over 5 to 10 kilometres. Both 
groups are dependent on borrowing animal-drawn transport and very 
few own their own. 
Both groups rely upon the radio for information and very few have 
any access to newspapers or the television. Clients have less access 
to the radio, which is consonant with their greater poverty. An 
additional source of information is through community groups such 
as agencies: clients have much greater access to such groups. 
However the majority of these contacts are with agencies dealing 
with disabled and disadvantaged persons. The controls have much 
greater frequency of contact with Agritex, which may be a factor in 
this group's relatively better agricultural performance. 
There were differences in the spending power between the two 
groups, with the controls reporting greater spending power than 
clients. This is consonant with the earlier findings of greater income 
162 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 No. 2 JULY 2000
Conclusions 
Reeler etal. I PSYCHOLOGICAL EFFECTS OF VIOLENCE 
earning by the controls. The groups were different in their views on 
the causes of poverty. The clients' reasons were closely related to 
OVT - war destruction, confiscation of cattle (lack of draught power) 
and poor accommodation - while the controls put forward more 
traditional reasons such as land shortage, food shortage and the 
inability to pay school fees. There were also marked differences in 
their attributions of the ways out of poverty. The controls gave 
reasons that were much more consistent with greater self-reliance 
than the clients. This may be an important difference, leading to a 
poorer performance in life, which is one of Bandura's most central 
findings (Bandura 1989), as well as reflecting one of the more 
common effects of OVT, that is, a diminished belief in one's one's own 
ability to change things (Basoglu et al. 1996) 
APAR 
psychosocia 
T FROM studie 
l consequence 
s on refuge 
s o
e
f 
populations 
OVT in Afric 
, informatio 
a is scanty
n
. 
abou 
This
t 
i 
th 
s a
e 
new field and still bedevilled by the hostility and suspicion of many 
governments towards examining the consequences of gross human 
rights violations. Increasing numbers of studies have examined the 
medical and psychological consequences of OVT, but it is necessary 
to understand the psychosocial consequences, especially in settings 
where the violence has been of epidemic proportions. There remains 
a pressing need to understand the broader consequences of OVT 
and, as many authorities stress, the medium and long-term 
consequences may be as devastating of the short-term medical and 
psychological consequences (Danieli 1998, Kordonei al. 1998, Lagos 
1996). The present investigation, despite its limitations, provides the 
beginning of an empirical understanding of these issues. 
limitations of the study BEFOR 
the limitation 
E DISCUSSIN
s
G 
an 
th
d
e 
th 
finding 
e strength 
s we shoul 
s of th 
d mak 
e study 
e brie 
. Firstly 
f commen 
, thi
t
s 
upo 
wa
n
s 
very small sample drawn from a single district, so it can be argued 
that it is hard to make generalizations based on these findings. 
However, the sample interviewed here corresponds to the major 
sample of clients seen previously, was chosen by random sampling 
JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 2000 163
and the criteria used for selecting were valid; so it is also the case 
that the findings are robust. 
Secondly, there was a degree of confounding due to the fact that 
many of the controls also reported a history of OVT. This is not 
surprising in view of the scale of the violence in the 1970s. Therefore 
the comparison is between persons who seek help and those who do 
not and probably indicates a difference between those who can cope 
and those who do not. Several of the findings support this view: for 
example, the controls have fewer health complaints related to pain 
or depression, spend less money on health care and have greater 
belief in their own efficacy. Against possible confounding is the clear 
demonstration that the two groups live in the same environment have 
access to all the same social and economic facilities and have been 
subject to all the same stresses since 1980. 
Thirdly, the conclusions drawn have been inferential rather than 
statistical and thus may be unwarranted in certain respects. However, 
we point out against this that the magnitude of the some of the 
differences found gives us considerable confidence that the 
differences observed were real. 
Major findings 
WE
above 
CAN
, 
NO 
we
W 
hav 
briefl 
e draw 
y outlin 
n thes 
e the
e 
majo 
conclusion 
r finding
s
s 
unde 
of the
r 
study 
each
. 
heading 
As noted
. 
Firstly, there were a number of differences in social and economic 
factors between the survivors and their neighbours. Among the 
survivors was found: 
• Lower literacy and higher unemployment levels; 
• They had less access to information; 
• They spent more money on health care, earned less income in the 
past week and less earnings in the past year; 
• They had a lower household expenditure and were more 
dependent upon credit (and thus potentially more vulnerable to 
indebtedness); 
• Their housing was poorer (both structurally and as regards its state 
of repair); 
• They tended to travel further for fuel wood, grow less maize, cotton 
and tobacco, were less likely to have fruit trees or woodlots and 
164 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 No. 2 JULY 2000
Reeler et al. I PSYCHOLOGICAL EFFECTS OF VIOLENCE 
made less use of natural resources: consequently they had less 
food security (measured by the number of months during which 
food was available); 
• They had more frequent recourse to drought relief and were more 
likely to use charity or social welfare. 
Secondly, the survivors showed many signs of having lower selfesteem 
and greater apathy than their neighbours. They were: 
• more likely to see war as a reason for poverty; 
• less optimistic that the situation can be changed and 
• more dependent on outside help (believing they need money help 
as opposed to empowerment help). 
The first group of differences represent real and substantial 
differences in the social and economic well-being of the two groups. 
The survivors (clients) are markedly less well-off than their 
neighbours in many areas and it seems fair to conclude that survivors 
experience greater social adversity than other groups in the same 
community. This is probably not surprising and would be found to be 
true of other disabled populations. However, it does mean that 
survivors are more vulnerable to ongoing stress that will, in turn, 
exacerbate their medical and psychological problems. It is noteworthy 
that this is exactly the interpretation that is given by the survivors 
themselves and it was indeed their preoccupation with the practical 
problems of their lives that originally alerted us to the significance of 
social adversity. As several commentators have observed, the problem 
for survivors of OVT is not post-traumatic stress but ongoing stress 
(Straker 1987, Lopez and Marcelino 1995). 
The second group of findings speaks to the psychological 
consequences of OVT and social adversity. Survivors have low selfefficacy 
and this is due in part to the original violence they 
experienced and in part to their failure to overcome social adversity. 
Above we rather commented that nothing breeds success like success, 
but this is a truism with a powerful application here. It is endlessly 
demonstrated by studies on individuals that OVT creates a feeling of 
powerlessness and many commentators point out that this is 
replicated in the social and political arena (Kordon et al. 1998, Lagos 
1996). Our findings speak to the heart of this problem: survivors are 
JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 200O 165
traumatized into beliefs of powerlessness; they perform less well in 
the many tasks of life and their failure compounds and reinforces the 
lack of self-efficacy. It takes little imagination to see how this then 
translates into community, social and political apathy and provides 
severe problems for the development of rural areas. This point has 
been made again and again by refugee workers and community 
workers in areas that have experienced epidemic violence. 
This must all be viewed within the context in which these survivors 
are living. In real terms the supporting facilities - health and social 
welfare - around them are eroding at an alarming rate. The problems 
being experienced by the health delivery system have been 
highlighted by the government's own Poverty Assessment Study 
Survey (Ministry of Public Service, Labour and Social Welfare 1997) 
and this study, which is now completely accepted by many authorities, 
estimates that more than 45% of Zimbabweans experienced regular 
food shortage - below the food poverty line - and moreover, that 
only 26% of Zimbabweans are "not poor" in economic terms - 74% 
are classified as poor at the time of writing as compared to 62% in 
1995. If survivors are below these thresholds, this is indeed alarming 
and suggests that some pro-active measures will be necessary to 
aEeviate the consequences of disability and poverty in survivors. It is 
as well to remember that this is also a population at risk for HIV and 
AIDS like all other population groupings in Zimbabwe and that these 
are families that must develop the parents and workers of the future. 
REFERENCES 
AMANI 1997 (a). Survivors of Torture in Mount Darwin District, 
Mashonaland Central Province: Overview of Report and 
Recommendations. Legal Forum 9, 49-60 
AMANI Trust 1997 (b). Report on Psychological Disorders in 
Clinics and Hospitals in Mount Darwin District, Mashonaland 
Central Province. Harare 
AMANI Trust. 1995. The Chiweshe Nurse-Counsellor Programme: 
Resource Manual revised. Harare 
AMANI Trust. 1995. Assessment of the Consequences of Tbrture 
and Organized Violence: A Manual for Field Workers. Harare 
166 JOURNAL OF SOCIAL DCVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 2000
Reeler etal. I PSYCHOLOGICAL EFFECTS OF VIOLENCE 
Amnesty International. 1984. Against Torture. London, Amnesty 
International Publications 
Arcel, L.T. 1995. PsychoSocial Help to War Victims: Women 
Refugees and their Families. Copenhagen, IRCT 
Bandura. A. 1989. Perceived self-efficacy in the exercise of personal 
agency. The Psychologist 2, 411^25 
Basoglu, M. et. al., 1996. Appraisal of self, social environment and 
state authority as a possible mediator of post-traumatic stress disorder 
in tortured political activists. Journal of Abnormal Psychology 105, 
232-236 
Catholic Commission for Justice and Peace 1975. The Man in the 
Middle. Salisbury 
Catholic Commission for Justice and Peace 1976. Civil War in 
Rhodesia, Salisbury 
Central Statistical Office 1992. Census Report. Harare 
Central Statistical Office 1993. Census Report. Harare 
Combat Poverty Agency 1989. Pictures of Poverty, Twelve 
Accounts of Low Income Dublin. Angus Press 
Danieli, D. 1998. International Handbook of Multi-generational 
Legacies of Trauma. New York, Plenum Press 
Kordon, D., L. Edelman, D. Lagos and D. Kersner, 1998. Argentina: 
Psychosocial and clinical consequences of political repression and 
impunity. Torture 8, 43-47 
Lagos, D. 1996.Argentina:Psydioso(Mard(Jracalcorfiequencesofpditical 
repression and impunity in the medium term. Torture 6,13-15 
Lopez, J. and E. Marcelino, 1995. Torture Survivors and 
Caregivers: Proceedings of the International Workshop on 
Therapy and Research Issues. Philippines: Center For Integrative 
and Development Studies 
Ministry of Public Service, Labour and Social Welfare 1997. Poverty 
Assessment Study Survey: Main Report. Social Dimension Fund 
SDF. Harare, Government of Zimbabwe 
Mwanza, L.K. 1995. Participation of non-governmental organizations 
in social development process in Africa: Implications. Journal of 
Social Development in Africa 10 
Mundy, V. 1995. The Urban Poverty Datum Line in Zimbabwe. 
Harare, Catholic Commission for Justice and Peace 
JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 2000 167
Patel, V., C. Todd, M. Winston et al., 1997. Common mental disorders 
in primary care in Harare, Zimbabwe: Associations and risk factors. 
British Journal of Psychiatry 171, 60-64 
Patel, V., C. Todd, M. Winston et al., 1998. Outcome of common 
mental disorders in Harare, Zimbabwe. British Journal of 
Psychiatry 172, 53-57 
Reeler, A.P. 1998. Epidemic violence and the community: a 
Zimbabwean case study. Community Development Journal 33, 
128-139 
Reeler, A.P. 1998. Compensation for gross human rights violations: 
Torture and the War Victims Compensation Act. Legal Forum 10, 
6-21 
Reeler, A.P. 1994. Is torture a post-traumatic stress disorder? Torture 
4,59-65 
Reeler, A.P. and R. Immerman, 1994. An initial investigation into 
psychological disorders in Mozambican refugees: Prevalence and 
clinical features. Central African Journal of Medicine 40, 309-315 
Reeler, A.P. and M. Mupinda, 1996. An Investigation into the sequelae 
of torture and organized violence in Zimbabwean war veterans. Legal 
Forum 8, 12-27 
Reeler, A.P. and M. Mupinda, 1995. Report of a pilot project to assist 
survivors of torture and organized violence in Mount Darwin District, 
Zimbabwe, April-July 1995. Harare, AMANI 
Reeler, A.P. and C.H. Todd, 1995. An overview of psychological 
disorders and psychiatric services in Zimbabwe in Y. Pillay and A. 
Bhana eds., Proceedings of a Primary Mental Health Care 
Workshop. Durban: University of Durban-WestvUle 
Reeler, A.P., H. Williams and C.H. Todd, 1993. Psychopathology in 
primary care patients: A four-year study in rural and urban settings. 
Central African Journal Of Medicine 39, 1-8 
Sarantidis, D., et. al., 1996. Long-term effects of torture of victims 
during the period of dictatorship in Greece. Torture 6, 16-18 
Straker, G. 1987. The continuous traumatic stress syndrome - the 
single therapeutic interview. Journal of Social Development in 
Africa 8, 48-78 
168 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 No. 2 JULY 2000