1. Introduction
Agriculture is one of the most important human endeavours [
1,
2]. It serves as the main source of food, which is crucial for the survival of the human race. Today, however, the global food system is at a crossroads [
3]. Although agricultural systems supply large quantities of food through intensification, their sustainability is questionable [
4]. As climate change continues to dominate scientific discourse, the expectation that agricultural systems will need to adapt is widely accepted [
5,
6]. Resource intensive agricultural practices have catastrophic impacts on the environment and societies at large [
4,
5,
6].
A transition to sustainable agricultural systems is required for social and economic equity, food security, conservation of biodiversity, and provision of ecosystem services [
6]. Permaculture is an approach that could contribute to the sustainability of social and ecological systems [
5]. Although it is difficult to define permaculture, it can be understood in terms of four characteristics that are interrelated [
7].
The first emphasizes the potential of synergistic effects on ecosystem health and human well-being. At the same time, permaculture presents a model of change that stresses the importance of individuals’ and local communities’ ability to act [
8]. Second, permaculture draws on systems thinking and ecological principles, which offer different approaches for analysis, and relevant practices that are unique to each site. Third, permaculture is described as a framework that aims to optimize ecological systems. Finally, fourth, due to its focus on individual responsibility and small-scale solutions, permaculture is less institutionalized than other agricultural approaches [
8]. For the purpose of this study, the definition of permaculture we adopted focuses on the first and third characteristics, which emphasize synergistic effects on ecosystem health and human well-being. Given the focus on rural livelihoods in this study, the following definition of permaculture is used: Permaculture is a design system that uses ecological management practices and locally adaptive solutions for sustainability in all aspects of human endeavour. Such practices aim to optimise the interactions in the soil-plant system for an efficient use of their ecological functions and ecosystem services, while promoting diverse, resilient, and regenerative agricultural systems [
9].
While permaculture is transdisciplinary in its nature, it places emphasis on ecosystem health [
10]. What distinguishes permaculture from other approaches is its comprehensive design process. Each site, whether a garden, farm, school, or household, has a unique set of elements and design considerations [
9]. Its approach to production focuses on multi-functionality and diversity through the use of integrated water management, land use diversification, polycultures, and perennial cropping [
8]. Twelve principles (
Table A1) and three ethics, care for people, care for earth, and fair share, inform the practice of permaculture [
11]. As proposed by these ethics, permaculture is the design of sustainable socio-ecological land use systems. Permaculture recognizes the interdependence of land use systems and social systems [
12].
The design principles that inform permaculture draw on the concept of systems thinking [
13]. Examples of systems thinking include ecosystems in which soil, water, animals, plants, and air all work together. The principles are short statements that give a wide range of options for action when dealing with complex systems. Krebs and Bach (
Table A1) give a summary of the twelve principles of permaculture with broad examples.
There is limited information on the benefits and challenges of permaculture [
8]. However, it has been argued that the application of permaculture is likely to succeed in countries of the Global South [
14,
15,
16]. This has been attributed to the extensive use of traditional agricultural practices and labour-intensive practices in many countries in Southern Africa [
14]. Permaculture also has the potential to galvanize action towards the achievement of multiple sustainable development goals (SDGs) [
17]. Some of these include zero hunger, good health and well-being, and life on land. The uptake of permaculture by smallholder farmers in Southern Africa is, however, slow [
18]. Therefore, more research is required to assess the benefits and constraints associated with the practice.
This study attempted to answer the following two questions. First, what is the effect of permaculture on rural livelihoods in Southern Africa? Second, what are the enablers and barriers to the adoption of permaculture in Southern Africa? Semi-structured interviews were used to collect data. We used both quantitative and qualitative methods to analyze the data.
2. Study Area
This study was conducted in rural parts of South Africa and Zimbabwe. The study areas were Middledrift and Ngqeleni, towns in South Africa, and the Chimanimani district in Zimbabwe. The study areas were chosen as they represent the major agro-ecological regions of both countries.
Middledrift is a small town located 90 km north-west of East London and is situated in the Raymond Mhlaba Municipality in the Amathole District, Eastern Cape (EC), with a population of 20,600 individuals. The region has a mild and temperate climate and receives 412 mm of rainfall per year [
19]. It hosts a range of vegetation types including thornveld, semi-arid Karoo, thicket areas and succulent plants. Communal farming is a dominant practice where vegetables, maize, and livestock are the main agricultural commodities [
20]. Each household has access to 0.08 ha of cropland, comprising of large crop fields as well as small vegetable plots. Key contributors to employment in the district include trade (25.5%) and agriculture (15.1%) [
20].
Ngqeleni village is 40 km north-west of Coffee Bay and is situated in the OR Tambo District, EC, with a population of 2600 individuals. The region has a warm temperate climate and receives 652 mm of rainfall per year [
21]. The region hosts a diversity of vegetation from bushveld to thicket forests. Agriculture forms the basis of primary activities and is mainly subsistence. Vegetables, maize, and livestock are the main agricultural commodities produced in the region [
22]. Key contributors to employment in the district include community services (34%), trade (24%), and agriculture (4%).
Chimanimani is located in Chimanimani district, Manicaland Province in south-eastern Zimbabwe, with a population of 2700 individuals. The south-western parts of the district fall under the semi-arid agro-ecological zone with a harsh environment due to variable annual rainfall ranging from 450–800 mm [
23]. Millet varieties, sorghum, and drought resistant maize are mostly grown in this area, where subsistence farming is the major land use.
2.1. Data Collection
This study adopted a mixed methods approach, which combines philosophy, research design, and orientation methods [
24]. It is a procedure for collecting, analyzing, and mixing both quantitative and qualitative research methods within a study. The use of such an approach allowed for a greater understanding of research problems, in comparison with the use of only one approach [
24].
We interviewed 94 key informants using semi-structured interviews in June 2018. In South Africa, 44 interviews were conducted in Ngeqeleni () and Middledrift () in EC. In Zimbabwe, 50 interviews were carried out in the Chimanimani district. The sample size was limited to 94 respondents due to budget constraints and a limited number of people who had received permaculture training.
Permaculture practitioners were selected as respondents, given the nature of this study. The key informants had received some form of permaculture training, either formally (certified) or informally (non-certified). These key respondents were identified through the help of the two following organisations: Hope Permaculture and Chikukwa Ecological Land Use Community Trust (CELUCT). Hope Permaculture and CELUCT are organisations and training centres that support community-based permaculture projects in South Africa and Zimbabwe, respectively. In addition to identifying key informants, both organizations aided in selecting enumerators from the community, whom we trained on how to conduct the interviews.
Before conducting interviews, all participants signed a consent form which assured anonymity, the right of withdrawal, and confidentiality of their information. Ethics approval was obtained from the Environmental Science Department of Rhodes University and the signed document was in accordance with the Rhodes Ethics guidelines [
25]. We assumed that the respondents gave honest information.
The semi-structured questionnaire consisted of five sections. The first section asked questions pertaining to demographic characteristics such as age, gender, and education. The second section contained questions regarding permaculture training and required information on the kind of training received (i.e., formal or informal). This was followed by a section on livelihood activities, which gathered information on income before and after practising permaculture and other livelihood strategies. In addition to average income earned from other activities, respondents reported how much they earned or saved from practising permaculture. Respondents were also asked to report how permaculture changed their income. This was grouped into the following categories: improved substantially, moderately, slightly, no change, or decreased. The next section required descriptions on the crops grown and inputs used before and after practising permaculture. It also included questions on place of practice, hours of labour, and the size of the cultivated area.
Crops were grouped into cereals, vegetables, legumes, tubers, and fruits. The average number of crop varieties grown in each category was calculated. The final section contained open-ended questions that allowed respondents to share their views on the challenges and benefits associated with permaculture.
2.2. Data Analysis
In addition to standard descriptive procedures, all statistical tests were performed using RStudio version 1.0.153 [
26]. A chi-square test was used to test for association of categorical data, such as change in income.
Quantitative data, such as income, labour hours, and size of cultivated area, were first tested for normality using the Shapiro test. Non-parametric tests were used for non-normal data. Wilcox signed-rank tests were used to test for differences in mean income and the size of cultivated areas between South Africa and Zimbabwe. Furthermore, a paired Wilcoxon signed-rank test compared the average number of crop varieties grown before and after permaculture.
Exploratory factor analysis (EFA) was performed to identify latent factors that were explained by the measured variables. It is a multivariate analysis technique that models underlying relationships among measured variables [
27]. The optimum number of factors was determined by analysing the scree plots (
Figure A1 and
Figure A2) [
28]. The point at which the steep curve flattened was an indication of the optimal number of factors required. The variables used for EFA included age, gender, education, type of permaculture training, average income, proportion of income contributed by permaculture, size of cultivated area, and hours of labour.
Content analysis was used to analyse the challenges and benefits associated with permaculture. Moreover, recurring themes were assessed that emerged from the participants’ responses. These themes were grouped into categories. The chi-square test was used to test for associations of the type of challenge or benefit mentioned.
3. Results
3.1. Demographics of Permaculture Participants
In South Africa, 53% of the respondents were female, whereas there was an equal representation of male and female respondents in Zimbabwe (
Table 1). The majority of respondents from both countries were middle-aged, ranging from 35–54 years old.
Most respondents in Zimbabwe received high school education, whereas the majority of respondents in South Africa attended a technical college (
Table 2). All participants received some form of permaculture training, either certified or non-certified. In both South Africa (63%) and Zimbabwe (60%), the majority of the participants received formal (certified) training, whereas almost 40% of respondents from both countries learned informally through social/peer learning.
3.2. Years Spent Practising Permaculture
Zimbabwean practitioners had 1.75 times more years of experience than South African practitioners. There were highly significant differences in the duration of practice between Zimbabwean and South African respondents (). Respondents in South Africa were fairly new permaculture practitioners, with years of experience. On the other hand, Zimbabwean practitioners had years of experience.
3.3. Place of Permaculture Practice
Respondents practised permaculture in three different places: farms, home gardens, and school gardens (
Table 3). Permaculture was most commonly practised on home gardens in South Africa (61%) and Zimbabwe (46%). An association was found between the place of practice (
) and country. While a small percentage of South African respondents (12%) practised permaculture on school gardens as well as on municipal land, no Zimbabwean respondents were found to practise permaculture in these areas. A significantly higher proportion of Zimbabwean respondents (34%) than South African respondents (9%) practised permaculture on both their home gardens and farms. Those that practiced permaculture on farms did so for both home and commercial use.
3.4. Size of Area Cultivated Using Permaculture Practices
A significant difference in the average size of cultivated area between the two countries was found (
). The smallest cultivated area was found to be 12
in South Africa and 42
in Zimbabwe (
Figure 1). In contrast, the largest areas were 50,000 m
and 30,000 m
for South Africa and Zimbabwe, respectively.
3.5. Hours of Labour Practising Permaculture
A highly significant difference was found in the number of hours spent practising permaculture between the two countries ). On average, South African practitioners put in h of labour per week. On the other hand, Zimbabwean practitioners put in h per week.
3.6. Income of Participants
The average income of South African participants was $ and $ for Zimbabwean participants. The exchange rate was USA $1 to R14.51 South African Rand at the time of the study. A Wilcoxon test revealed a highly significant difference in income earned by the participants of the two countries (). South Africans earned 2.4 times more than Zimbabweans. On average, permaculture contributed 47% and 45% towards total income for South African and Zimbabwean practitioners, respectively. There was no significant difference in the mean proportion of income contributed by permaculture between the two countries ().
Significant associations were found in income change categories after applying permaculture between South Africa and Zimbabwe (
). In South Africa (35%) and Zimbabwe (50%), the majority of respondents noted that their income improved slightly. Highly significant associations were found in income change categories in Zimbabwe (
). Permaculture substantially improved the income of 18% of the respondents (
Figure 2). No significant associations were found between the income change categories within South Africa (
= 4.55,
p > 0.05). Individual counts and chi-squared values are given in
Table A2.
3.7. Livelihood Strategies
In addition to permaculture, there were two main sources of income, income generated from conventional agricultural practices and non-agricultural related practices – from activities such as knitting, building, and formal employment (
Figure 3). It was apparent that respondents from both countries did not rely on permaculture as their dominant source of income. Of the 44 South African participants, only four relied on permaculture as their main source of income. Of the 50 Zimbabwean participants, only one relied on permaculture as their dominant source of income. The most common source of income for South African respondents was teaching. Whereas Zimbabwean respondents relied heavily on conventional agriculture and vending as additional sources of income.
3.8. Factor Analysis of Socio-Economic Variables
3.8.1. Correlation of Different Variables for South African Permaculture Practitioners
A strong correlation between hours spent practising permaculture and income was identified using EFA (
Figure 4). Participants that put in more labour hours and received formal permaculture training gained more economic benefits from permaculture than those who put in less hours and received informal training. The duration of practice was not correlated with any items. Based on the variables that loaded onto Factor 1 (MR1), the factor was identified as the economic returns from permaculture and accounted for 51% of the variance.
A realistic hidden variable could not be identified for Factor 2. Age had a strong negative relationship, but no meaningful associations could be drawn from the remaining variables, which all had weak relationships.
3.8.2. Correlation of Different Variables for Zimbabwean Permaculture Practitioners
Similar to South Africa, Factor 1 was also identified as economic returns from permaculture, as it loaded highly with size of area, labour hours and income (
Figure 4). Those that cultivated on large areas required additional hours of labour to achieve a positive change in income. Factor 1 accounted for 18% of the variance.
Factor 2 was identified as experience in permaculture as it loaded highly with age and duration of practice. This indicated that older practitioners had more experience. Factor 2 accounted for 13% of the variation.
Factor 3 indicated that the proportion of income was an independent variable and accounted for 11% of the variance. Lastly, Factor 4 accounted for 11% of the variance and indicated that more males than females received formal permaculture training.
3.9. Agricultural Inputs Before and After Practising Permaculture
In both South Africa and Zimbabwe, over 70% of the participants shifted from using chemical to organic fertilizers (
Figure 5). For participants that also employed conventional agricultural practices, chemicals were still used. Moreover, many South African (70%) and Zimbabwean (50%) participants shifted from using chemical to organic pesticides. Participants in both countries also shifted from using hybrid seeds to non-hybrid seeds, that they often saved.
3.10. Crop Diversity After Practising Permaculture
In South Africa, participants grew eight different types of vegetables, whereas in Zimbabwe, six different vegetables were grown (
Figure 6). Zimbabwean practitioners grew five different varieties of legumes compared with South African practitioners, who grew three different varieties of legumes. Respondents in both countries grew the same types of cereals, that is maize, millet and sorghum.
Highly significant differences were found in the average number of fruits (
V = 325,
p < 0.001) and vegetables (
V = 539,
p < 0.001) grown after practising permaculture in Zimbabwe. In South Africa, highly significant differences were also found between the average number of vegetables (
V = 507,
p < 0.001) and fruits (
V = 325,
p < 0.001) grown. In both countries, participants grew over twice the number of vegetables and nearly five times the number of fruits. Paired Wilcoxon values for all crop varieties are given in
Table A3.
3.11. Benefits Associated with Permaculture Practices
Perceived benefits of permaculture, among participants from South Africa and Zimbabwe, were grouped into three categories, as follows: quality of life, environmental, and economics, as illustrated in
Table 4. No associations were found between the type of benefits mentioned between the two countries (
). However, significant associations between the type of benefits mentioned were found within South Africa (
) and Zimbabwe (
). Few South African (14%) and Zimbabwean (16%) respondents viewed permaculture as an affordable practice. In contrast, over 65% of respondents from both countries reported feeling healthy because of eating more nutritious food.
With regards to environmental benefits, a higher proportion of South African (52.3%) than Zimbabwean (16%) respondents mentioned that permaculture fostered respect for nature and people (). Economic benefits included the reduction of input costs, which created the opportunity to save money. Lastly, over 25% of respondents from both countries indicated higher yields as a benefit.
3.12. Challenges Associated with Permaculture Practices
Challenges associated with permaculture were also grouped into three categories: education, environmental and economics, as illustrated in
Table 5. Highly significant associations were found within the type of challenge between the two countries (
). A significantly higher proportion of South African (43%) than Zimbabwean (6%) respondents mentioned that they lacked knowledge on how to deal with problems using permaculture practices.
With regards to the environmental category, the biggest challenge faced by participants of South Africa (41%) and Zimbabwe (52%) was drought. From an economic point of view, over 30% of respondents from both countries were concerned about the inefficiencies linked to permaculture, such as it being a labour intensive practice. A higher proportion of South African (25%) than Zimbabwean respondents (6%) found low yield to be a challenge. Some participants in South Africa (20%) and Zimbabwe (8%) also mentioned that permaculture required high initial capital.
5. Conclusions
Permaculture can be effective in supporting multiple objectives. As revealed in this study, permaculture may help support livelihood activities and improve the ability of farmers to deal with environmental problems. It also holds the key to increasing dietary diversity within households and enhancing social and ecological resilience. However, as with any agricultural system, it has its limitations. While permaculture on its own may not match the yields produced through conventional techniques, the prudent path towards reforming the global food system will require holistic approaches that have a neutral environmental effect and are economically viable.
This implies a transition from conventional, monoculture-based, and intensive production towards an array of sustainable regenerative production systems that improve productivity. Furthermore, shifting from a linear to a holistic approach in agricultural management is necessary. An approach that acknowledges the role of people as not mere producers of food, but also as managers of ecological systems that produce a suite of ecosystem services is needed.
A necessary step for this will require institutional support that favours transitions across all stages of food systems, which includes improving the resilience of rural livelihoods. A space for knowledge exchange on sustainable agricultural practices, coupled with a supporting environment and strong governance are also vital. This entails a strong emphasis on alternative agricultural practices in national strategies. Lastly, increasing funds to support such transitions will be required from community to national levels in order to advance socio-economic development.