Mediation Practice in the South African Construction Industry: The Influence of Culture, the Legislative Environment, and the Professional Institutions

Although the potential for dispute is inherent in any contractual relationship, disputes in construction are common and appear to be inevitable, given the unique and complex nature of most construction projects (Chan 1998; Clegg 1992; Delmon 1998; Hibberd and Newman 1999). Few would disagree with Sir Michael Latham's (1995) view that the best solution to the problem is to prevent disputes. However, despite good intentions, disputes do arise, and often the successful completion of a project relies on their expeditious resolution.

Avenues for dispute resolution in South Africa include negotiation, mediation, adjudication, arbitration, and litigation. The focus of this article is on mediation, and more specifically, on mediators, based on the results of a survey of mediators aimed at understanding how they perform this role. Mediation has been used in the South African construction industry for more than two decades, and was first introduced into the General Conditions of Contract (GCC) for Works of Civil Engineering Construction in 1982. The 1990 version of the GCC preserved this provision, but the 2004 version removed the mandatory requirement that parties must mediate before proceeding to arbitration. In 1991, mediation was also included as an option (provided that both parties were willing) in the Principal Building Agreement (PBA) published by the Joint Building Contracts Committee (JBCC). These two agreements, the 1990 GCC and the PBA, are the forms of contract predominantly used for civil engineering and building work. The PBA and the GCC are the two key forms of construction contract used in South Africa. They have been developed and endorsed by key industry stakeholders and the professional institutions involved in the construction sector.¹ However, contractual adjudication and Dispute Review Boards are slowly being introduced into the industry with the introduction in 2004 of the GCC for Construction and the increased use of the international Fédération Internationale des Ingénieurs‐Conseils contract documents (van Langelaar 2001).

The context for our investigation of mediation and mediators is post‐apartheid South Africa. In this article, we discuss the legal system and the role played by industry institutions in promoting mediation, and also explore whether or not South African culture has influenced the approaches to dispute resolution that have been taken in the construction industry.

In other countries, research has generally sought to establish the perceptions, attitudes, and experiences of industry participants toward mediation as an ADR mechanism (Brooker and Lavers 2000; Gould 1999; Stipanowich and Henderson 1992). Research on the mediation of disputes in the South African construction industry has been limited. B. W. Barth (1991), Rolf Schindler (1989), and Mark Watson (1996) have investigated the role of mediation and arbitration as construction industry dispute‐resolution mechanisms. Schindler (1989) focused on the awareness, experience, attitudes, and perceptions of architects, engineers, and contractors to mediation and arbitration, concluding that, despite little actual experience in mediation, they had developed negative attitudes and perceptions about it. Barth (1991), on the other hand, investigated the suitability of arbitration as a dispute‐settling mechanism, and concluded that mediation was considered by the industry participants (including attorneys) to be a more suitable mechanism than either litigation or arbitration. Watson (1996) analyzed forty‐four disputes to evaluate the effectiveness of the various dispute‐resolution processes, and established that 85 percent of the cases were resolved through mediation at a fraction of the cost and in considerably less time than was often the case when similar disputes were arbitrated.

Is there an African, or South African, way of doing business, avoiding conflict and settling disputes? To answer this question, one must first understand the post‐apartheid South African context, well described by Wim van Binsbergen as:

Despite the tensions van Binsbergen (2002) describes, South Africa is a country with a positive outlook. Its government is fundamentally concerned about humanity and dignity, and government‐led processes (in which we can include the transition to democracy itself, the drafting of the constitution, and the Truth and Reconciliation Commission process) in policy formulation and implementation are typically broad‐based, inclusive, developmental, and where appropriate, reconciliatory. This approach can be ascribed to the influence of ubuntu, a Southern African² cultural worldview that highlights the unity of humanity and places emphasis on the principles of empathy, sharing, and cooperation in efforts to resolve common problems (Murithi 2003). It is a concept used by academic philosophers to evoke forms of precolonial African social life (that may still survive to a degree in remote villages and intimate family relationships) that provides an “inspiring blueprint for the present and future” (van Binsbergen 2002: 1). Ubuntu societies have well‐developed mechanisms for resolving disputes and place importance on the maintenance of positive relations within the community (Murithi 2003). Further, emphasis is placed on restorative, rather than retributive, justice, thereby creating an environment for reconciliation in the process of conflict resolution.

The source of the ubuntu influence is the older, more traditional generation to which many of the country's leaders belong. It would, however, be incorrect to describe modern Southern African societies as ones in which ubuntu is standard. On the contrary, most young urban citizens of South Africa are as globalized as the people anywhere else in the world. Aided by the burgeoning reach of information and communication technologies and all that this implies, young South Africans have been exposed to “globally circulating cultural, productive, reproductive, and consumptive models”— to the extent that they have no real direct contact with the worldview of Southern African ubuntu societies (van Binsbergen 2002: 7). Some might question whether this particular worldview can be applied to all South Africans in such a diverse, heterogeneous society. We think it can because those population groups that may be considered “immigrants” are now at least fourth‐generation South Africans and are deemed to be as African as the indigenous tribes. In addition, it is commonly assumed that those South Africans who have chosen to remain in a post‐apartheid South Africa and participate in its transition to democracy are comfortable with the inclusive concept of ubuntu that is epitomized in the policies of the ruling party, the African National Congress.

It would similarly be incorrect to describe the South African construction industry as one in which an ubuntu approach is standard — in either project management or dispute resolution. Panteleo Rwelamila, Alfred Talukhaba, and Alfred Ngowi (1999) argue that on public‐sector projects in the Southern African Development Community (SADC) countries,³ much of the failure by project managers to deliver projects that meet client expectations of quality, cost, time, and utility derives from their failure to treat internal clients, who range from site cleaners to highly skilled specialists, with respect and dignity. The authors argue that, in the SADC countries, the majority of internal clients are African, that ubuntu is part of their culture, and that this implies that they must be made to feel “part and parcel of the whole building project‐management system,” in recognition of the respect and dignity to which they feel entitled. Failure to recognize this, the authors argue, results in an ongoing internal conflict to the detriment of the project, while recognition of it can help prevent conflict. The context of this discussion is ubuntu within the project team, but clearly, the authors also imply that, if the relationship between the project team and the employer (or external client) were an ubuntu relationship, the potential for conflict and disputes would be reduced.

Returning to the question posed at the beginning of this section, it seems that we can say that there is a Southern African way of doing business, preventing conflict, and resolving disputes. However, we cannot say that it is significantly in evidence in the construction sector. Many construction disputes that concern large amounts of money end up in court. It is widely agreed that this is the most expensive and time‐consuming route that could be taken, but the reasons why it is selected are universal (maximization of financial gain, pride and/or perception that justice will be done, little or no concern about an ongoing relationship between the parties, etc.).

In South Africa, there is an institutional commitment to the general use of ADR methods, evidenced by the clauses in standard construction‐industry contracts promoting their use. The institutions representing the formal sector construction and building industries include both client/ professional bodies and contractor organizations (see note no. 1), and the standard contracts we refer to are those referred to in the introduction of this article, namely, the 1990 JBCC PBA and the 1990 GCC (see note no. 1). While both of these contracts make it mandatory to take disputes that cannot be settled by an architect or an engineer to arbitration, only one of them stipulates that disputes must go to mediation. In the JBCC PBA, mediation is listed as an option if both parties agree, whereas in the 1990 GCC it is stipulated as a mandatory next step if the engineer's decision is not accepted. The 2004 GCC eliminates mediation as a necessary step, and introduces adjudication as the default method.

It is commonly understood that arbitration and litigation are the predominant forms of dispute resolution in the South African construction industry (although the exact number of arbitrations is all but impossible to determine because these are private proceedings and no formal records are kept). This is consistent with the type of firms found in the building and civil‐engineering industries — which we suspect are still predominantly white owned. (All government contracts require a 50 percent black partnership, but it is widely believed that there is a great deal of “fronting” where construction companies strategically acquire black partners as window dressing in order to secure public work. The only evidence of black entrepreneurs in the market ten years into the new South Africa is still at the small‐ and medium‐enterprise level.) In addition, save for those that are required to use labor‐based methods under government contracts, most of these firms are typically “first world” in their organizational structure and production methods. (Labor‐based contracts specify that the method of construction must involve human labor in preference to machines.) Thus, while we argue that there is little apparent institutional commitment to mediation and the principles of ubuntu, we also recognize that the membership of these institutions derives from the professional consulting firms and contracting firms that make up the building and civil‐engineering industries.

These industries still largely follow the colonially inherited traditional procurement system⁴ (Rwelamila, Talukhaba, and Ngowi 1999), in which the fragmentation of the project team, and in particular, the contractor's exclusion from the design stage, has been widely criticized as a recipe for conflict. Because the design and cost‐management consultants typically control — or at least heavily influence — the choice of procurement system, it can be argued that the predominance of the traditional procurement system and the use of the standard industry contracts (which have not changed much over the past two decades) that accompany this choice reflect an industry mind‐set that is not very interested in mediation. This predisposition arises from two factors. First, the use of the traditional procurement system puts the JBCC (PBA) and the GCC forms of contract in play, and these two contracts reflect a bias toward adjudication, arbitration, and litigation by the very nature of the clauses contained therein. Second, the design and cost‐management consultants are bound to specify these contracts as a direct result of their membership in the professional institutions that formulated them. An aggravating factor is that partnering is not very widespread in the South African construction industry primarily because government does not support it and all public money must be awarded through an open bidding process. This makes it less likely that more inclusive forms of dispute resolution that protect partnering relationships between employers and contractors, such as mediation, will be chosen.

Against this background, we present in the next section the results of a survey into the practice of mediation in the South African civil‐engineering industry.

The objectives of the survey were to:

1. establish the type and level of professional training and experience of mediators;

2. explore the perceptions of mediators regarding the role and function of a mediator in a construction dispute;

3. determine the procedural steps and activities that the mediators should follow during the mediation process; and

4. establish the extent of the mediators’ knowledge and use of skills and techniques associated with mediation.

The primary data were collected by way of sixty‐three questionnaires received from the two hundred six that we sent to mediators registered with the South African Institution of Civil Engineers (SAICE) and the South African Association of Consulting Engineers (SAACE), as well as from mediators listed by the Western Cape Branch of the Association of Arbitrators (Southern Africa) (AASA). The responses were analyzed using basic descriptive statistics.

The research indicates that 97 percent of the respondents possess a tertiary qualification, with qualified engineers holding bachelor's or master's degrees dominating the group (75 percent). As expected, the academic qualifications of the respondents closely correspond with the nature of the respondents’ businesses, with consulting engineers and civil‐engineering contractors making up 70 percent of the group. Because this survey sample was a non‐probability, convenience sample, the fact that the majority of these respondents have engineering backgrounds, as opposed to architectural, surveying, or legal backgrounds, does not mean that the majority of the mediators of construction industry disputes in South Africa are engineers, however. No significant differences in the views or opinions on research issues have been found to exist between those with an engineering background and those with other backgrounds.

When asked to rate their knowledge of the mediation process, 27 percent of the respondents reported that they consider it to be “average,” 35 percent consider it “adequate,” and 35 percent consider it “substantial.” The respondents reported significant amounts of experience: 29 percent have been involved in the mediation of disputes for more than twenty years, while an additional 32 percent have between ten and twenty years of experience. Their experience was cited by 95 percent of the respondents as one of the sources of their knowledge of the mediation process, followed by “workshops and seminars” (44 percent); “practice notes, journals, and other literature” (43 percent); and “formal training” (40 percent).

The findings reported in this next section relate to the most recent mediation in which the respondents had been involved. Fifty‐nine percent of the respondents reported that either or both of the parties appointed the mediator (32 percent by the parties jointly, 27 percent by a single party), while 40 percent reported that the mediator was appointed by a party other than the disputing parties, such as the president of the SAICE or AASA. Approximately two‐thirds of the mediations were initiated under the terms of a contract. In 60 percent of these cases, the clause made mediation compulsory, while for the remainder it was voluntary. Most respondents (59 percent) reported that whether the mediation was to fulfill a contractual obligation or not, the disputing parties had voluntarily chosen mediation, rather than being forced to do so because of that requirement.

A significant finding was that 41 percent of the respondents reported that the parties to the mediation had signed an agreement binding them to the mediator's opinion unless otherwise ordered in arbitration or litigation proceedings. This finding is highly significant because it seems to contradict the generally accepted view that a mediator should not make a binding decision, but instead should encourage the parties to reach a settlement by which they can agree to be bound.

Sixty‐two percent of those cases in which the parties agreed to be bound to the mediator's decision were disputes in which mediation was contractually compulsory, while in 39 percent of those cases the mediation was voluntary. No significant correlation was found between the manner in which the mediation had been initiated (i.e., by which party) and the agreement to be bound by the mediator's decision. We argue, however, that a process in which the parties agree, prior to the commencement of the mediation, to be bound by a third party's opinion or decision, as was reported in 41 percent of the cases, should not be called mediation. Terms like expert determination, contractual adjudication, or evaluative mediation, we believe, would better describe such a process.

The majority of mediations described by the respondents included the following activities (percentage of total mediation time in parentheses): preliminary or preparatory matters (9 percent), obtaining information (43 percent), problem solving (20 percent), and drafting the final decision/ opinion/agreement (27 percent). In the majority of cases, the first contact made by the mediator with the parties was personal and usually made over the telephone for the purpose of arranging a meeting to outline the procedure and program to be followed. Only 22 percent of the respondents made the first contact by communicating with the parties in writing. When obtaining information, the respondents predominantly relied on written sources of information about the dispute, either in the form of existing documentation and correspondence between the parties or from written presentations specifically drafted for the mediation. In 64 percent of the cases, oral presentations supplemented the written documentation and presentations.

Another significant finding was that, in a process aimed at assisting disputing parties to reach an agreement, only 20 percent of the total time of the mediation was allocated to problem‐solving discussions. The majority of the time was allocated to gathering information on the dispute (43 percent) and drafting the final decision/opinion/agreement (27 percent). In considering the role of problem solving during the process, it is significant that when asked how negotiation/bargaining took place, 58 percent of the respondents stated that no solution‐seeking discussions took place at all. This is because after the information on the dispute had been obtained, the onus was placed on the mediator to suggest or submit an opinion or decision, without any further involvement from the parties. Interestingly, 37 percent of the respondents stated that negotiations between the parties took place through the mediator, rather than directly between the parties. From this finding, we infer that, although the mediator facilitated the joint solution‐seeking discussions between the parties, his or her level of intervention was high. As expected, we found that face‐to‐face negotiations dominated cases where the mediator facilitated joint solution‐seeking discussions (41 percent) between the parties, but played a lesser role in mediations where the mediator offered an opinion or solution (11 percent).

A significant finding was that 25 percent of the respondents did not use any of the guidelines published by the SAICE (1994), SAACE (1993), and AASA (1992), while 33 percent used one of the three guidelines and another 33 percent used a combination of the three sets of guidelines. This suggests that, in deciding the mediation procedure, the respondents did not place great reliance on any particular published guideline, but instead relied on a combination of guidelines and their own experience. The aforementioned guidelines advise the mediator to consult with the parties on the procedure of the mediation, but it seems that many of the respondents did not consider the parties’ input regarding the process to be of much importance. The results showed that the disputants had a substantial (rated greater than “4” on a 5‐point scale) input into the determination of the procedure in only 25 percent of the cases, while in 44 percent of the cases their input was limited (rated “2” or lower on a 5‐point scale).

The respondents indicated that the parties’ willingness to participate in the mediation process and to reach a consensual settlement was important to the outcome of the mediation (78 percent). They neither, however, consider the parties’ understanding of the nature of the process to be very important (11 percent), nor did they place much value on the parties’ control of the process and its outcome (14 percent). Instead, the respondents were almost unanimous in their opinion that the mediator's expertise and authority on the issues of the dispute played the most important role in determining the outcome of the mediation process (84 percent), while the mediator's expertise in the process of mediation was of lesser importance (40 percent). These opinions correspond to the respondents’ attitudes regarding the parties’ level of input into the determination of the mediation procedure; 71 percent gave this input a value rating of “3” or less on a 5‐point scale. In the majority (81 percent) of the cases, the parties did not make use of witnesses, and only 20 percent of the mediators sought outside advice.

Of the four mediator roles that we suggested to the respondents, “to provide professional expertise on the content of the dispute and persuade the parties that the mediator's proposed settlement is fair, reasonable, and in everyone's best interests,” was the most popular choice (32 percent). “To provide professional expertise on the content of the dispute and persuade the parties that the mediator's opinion of the outcome would be within the range of a likely court/arbitration ruling” was the next most popular (28 percent). This was followed by “to facilitate a constructive dialogue between the parties and encourage the parties to negotiate their own settlement” (21 percent). The least popular choice was “to consider the information and evidence gathered, and give a decision based thereon, for or against the claimant” (18 percent).

We analyzed these findings from three different perspectives. First, we looked at them to examine what role technical experience and expertise played in the process, and found that evaluation of the dispute based on subject expertise was considered to be important 79 percent of the time, with facilitative skills being the determining factor in solving the dispute for the remaining 12 percent.

Second, we looked at the mediator's actions to move the parties toward settlement. We found that persuasion was used 60 percent of the time, followed by facilitating discussion (21 percent) and decision making (18 percent). Third, we looked at the mediator's approach to the parties’ interests and rights. In this case, interest‐based approaches were used 53 percent of the time, and rights‐based approaches used 47 percent of the time.

These findings reflect the respondents’ diverse perceptions of the mediator's role, although the evaluative approach seems to be generally favored over the facilitative approach. This proposition is supported by our analysis of the mediation functions that the respondents consider to be important.

The analysis of the perceived importance of ten suggested functions of the mediator indicates that a strong majority (75 percent) of the respondents perceived “the development and preservation of the trust and confidence of the parties in the mediator's role” to be very important. A smaller majority (61 percent) of the respondents’ reported that “evaluating the dispute and giving a reasoned opinion and decision” is very important. “Educating the parties about the mediation process” was not considered as important as the other functions, with only 40 percent of the respondents rating this function as very important.

Although the majority of the respondents responded that the mediator should possess skills specific to the mediation process and different from those required by an arbitrator or an adjudicator, the results to an open‐ended question indicate that the respondents do not often use specific mediation skills and techniques, but rely mostly on their communication skills. They also reported that in recent mediation cases, they relied on the authority of their positions and personal attributes and attitudes to develop the trust and confidence of the parties. The respondents reported that they use communication techniques to gather information (careful reading and listening), while negotiation techniques are limited to “promoting reality by predicting outcomes” and “assessing the strengths and weakness of the parties’ cases with a view to proposing a settlement.” Generally, the respondents consider “subject‐matter expertise” to be more important than expertise in the mediation process, thus asserting their evaluative approach to mediation.

Finally, when asked how frequently the dispute was actually settled using mediation, 62 percent of the respondents reported that disputes were settled by mediation in more than 80 percent of the cases they had been involved in, (meaning, the final resolution of the dispute was achieved without going to arbitration or litigation). This compares well with Watson's (1996) finding that 85 percent of his cases were resolved through mediation.

Our survey results indicate that the construction‐industry mediators in our sample do not generally assist the parties to achieve their own settlements. Instead, these mediators focus mainly on collecting information regarding the dispute, and formulating a solution. These mediators’ knowledge and utilization of specific mediation‐process skills and techniques were found to be limited.

Mediation generally seeks to assist disputing parties to reach an agreement and settle a dispute finally and conclusively. The goal is to avoid the expense (in both money and time) and ill will that is often generated by arbitration and litigation. The belief that an agreement reached between two disputing parties, where the parties believe such an agreement is in everyone's best interest, is a lasting agreement that preserves amicable and long‐lasting relationships, is one of the ideas that underlie the process of mediation. This belief is compatible with the South African cultural philosophy of ubuntu. Therefore, we argue, it is preferable that agreements not be imposed on the parties, but instead that the parties should be skillfully led to such an agreement through a delicate process of perceptual change and understanding. It is under such circumstances that mediation results in final and conclusive settlement of a dispute.

Conflict and disputes are common in the South African construction industry, not least because of the tendency for adversarial relationships between contractors and employers that are formed under the traditional procurement system. Every effort should be made to reduce the levels of conflict and dispute. The settlement of disputes by mediation, practiced in accordance with the principles and objectives generally recognized as constituting the process, has a major role to play in improving the climate of the industry and improving its sustainability. Indeed, the more wide‐spread adoption of properly practiced mediation would be more consistent with the ideal of nurturing South Africa's innate “ubuntuness” than the predominant use of more adversarial methods of dispute resolution.