HCI and Multiculturalism in Southern Africa

Jacques Hugo, CHI-SA
February 2002

Introduction

The quest for socio-economic empowerment in any country requires that we emphasise education, training and resource accessibility. Regrettably, almost the whole of Africa has an enormous skills deficit that needs to be made up. In the knowledge society that will characterise the era ahead, it is not enough just to concentrate on literacy and primary education. We need to make a quantum leap and actively embrace methodologies and technologies that will help leverage South Africa's commitment to revival and growth.

Although the local and international press is making no secret of South Africa's comparatively poor international competitiveness, innovation and productivity, it is encouraging that the much-publicised "African Renaissance" includes among its many objectives the "active stimulation of information transfer, intellectual inputs, patriotic investments, and the promotion of Africa and things African".

In response to this, various institutions in South Africa are calling upon local and international experts to become involved in the revival and growth of African economies and industry. Foremost among the areas singled out for urgent and immediate attention are the most powerful enablers of socio-economic growth in the new millennium: education, skills development, and the development of the Information and Telecommunications Technology (ICT) industry.

The promotion of "things African" refers particularly to the recognition and exploitation of cultural values in all areas - government, education, commerce and industry. Although HCI is still in its infancy in South Africa, the changing socioeconomic landscape in South Africa is where some of the most exciting and challenging opportunities for practitioners in interdisciplinary fields like HCI lie. In fact, South Africa can be regarded as a virtual laboratory where the many complex socio-politico-economic issues can be investigated at first hand. We know that there is a link between culture and economical and technological development, but there are still many unanswered questions and problems related to the relationship between multiculturalism and HCI.

It is not the purpose of this article to attempt to answer these difficult questions. I will rather highlight the opportunities presented to researchers, educators and developers in the HCI community by the unique perspective on "multicultural HCI" in South Africa.

HCI in South Africa - problems and opportunities

At present HCI in South Africa can be characterised as follows:

• A scarcity of qualified practitioners and educators. There are only about 20 people involved full-time in HCI. The majority are academics and only three or four are working as full-time practitioners in the industry.
• A lack of awareness and implementation at industry level - the evidence for this is found in the lack of press coverage of HCI or usability matters and the lack of membership of professional human factors societies. At present there are only sixteen SIGCHI members in South Africa.
• Isolation and fragmentation between academia, industry, private Research and Development, and government. There is little coordination between universities to ensure conformity and standard of HCI curricula. HCI in industry is largely limited to usability testing, but, as laudable as this might be, the scientific rigor of the methods employed is not very high yet. Usability principles are often enthusiastically implemented, but due to a lack of resources and inadequate training, guidelines are blindly adopted from literature and few structured methodologies are followed.
• Inadequate or non-existent funding of HCI research. At present funding sources are limited to university research budgets and so far no large-scale formal studies outside academic environments have been undertaken. However, the South African government has approved significant funding initiatives for IT applications, especially for community development projects. We are hoping that an appeal to funding agencies in government for inclusion of HCI initiatives will reach sympathetic ears.
• In certain HCI research studies we often find that researchers blend concepts of culture with race and also incorrectly equate language with race. This could be due to the fragmentation mentioned earlier - vested interests often prevent interdisciplinary collaboration. The result is a tendency to approach questions about the relationship between culture and technology from either an anthropological, or from a technology angle.

What is culture?

Before we continue, it is necessary to first consider the meaning of the term "culture" in this context:

Culture can be defined as "the shared patterns of behaviours and interactions, cognitive constructs, behavioural norms, expectations and affective understanding that are learned through a process of socialisation. These shared patterns identify the members of a culture group while also distinguishing those of another group." [1]

From this we can infer that the term "multicultural" refers to different cultural patterns found among different groups of people in a particular country of region. Although "cultural diversity" is often used as a synonym for multiculturalism, it actually refers to differences between different cultural groups in a specific environment, for example multiple groups within a defined environment, such as a workplace or township.

When we apply this definition to the shared patterns in a nation, we can speak of a "national culture" which is a set of values, attitudes, beliefs, and norms shared by the majority of the inhabitants of a country. These shared patterns become embodied in the laws and regulations of the society, as well as in the generally accepted norms of the country's social system.

Although it might be possible to "unpack" the shared patterns into a generic list of components, this will always result in artificial divisions because all components are interdependent. I have nevertheless found the following classification of components useful for discussion purposes:

• Learned behaviours
  • Habits
  • Communication habits
  • Customs
  • Locus of control (internal or external)
• Ethnicity and Social organisation
  • Race
  • Language
  • Religion
• Environment
  • Economic conditions
  • Physical environment (climate, geography, etc.)
  • Diet
• Values and beliefs
  • Work ethic
  • Dress, fashion
  • Politics

The South African cultural orientation

In any multicultural society learning about culture is an exciting and creative process. But it can also provoke anxiety because of the painful realities that constitute the histories of South Africa and the ongoing interactions with the content that we have with diverse cultural realities. Cultural content and encultured learning, as one example, is a complex sociological and psychological process and one topic that is already being researched by South African institutions. Researchers are discovering that culture is far more exciting and interactive than simply digitising various cultural artefacts.

I believe that certain attributes of culture cannot be divorced from a certain group. Others, like language or race, are not necessarily unique to a culture. Moreover, the framework, according to the definition above, is not static and certain aspects of culture, like communication habits, can change over time, especially under pressure of influences like information technology which tends to change life styles and expectations, and hence behaviour. Also, a large part of culture is learned behaviour common to a given society and this acts like a template in the sense that it shapes behaviour and consciousness in that particular society from one generation to the next. [1]

The influence of culture

In South Africa the influence of culture on technological innovation and deployment is profound and complex, but evidence of this is often only visible to those who search for it. I suspect that most people only become aware of these influences when they are at the receiving end of designs based on ill-conceived population stereotypes. For us as practitioners, the development of a unique South African HCI perspective should begin from the recognition that users are cultural beings, and for this reason, HCI researchers and practitioners need to pay attention to the processes of acculturation [2], enculturation[3] and formation of cultural identities (sometimes multiple identities!) under globalisation pressures.

South African society is shaped by indigenous socio-economic forces and cultural practices as much as it is by globalisation forces. Computer users are not a homogeneous group and for that reason, it is important to acknowledge and understand the role played by human factors, not only in the design of software for diverse user groups, but also in our efforts to build a healthy ICT industry that can enrich people's economic, social, cultural, and political lives.

There is no doubt that ICT, as one of the major drivers of globalisation, plays a significant role in hastening change, particularly in terms of hastening acculturation from an indigenous cultural system towards that of individualism. We know that ICT is both a product of change and an agent of change, in the developed world and also in developing countries. HCI practitioners should therefore understand how these technologies (whether device, interface or the institution represented by the technology) might enhance or undermine the indigenous cultural systems of society.

In his address at the African Renaissance conference [5], Professor Ali A. Mazrui [4] emphasised that culture is relevant for development in all spheres of society because of seven key functions:

1. Culture functions as lenses of perception - it influences how people view themselves and their environment.
   
   
2. It serves as a spring of motivation - what people respond to as incentives or disincentives for certain patterns of behaviour is a phenomenon which is greatly influenced by culture.
   
   Comment: The work ethic is often a product of cultural configurations. Is the work ethic in Africa cultivated or stifled by culture?
   
   
3. Culture provides a standard of judgment. What is right or wrong, what is virtuous or evil, what is beautiful or ugly are all greatly conditioned by culture.
   
   
4. Culture forms the basis of socio-economic stratification. Rank, caste, and class are all profoundly conditioned, if not created, by culture.
   
   
5. Culture is a means of communication - it provides all sorts of nuances in communication and intimation. But above all culture provides language in the literal sense of the legacy of words and lexicon.
   
   Comment: Can any country approximate first rank technological and economic development if it relies overwhelmingly on foreign languages for its discourse on development and transformation? Will Africa ever effectively take off when it is held hostage so tightly to the languages of its former imperial powers?
   
   
6. Culture defines and influences production and consumption. Cultures differ widely in productivity, not only in Africa but the world as a whole.
   
   
7. Culture is a basis of identity - it is crucial in defining the "we" and "they" and marking the frontiers of solidarity.
   
   Comment: How can development tap into the complexities of identity to achieve results? And how do differences between men and women affect issues of identity?

When one considers the issues of science and technology, it is this complex configuration of cultural functions that the African Renaissance has to respond to. When we face the developmental challenges of the future, we must understand that for South Africa "development" inevitably means "modernisation", but without dependency on a legacy of foreign aid, resources and products. At the recent World Economic Forum Deputy President Jacob Zuma again emphasised that Africa needs investments, not handouts. How to become self-sufficient without either sacrificing indigenous values, or alienating your allies, is one of the key issues the African Renaissance has to deal with. As Professor Mazrui points out, " modernisation is change that is compatible with the present stage of human knowledge, which seeks to comprehend the legacy of the past, which is sensitive to the needs of the future, and which is increasingly aware of its global context." [5] This is the positive interpretation of modernisation. Development of indigenous skills and values are at the core of the African Renaissance.

Again we must ask, where does culture enter into this? If development equals modernisation without dependency, there is no doubt about the relevance of the African Renaissance in so far as it aims to stimulate the regeneration of indigenous values. African culture is central to this process of reducing dependency in the dialectic of modernisation.

To date, most researchers have relied on Geert Hofstede's (1997) dimensions of culture to help analyse and describe cultural variables in a particular region, community or nation. In most countries, as Hofstede points out, a dominant culture exists and it should be possible to identify the characteristics of such a culture to determine how we should design our technological artefacts to suit the majority of users. This might be true for some countries, but, due to the dramatic socio-economic changes we experience at present, South Africa does not have one dominant culture that is so mature and powerful that it can absorb different cultures and also be enriched in the process. As in many other "multicultural" environments, we find different groups of people with different religions, race, and ethnicity and they all speak different languages. Several cultural forces are in evidence and these forces are constantly in flux, for reasons that should best be explained by anthropologists and sociologists.

This has a dramatic effect on the development of a globally competitive and nationally relevant ICT industry in our country. The history of information technology development elsewhere in the world is repeating itself in South Africa. For the vast majority of IT practitioners in South Africa, the field of HCI is very obscure and far removed from their everyday work and interests. The prevailing approach to application development is still characterised by the application of engineering techniques to technical problems, without much, if any, attention to human factors. As elsewhere in the world, this results in poor application usability, poor user performance, low productivity levels, and users' dependence on support and training. The realisation of the importance of a user-centred approach in software design has not yet filtered through to the South African IT community. Combined with the lack of attention to cultural differences, this has far-reaching implications for empowerment of the disadvantaged masses, for education and training, and for economic development.

Seen against the background of South Africa's social and political conditions, it seems that even poor usability can be blamed on factors like linguistic, cultural, literacy and economic diversity. These realities can in turn be located within an historical context, for example the legacy of apartheid education that has now forced South Africa into trying to catch up with the competency bases of developed countries.

In South Africa, as in most developing countries, a digital divide exists between the technological "haves" and "have nots". Of course this is not a new idea, and worldwide it is generally recognised that for the "haves", computer-based commerce and dynamic networked communication is not only possible but also essential. Also in South Africa this community has eagerly adopted ICTs and moved beyond e-mail and the web; we are seeing signs of new electronic communities that are evolving from the convergence of domains and technologies. For example, young people are finding it more and more difficult to distinguish between the use of handheld devices (PDAs, cell phones, MP3 players, etc.) for entertainment, work or communication. People's attitudes toward new devices are changing rapidly as they discover new benefits and it is generally assumed that these devices will have great personal or social value, and therefore inevitably result in cultural paradigm shifts.

Signs of this paradigm shift are becoming more evident by the day; the population that could potentially use and benefit from information technology, seems to grow culturally, economically and educationally more diverse. Particularly for the "have nots" the situation is very different. Users from these communities have different expectations of technology. These expectations are coloured by their frame of reference, educational level, cultural prejudices, career expectations, income, and many other variables that are very poorly understood by software developers, nor accommodated within current design practices in South Africa.

Avoiding technological determinism

Seen from from a human factors point of view, traditional designers and most developers of new technology have been making too many assumptions for too long about user prejudices and preferences in the adoption of innovations. We find that they either ignore these perspectives entirely, or they over-compensate. We find such assumptions about the prejudices and preferences in a particular region or community in the form of ideas about colour associations, how messages should be structured, forms of address and honorifics, preferences for "ethnic" designs, and many more. This emphasises the importance of understanding the differences and also the similarities through the eyes of those who should benefit from the product.

Without this understanding, the marketplace will impose a "one size fits all" attitude on South African users. This will inevitably result in excluding various sectors of society from access to, and benefiting from information technology. Without proper understanding of the cultural variables involved, new media and communication devices threaten to expand, rather than resolve, the divide between haves and have nots.

Many IT missionaries envision the development of an information infrastructure that can enrich people's economic, social, cultural, and political lives. For such an infrastructure, or, in fact, many other applications of computer and network technologies, to be successful, requires solutions to all these issues: how to build effective and appropriate human-centred systems. The time has come to focus more on designing active information spaces that exploit the humanisation potential made possible by information technology, and thereby avoid technological determinism.

To avoid elitist, exclusionary and discriminatory practices, ICT practitioners should focus more on designing interactive information environments that will facilitate access, learning and empowerment, and in so doing, exploit the new cultural and work representations for all, leveraged by information technology. Appropriately designed systems should not only acknowledge our unique South African characteristics, but should also help to open the doors to collaboration and integration with the international community.

Avoiding imitation

When we consider the urgent need for HCI education in South Africa, we are also faced with the challenge of closer cooperation between industry and academia. In developing countries there is often a tendency to regard countries that are seen as technologically and economically more advanced, as the source of superior ideas and values - an attitude that might be described as a "technological halo effect". But what is good for one country or one user is not necessarily good for another - "imported" doesn't necessarily equate with "better" and is often an insult to the skills and capabilities of local practitioners. We therefore need avoid the obsolete tendency to simply transplant so-called tried and tested methods and products. We need to develop our own resources and strengthen our own competence.

The impact of software design

The cultural diversity of South African society has important implications for software as far as user-centred design is concerned. For example, under normal circumstances the user and task analysis techniques offered by HCI provide adequate information about users and their work. However, these techniques are inadequate when a large number of cultural variables must be examined. To cope with cultural diversity and still ensure optimum performance, a designer needs to know about a wider range of factors that will affect a person's work and social behaviour in a technological environment. This implies that the South African HCI practitioner cannot function effectively without including ethnographic techniques in his or her toolbox.

For example, we must be careful not to relegate users' perceptual idiosyncrasies and varying cognitive strategies to simple "user preferences" that have no real technical merit. Since perceptions affect behaviour, we must determine to what extent this will affect the intended functionality and usability for this class of user. This question must be extended to a consideration of the impact of the system on the quality of the user's work life.

HCI advocates familiarity with the users' task domain. Users' computer-mediated tasks are embedded in their work, and their work is embedded in their work culture. This implies that in multicultural environments analysts need to do more than "standard HCI practice" to understand these broader domains. Common HCI practice is typically concerned with integrating diverse sources of information. It is common to hear advice such as "know thy users, for they are not you," and to study the users' information- or task flow. As praiseworthy as it is to be sensitive to, and have a respect for the users as skilled and competent members of their cultures, analysts should be careful not to claim that they "learned the user's job well enough to do it myself". This is an obvious fallacy - no matter how much you feel yourself part of the user's culture - the mere fact that you are the analyst or designer automatically creates an ideological and cultural distance between you and the user.

A cultural orientation (perhaps more correctly an ethnographic orientation) is therefore a logical approach to contextual analyses in multicultural environments. This approach emphasises the users' competence in a dialogic setting with members of other disciplines. In this sense, users are not only competent members within their own cultures - they are also competent interpreters of their own and of others' cultures.

The aim of participatory design has always been to encourage direct work with users, as well as sharing knowledge elicitation and decision-making. An ethnographic orientation extends this approach through its concern that artefacts and interpretations from the users' domain be understandable and verifiable by the users and that HCI professionals consider whether HCI usage of the users' materials or knowledge might be contrary to the users' wishes or interests.

Current work on multiculturalism in HCI

In response to the situation sketched at the beginning of this article, we now need to ask:

1. South African is a multicultural country, but what makes us different from countries like Brazil or India, or even the USA, and do we have something to offer the international HCI community?
2. What are the variables of culture that seem to have an effect on the way people assimilate technology and how do these variables differ from other countries, if at all?
3. If we know these variables, how should they be incorporated into our analysis, design and development techniques?
4. It appears that users in multicultural environments are often very resilient and more likely to make allowances for software that does not quite represent their own cultural values. Does this mean that we need to pay less attention to "localisation"?
5. To what extent can we rely on the results of multicultural studies in other countries?

In a bold initiative, South African practitioners have started to tackle some of these issues, and we will report on the progress at the CHI 2002 Development Consortium:

1. Web usability in a multicultural environment. Lizette deWet, Pieter Blignaut & Andries Burger (University of the Free State) have found that South African web site developers should take cognizance of the fact that Afrikaans-speaking people find it easier to search for information in Afrikaans (in contrast to English). It is suggested, however, that there is no need to translate web sites into an African language. Carina de Villiers and Arcilia Masoeu have done similar work on visual and other representational issues.
2. Customization of interfaces for different cultural groups. Diane Norton's results suggest that customized user interfaces may not be necessary for users of the same computer based systems in South African businesses due to shared representations.
3. Empowerment of semi-literate users. Edwin Blake's research and development project aimed to provide semiliterate animal trackers in nature reserves with a hand-held device to gather complex data on animal behavior. Trackers are experts in their own right and have access to very sophisticated and complex information about the environment. This knowledge is not available to the wider community, mainly because of the barrier of illiteracy. They benefit from greater recognition, while the wider community gains from access to the knowledge of the trackers on animal behavior. The system has become a successful commercial product with a number of applications.
4. Using digital technology to archive and access cultural treasures. Gary Marsden, Katherine Malan and Edwin Blake (University of Cape Town) showed that to understand the environment in which African art is created, it is essential to build systems which present uncertain data about scarce and valuable resources in a meaningful way. Africa also has unique problems in distribution of information, which makes it essential to create interfaces for the widest possible range of distribution. There are unique challenges in creating, and providing access to, a database of African cultural artifacts. Such an archive will serve not only as a store, but also as a way in which some of the African artwork stored in foreign museums can at least be virtually repatriated. It will also serve as an educational tool for sharing artwork across the continent. There is a great potential for Africa to present its rich culture to the rest of the world and to recapture (in digital form at least) those artifacts which are currently housed in museums outside our borders.
5. The future of HCI Education, Research, and Practice in Southern Africa. Paula Kotze's (University of South Africa) study emphasizes that, while the limits of technology seem endless, the real limits of scientific and economic growth lies in humankind's ability to absorb and apply new information and technology. The human-computer interface can contribute to the successful impact of information and technology on society by making it accessible and usable by the wider population. In Southern Africa this population is characterized by a multiplicity of ethnic, cultural, language, education, economic, and other backgrounds. A concerted effort by all role players is required to draw together of all forces within the fields of academia, research and industry to ensure the successful integration of HCI and it related principles and approaches.
6. Visual literacy. In this fascinating study, Marion Walton, Vera Vukovic & Gary Marsden (University of Cape Town) question western cultural assumptions underpinning the web's evolving navigational conventions, and investigate to what extent certain communities (like university students) command the currently dominant Western conventions. Many previous studies originate from a concern with the 'export' of software to culturally different contexts, and consequently focus on the relatively crude practicalities of "internationalization" and "localization". This study particularly investigates an understanding of visual communication beyond the individual icon, at the level of the complex, culturally-relevant combination of both representational (iconic) and non-iconic elements of screen designs.

Conclusion

We see cultural factors at work in the design of interactive software primarily in intra- and intercultural and class differences in the use of abstractions and generalisations, where such generalisations are required to successfully use applications. We also see it in cultural variants based on differences in "locus of control" (i.e. internal or external) and in visual literacy which includes the understanding of colours and symbols, and the role of ethnicity, class, gender and age in the design of visual communication.

To a greater or lesser degree all of these factors may be predictors of individual task performance. If user-centred design does not include a cultural dimension, we in South Africa will remain at the receiving end of foreign computing developments. Software needs to reflect and acknowledge our diversity and it must accommodate the true needs of our upcoming generations.

South African HCI practitioners and their international collaborators should be encouraged to give high priority to all projects aimed at addressing the bigger problem of eradicating elitist, exclusionary and discriminatory practices in the South African ICT industry, and even worse, in products inherited or acquired from other countries. All of our energies should be focuses on promoting products that facilitate access, learning and empowerment, and in so doing, exploit the new cultural and work representations for all, leveraged by information technology.

References

Bourges-Waldegg, P. and Scrivener, S.A.R.  Meaning, the central issue in cross-cultural HCI design.  Interacting with Computers. 9, 1998, 287-309.

Hofstede, Geert. 1997. Cultures and Organizations: Software of the Mind, McGraw-Hill, New York.

Masoeu, A. & de Villiers, C. 2001. Web usability in a multicultural environment: a concern for young South African web users? Proceedings of CHI-SA 2001, Pretoria.

Muller, Michael J. 2000. Ethnocritical Heuristics for HCI Work with Users and Other Stakeholders. http://iris.informatik.gu.se/conference/iris18/iris1844.htm

Onibere, E.A., Morgan ,S., Busang, E.M., Mpoeleng, D. Human Computer Interface Design for a multi-cultural and multi-lingual English speaking country - Botswana. Interacting with Computers. 13, 2001, 497-512.