Sailing between Scylla and Charybdis: Mayelane v Ngwenyama

Mayelane v Ngwenyama 2013 4 SA 415 (CC) is arguably the most important judgment concerning the recognition of customary marriages in recent times. This article attempts to unpack some of the many issues that arise from the case, namely: (a) the practical difficulties associated with ascertaining living customary law and the problems of identifying legal versus social norms; (b) the meaning of consent as a requirement of a customary marriage; (c) the implications of the case for equality between multiple wives in a customary marriage, and as between wives across customary marriages of different cultural traditions; and (e) the implications of the case for equality considerations more broadly. While the authors sympathise with the court in respect of the complex decision before it, it questions the Court's method and result, specifically for the equality rights of a second (or further) "wife" in a Vatsonga customary marriage. The authors suggest that the issues should be put to democratic deliberation by the legislative arm, rather than leaving courts in the unenviable position of having to decide these matters.

written contract which will regulate the future matrimonial property system of his marriages". 4 Both the court a quo and the Supreme Court of Appeal based their findings on the latter contention, without interrogating the former averment that the requirements of custom had not been complied with. As noted in both courts a quo, the problem with section 7(6) (and which has been the subject of debate among commentators) 5 is that it does not set out the consequences of a failure to obtain a court-approved contract. It was clear from the evidence before the court that Mr Moyana had not made such an application in respect of his 2008 marriage, and it was the failure to do so that occasioned the dispute. In finding for Ms Mayelane (the first wife), Bertelsmann J in the court a quo found that the failure to approach a court for approval invalidated the second marriage. The court based its finding on the preemptory wording of section 7(6), but was swayed more trenchantly by the likely prejudice that would be suffered by the first wife due to the non-disclosure of the subsequent marriage: 6 The failure to comply with the mandatory provisions of this subsection cannot but lead to the invalidity of a subsequent customary marriage, even though the Act does not contain an express provision to that effect. Cronje and Heaton argue in South African Family Law 2ed at 204, that the court's intervention would be rendered superfluous -which the legislature could not have intended -if invalidity did not result from a failure to observe ss (6).
The learned judge continued at para 25: A further argument … arises from the peremptory language of the provision: the word "must", read with the provisions of subsection (7)(b)(iii), empowering the court to refuse to register a proposed contract, indicates that the legislature intended noncompliance to lead to voidness of a marriage in conflict with the provision. 7 4 The judgment in the court a quo is reported as MM v MN 2010 4 SA 286 (GNP). para 32, where it was held that "language of a predominantly imperative nature such as 'must' is to be construed as peremptory unless there are other circumstances which negate this construction". As hinted at above, Bertelsmann J found that: [t]he most persuasive consideration must however be the gross infringement of the first or earlier spouses' fundamental rights: to respect of their dignity, physical and emotional integrity; their right to protection from abuse -in this instance both emotional and economic or material; their right to be treated on an equal footing with their husband, as decreed by the Act; their right to equal status as marriage partners, arising from the Act; their right to marital support from their husband; and their right to marital intimacy and trust, which rights flow naturally from those guaranteed by the Act and the Constitution. A gross infringement of these rights would be committed if the husband were to be allowed to enter into a further marriage without their knowledge and acquiescence. 8 On appealing the ruling to void her marriage to the Supreme Court of Appeal, Ms Ngwenyama (the "second wife") succeeded. 9 The SCA (per Ndita JA) found that section 7(6) was not peremptory and that it could not have been the intention of the legislature to effect so fundamental a change to the customary law of polygamy by subjecting the validity of a second marriage to prior consent by a (secular) courtwhich could withhold it. The court also highlighted the discrimination that would be visited upon the second wife if Judge Bertelsmann's reasoning was upheld, as she would not have her marriage recognised because of her husband's omission. 10 The court then came to the conclusion that section 7(6) had nothing to do with the validity of the marriage (which was the proper concern of section 3(1) of the Act), but merely affected the proprietary consequences of the polygamous marriage. 11 8 Minister of Environmental Affairs and Tourism v Pepper Bay Fishing (Pty) Ltd; Minister of Environmental Affairs and Tourism v Smith 2004 1 SA 308 (SCA) para 27. As an aside, it is interesting that Bertelsmann J ascribes marital intimacy and trust as "rights which flow naturally from those guaranteed by the Act and the Constitution". It implies a western form of consortium omnis vitae which, we believe, is not axiomatic, especially if one considers the original foundation of customary marriages as a union between families. Bonthuys 2007 SAJHR 534 argues that the customary concept of marriage is "flexible enough to accept different family formations" in demonstrating that the Civil Union Act is a poor cousin to the Marriage Act in South Africa. Woman-to-woman marriages with no "sex" is an example of such a marriage. See Bertelsmann J's decision in his own division. In coming to this conclusion, Moshidi J stated that once there is a valid subsequent customary marriage, the second wife also acquires certain rights. He further questioned why the second wife should be penalised or prejudiced on account of the failure of the husband to comply with s 7(6).
Importantly, and in a separate judgment which was concurred by all, 12 Ponnan JA held that non-compliance with 7(6) would not invalidate the second marriage, and went further to find that where the husband had not complied with the provision requiring court sanction of a second marriage, this marriage would have to be one out of community of property: "It plainly cannot be a marriage in community of property as that would imply the existence of two joint estates, which it is clear cannot co-exist." 13 Notwithstanding the fact that the section 3(1)(b) requirements for a valid customary marriage were not the locus of debate in the High Court and Supreme Court of Appeal decisions, the Constitutional Court -on the final appeal discussed heregave two sets of practice directions 14 to the litigants and amici 15 prior to the hearing.
The Court requested the parties to deal specifically with the issue: "whether under Tsonga customary law the first wife's consent was required before the husband could enter a second marriage". This question redirected the focus of the partiesand the dispute -from section 7(6) to section 3(1)(b) of the Act. Section 3(1) provides for express validity requirements in section 3(1)(a), namely majority age and the consent of both parties, but importantly it introduces the further requirement in section 3(1)(b) that the marriage must be "registered and entered into or celebrated in accordance with customary law". The logical inference to be drawn from these directions was that if the appellant could establish that Vatsonga 12 The strange way in which the judges concur in each other's judgment is an issue in and of itself.
Ndita JA gives the main judgment and order and ends with saying in Ngwenyama para 27:"It remains to mention that I have had the privilege to read the concurring judgment of Ponnan JA, and I find nothing different from what I have already said. It is substantially a repetition of what I have said, except for what is contained in paras [5] and [10]. For that reason, I concur in it." Then Ponnan J gives his judgment where he uses the analogy of the consent of a minor in terms of the Marriage Act (para 5) and the finding that a subsequent marriage would be out of community of property (para 10). There is no reference to a different order, as he has concurred in Ndita's order. Then finally, Mthiyane DP says at the end in para 39: "I have had the privilege of reading the judgments of my colleagues Ndita AJA and Ponnan JA. I concur in both, save for para [27] of the main judgment." It is not clear what Mthiyane DP is disagreeing with in para 27. One presumes it is that Ponnan J's judgment is not substantially a repetition, but this is by no means clear. For another interesting but more fraught "concurring" issue in the SCA, see Phodiclinics (Pty) Ltd v Pinehaven Private Hospital (Pty) Ltd 2011 4 All SA 331 (SCA). 13 Ngwenyama para 38. 14 Practice directions given on 1 August 2012 and 25 February 2013. 15 The Women's Legal Centre had participated as an amicus in the SCA, and again applied (and was admitted) as an amicus in the Constitutional Court, together with the Commission for Gender Equality and the National Movement for Rural Women in a combined brief. law required the first wife's consent (as set out in section 3(1)(b)) as part of her custom, it would follow that this would render the 2008 marriage void for want of compliance with section 3(1)(b) (regardless of compliance or not with section 7(6)). 16 On the facts, it was common cause that no consent had been sought from or given by Ms Mayelane to her husband's taking a second wife.
In heads of argument filed with the court, the amici's representations can be broadly summarised as follows: the Women's Legal Centre Trust supported the findings of the SCA, and argued that the High Court's approach was "unduly and unnecessarily harsh" on subsequent wives whose unions do not have the consent of the first wife (the decision deprived the subsequent wives of important "legal and constitutional protections", they argued). The Rural Women's Movement and the Commission for Gender Equality (in a joint brief) submitted that insufficient evidence was presented at the High Court to entertain the question on whether or not prior consent from the first wife was part of the customs of the Vatsonga people. Accordingly, they argued the matter should be remitted back to the High Court for reconsideration. 17 As is apparent, the court decided against remitting the matter back to the High Court because of the important interests involved and the need for finality. Instead it called for, and considered, additional evidence itself on the question raised in the practice direction. Ultimately the judgment was then centred on this question (viz if Vatsonga customary law required the consent of the first wife in order for the second marriage to be valid). In turn, it was necessary to establish the content of the custom in order the determine if the requirements of the applicable statute had been met, ie if the conditions in section 3(1)(b) had been satisfied. The majority (six justices), found that the Vatsonga law requires the first wife be "informed" about the 16 Himonga and Pope 2013 Acta Juridica 321 correctly point out that the Court's reliance on and reference to s 3 of the RCMA was in fact unnecessary since the first marriage pre-dated the Act. That being so, the essential question (establishing the content of the custom for marriage) remains the same. second marriage. 18 Two separate dissenting judgments, a single judgment penned by Zondo J and another supported by three judges, reflect the divided responses to the issue. While the Constitutional Court "endorsed" the SCA's interpretation of section 7(6), 19 it overturned the order of that Court, preferring to deal with the matter in terms of the first basis mentioned originally in Ms Mayelane's application, that is, whether Vatsonga custom requires the consent of the first wife. As already said, in order to determine this point the court called for and considered largely contradictory evidence on the custom (although the court felt otherwise, a matter to which we will return to later). 20 Justice Froneman, writing for the majority, was at great pains to point out that evidence of consent was a question of law rather than of fact. As such, it could not simply be settled on the papers, and on the normal rules of civil procedure. 21 This conclusion also enabled the court to find that the custom, as a matter of law, did indeed require that the first wife needed to be informed about the second marriage. 22 Given that she knew nothing of the second marriage, the court found that the second marriage was invalid. In addition, the court determined thatprospectively -the lack of consent by a first wife in Vatsonga customary marriage would invalidate any further customary marriage. The court explicitly sets out that this "development" of Vatsonga custom was "in accordance with the demands of human dignity and equality" set out in both the Constitution and the RCMA itself 18 Ngwenyama para 87. 19 Ngwenyama para 41. 20 See s 3.1 below. 21 Ngwenyama para 47. This point was made in obvious contradistinction to the minority judgment of Zondo J, who felt that the matter should have been decided on the papers (viz records from the High Court and SCA) and that the new evidence was contradictory and therefore ran afoul of Rule 31 of the Court Rules, which require new evidence to be introduced only if it is common cause or otherwise incontrovertible; or is of an official, scientific, technical or statistical nature capable of easy verification. As an aside, it is hard to understand why the Court was prepared to accept the affidavits in casu in terms of Rule 31, but reject a social science study by an amicus (CALS)  (and the issue of the "development" of customary law as addressed in this matter is also discussed further below). 23

Proof of living customary law
As is apparent, this dispute was aired in three different courts at all three levels of the South African judicial system, with all three courts arriving at different conclusions. Moreover, in the final Constitutional Court decision, only a bare majority of the justices could agree on the issue, with two justices concurring with their decision on a different basis and a dissenting judgment penned by Jafta J and concurred to by Mogoeng and Nkabinde JJ. This emphasises the difficulties and challenges that face courts when dealing with the application and enforcement of an undocumented "living" law such as customary law. Langa DCJ (as he then was) captured this problem in the Bhe 24 matter when he commented that "[t]he difficulty lies not so much in the acceptance of the notion of 'living' customary law … but in determining its content and testing it, as the Court should, against the provisions of the Bill of Rights". 25 It is well known that the legislature attempted to ameliorate this problem when it amended the Law of Evidence Amendment Act 26 even prior to these cases in 1988. 27 The amendment provided that all courts may take judicial recognition of indigenous law. However, and as the Mayelane court (and other courts) have recognised, taking judicial notice has its own difficulties, not the least of which are its unwritten nature, as well as the tendency of the civil courts to view customary law through the prism of legal concepts that are foreign to it. 28 23 Ngwenyama para 75. The application of section 1 of the Law of Evidence Amendment Act is possible only where the customary law is "reasonable and certain". We argue that the issue of consent in Mayelane was anything but certain. In these circumstances, the court correctly called for evidence (as is permitted under the Law of Evidence Amendment Act), 29 but how the evidence was dealt with leaves one with a sense of unease. The evidence, in the form of affidavits, can be split into four categories (as the court itself found): 30 (a) evidence from three individuals in polygamous marriages; (b) evidence from one advisor to traditional leaders and a male commissioner in the Commission on Traditional Leadership Disputes and Claims; (c) evidence from three traditional leaders (one of whom also falls into the first category above); and (d) two academic experts (from different disciplines), who purportedly drew conclusions from available primary material.
In the first category, two men (Hosi Bungeni and Mr Shirinda) and one woman (Mrs Rikhotso) confirmed that, while the consent of the first wife was required for the validity of a subsequent marriage in Vatsonga custom, the legitimacy of the children and rights of inheritance would not be affected. In this evidence, at least one of the individuals indicated that while consent should be sought, if the first wife did not have a "good reason" (not explained) for refusing her consent, then the elders would impress the point upon her and that issues would usually be resolved that way. 31 In the second category of informants, the advisor to traditional leaders: Mr Mayimele, indicated that it is the husband who makes the decision. According to the first deponent in this category, the wife is usually involved only where her daughter's lobolo is used as lobolo for the subsequent wife. confirmed that where the decision to marry comes from the husband 33 or from his relatives, the first wife must be informed, but if she disagrees with his decision and his family supports him, he may proceed without her blessing. It is only where the husband acts without his relatives' support that the second wife will be regarded as a concubine.
In the third category (that of traditional leaders), a headman (Acting Headman Sithole) reported that the wife is "always" expected to agree to a husband's taking a second or third wife. If she unreasonably (again not explained) withholds consent she would be sent to her parents homestead to reconsider. If she returns to her husband and remains "unreasonable" the husband may marry without consent or divorce her. A second headman (Headman Maluleke) indicated that the consent of the first wife is not necessary, and that the openness of the lobolo negotiations (which are a precondition to the marriage) makes it unlikely that the first wife will not know about the impending marriage. 34 In the final category, the experts came to different conclusions. An anthropologist with extensive research in the field of Vatsonga customary law (Prof Boonzaaier) indicated that informing or obtaining consent is not a requirement for validity. A law academic (Dr Mhlaba) indicated that it is uncertain what the consequences would be regarding the validity, but he confirmed that where the wife does not consent, the families become involved in order to resolve the matter.
In taking into account this evidence, the court then commented that it could "safely find the following": (a) although not the general practice any longer, VaTsonga men have a choice whether to enter into further customary marriages; (b) when VaTsonga men decide to do so they must inform their first wife of their intention; (c) it is expected of the first wife to agree and assist in the ensuing process leading to the further marriage; (d) if she does so, harmony is promoted between all concerned; (e) if she refuses consent, attempts are made to persuade her otherwise; (f) if that is unsuccessful, is said to be to restore the dignity and integrity of the traditional leaders and traditional communities and the entire institution of traditional leadership in South Africa. 33 The male commissioner indicated that the decision could also come from the wife, but obviously this poses no problems. 34 Mayelane para 57. the respective families are called to play a role in resolving the problem; (g) this resolution process may result in divorce; and finally, (h) if the first wife is not informed of the impending marriage, the second union will not be recognised, but the children of the second union will not be prejudiced by this as they will still be regarded as legitimate children.35 Is this a fair finding (especially the finding in para (h)? A tabular-format of the evidence as described by the court looks as follows: Does the first wife need to consent to her husband's intention to marry? 36

Yes
No Uncertain 4, subdivided into: Mayelane para 61. Our emphasis. Given the disappearing language of legitimacy in South Africa and worldwide (recognising that the word has pejorative consequences), the amici and court's reference to this status is somewhat unfortunate. See Schafer Child Law 26-27. 36 As mentioned earlier, the practice directions requested the amici to provide evidence to answer the question of whether or not the consent of a first wife was a requirement (para 53, fn 51 of the judgment). The practice directions further stated that the amici should attempt to ascertain the "manner and form" of this consent. See the discussion infra regarding the finding of the court as to "informing" the wife versus her "consenting". 37 The affidavit of a Hosi Nkanyani from the Vhembe district is referred to in both the judgments of Zondo and Jafta JJ (Mayelane paras 119 and 137 respectively). It appears as if his affidavit forms part of the submission by the CGE and NMRW. Notwithstanding, this affidavit was not referred to in the majority judgment and is therefore discounted for the purposes of this table. 38 See Zondo J's judgment at para 116, which posits that the case study upon which Boonzaaier relied to support his opinion actually supported the view that informing the first wife was necessary. 39 Mayelane para 56. the wife will be strongly impressed upon to accede.
While it is unwise to determine the content of customary law through a "numbers game", it seems that the evidence before the court simply did not support its finding as set out above.
More importantly, and in the light of the disagreements amongst informants, the court also failed to ask the antecedent question: if there is a social normative practice of informing the first wife of the intention to conclude a further marriage (which does not look to be set in stone), then at what point does it become the basis of a legal norm with dire consequences regarding the validity of the subsequent marriage? It appears that the court's view of the evidence before it presupposed that a requirement that the first wife be informed would be tantamount to a legal norm, such as those traditionally accepted legal norms requiring the payment of lobolo (the bride price) and that of ukumekeza (the formal handing over of the bride). 40 If courts are not alive to the finer distinctions between behavioural norms, there is a concern that "law" and "customary law" will lose any distinctive meaning: everything will be swallowed up to become law. 41 Here the dissenting judgment appears to provide further ballast, as it seems that the custom of one community within a cultural group (ie in rural areas) may not be echoed by others belonging to the same clan (eg with more access to urban and peri-urban areas to an affidavit alluding to a requirement that the first wife be informed, the court is not able to interrogate the fuller meaning of such an act for that community as a whole, and as has been the case in some other courts, the court has glossed over the normative status of the social practice asserted by the parties. 47 As mentioned by Bekker and Koyana 48 in the context of succession, "incidents of living customary law cannot without more ado be elevated to a general rule of law." In setting out the process of recognising customary law, they argue that a court must find that a fixed line of behaviour is followed by more or less a constant group of persons for a certain period, and in particular, "a custom, in order to be law, must be commonly believed to be obligatory". An example of where this is put in doubt, is provided by Mabena v Letsoala. 49 The case dealt with the question of whether a customary marriage had been validly celebrated where the groom had negotiated the lobolo with his prospective bride's mother, in contrast to the generally accepted rule that such marriages require the 44 This relates to the desirability of formal law taking as legally authoritative some pre-legal normative practice, without more inquiry into the matter. 45 Bennett 2009 Am J Comp L 13. See also Himonga and Bosch 2000 SALJ 321. 46 The repeated references to "unreasonable" refusal, getting elders to approach the first wife, and sending the wife back to her parents imply some effort at achieving "peace for peace's sake" rather than any autonomous decision making by the wife. See below. 47 This is important in the context of understanding that marriage in customary law cannot be viewed simply in linear terms: certain acts (such as the so-called consent issue) could "suggest a marriage or movement towards such a state" (Griffiths In the Shadow of Marriage 56). The legal anthropologist, Griffiths, makes this point when describing the difficulties in establishing the requirements for a customary marriage, and the fact that customary marriage is a process rather than a single event. 48 Bekker and Koyana 2012 De Jure 571. 49 Mabena v Letsoalo 1998 2 SA 1068 (T) 1070, 1074. groom and his guardians to negotiate with the bride's male guardian. In this matter the respondent said that although the parties followed Pedi custom: "My people and I, we do not engage in these customary traditions. We did it as it pleased my mother." The court did not attempt to ascertain if the so-called rule in issue was not simply the practice of two people entering a contract, being the woman and her future son-in-law. As Bennett explains, the court merely referred to the respondent's testimony and analogous situations described in a South African Law Reform Discussion Paper, and then concluded that the case was not an isolated one. In Bennett's view, the court had virtually no evidence of a rule of living law. … [The woman's] action might have been purely idiosyncratic and not in keeping with a general pattern of behaviour: she might have been a particularly strong-willed character, having no truck with old-fashioned, sexist behaviour. Or her action might have been the result of temporary conditions: adult men were away from home at work, thereby compelling women to assume male roles for the time being. Was there any sense of legal obligation in this practice? 50 It is telling, for example, that one deponent in Mayelane (Headman Maluleke) in an additional affidavit filed by the amici in the case stated that: … the prospective bridegroom should inform his existing wife about his intentions to marry another wife. He informs her so that she should not be surprised in seeing another wife. It is not a requirement to even advise as to the identity of the prospective subsequent wife. Whether she gives her consent or not, the prospective bridegroom will proceed with his plan to marry another wife. 51 In this version, the evidence suggests that the so-called requirement to inform the first wife is much more akin to a housekeeping rule than to a legal norm to be adhered to in the face of the invalidity of the second marriage.
Finally, it is interesting (and the court does not raise this issue) that it is mostly the traditional leaders and the experts (bar one) who doubt that informing the first wife is a legal requirement. 52 While the court was correct to indicate that their finding was one of law and not of fact, it could be argued that the court should have given more weight to the testimony of the so-called experts (from a law-making or law-telling 50 Bennett 2009 Am J Comp L 13, 18. This concern is also shared by Bekker and van der Merwe 2011 SAPL 125. 51 Recorded in Mayelane para 122 of the judgment. Our emphasis.

52
Mayelane paras 54-59. perspective) who supposedly deal with these issues as part of their responsibilities or research, 53 than to the accounts of the personal experiences of less authoritative individuals.

Informing or requiring consent?
While it is conceded that the court was correct to emphasise that consent in the context of customary marriage must not be seen through the prism of the common law of contract, 54 it is hard to understand in what way consent can be interpreted otherwise -for the purposes of legal regulation -than as some form of conscious acquiescence. We believe the court's finding that the provision of information to the first wife -rather than requesting her consent -was the custom is somewhat disingenuous, since it is clear from some of the informants that even when pressure is brought to bear by relatives upon a recalcitrant first wife and she remains steadfast, the marriage can proceed. It appears then that the Court missed an opportunity to set a minimum threshold for consent (or at least to establish its meaning in a customary context if there are more wives? Would the consent of the second and third wife be required for the valid conclusion of subsequent marriages? Or is the first wife alone privileged in this regard? In addition, would a first wife in another type of customary union be required (or entitled) to go to court to ask for the same consent "development" to apply in her custom? While the court avoided these questions, given the specific factual matrix before it, the avoidance leaves these questions unresolved and consequently impacts on women's access to justice in ascertaining the legal position of their relationship, which in turn affects their financial security. 58 It may be ironic that in this situation the application of the "official customary law" of the past 59 GN R200 of 6 February 1987. Section 2(d) sets out that: When any deceased Black is survived by any partner-(i) with whom he had contracted a marriage which, in terms of subsection (6) of section 22 of the Act, had not produced the legal consequences of a marriage in community of property; or (ii) with whom he had entered into a customary union; or (iii) who was at the time of his death living with him as his putative spouse; or by any issue of himself and any such partner, and the circumstances are such as in the opinion of the Minister to render the application of Black law and custom to the devolution of the whole, or some part, of his property inequitable or inappropriate, the Minister may direct that the said property or the said part thereof, as the case may be, shall devolve as if the said Black and the said partner had been lawfully married out of community of property, whether or not such was in fact the case, and as if the said Black had been a European (Our emphasis).

Parity betw een w ives across custom ary marriages of different cultural traditions
The SCA's interpretation of section 7(6) was (in its own words) endorsed by the CC. 62 However, the CC overturned the SCA's order due to the finding that the deceased had not complied with section 3(1) as required in the RCMA. Strictly speaking this means that the SCA's finding relating to section 7(6) is no longer binding as a ratio decidendi, but it remains highly persuasive given the Constitutional Court's endorsement of its reasoning. So far so good. However, the finding that the second marriage concluded without prior court approval of the matrimonial property regime to govern the marriage(s) will therefore have to be a marriage out of community of property, while seemingly pragmatic, is in our view problematic for those customary marriages where informing the first wife is not (or is not yet) a legal requirement for the validity of the second marriage.
The obvious concern is that parity between wives across different customs must run on the issue of the proprietary consequences. In the Constitutional Court decision of Gumede v President of the Republic of South Africa, 63 the Court found section 7 (1) and (2)  Gumede v President of the Republic of South Africa 2009 3 SA 152 (CC). 64 We use the term de facto here since pre-Act all customary marriages were subject to customary rules which were neither in nor out of community of property, but were based on patriarchal principles. Given the perpetual minority status of all women under customary law, they did not have legal capacity. As a result women could not own, acquire or dispose of property. In effect then, one can characterise the marriage as out of community of property.
features between the two cases. Gumede dealt with a de facto monogamous marriage where the dignity and equality rights of only one wife and one husband had to be considered. However, the result is that courts have a discretion to impose an in community of property matrimonial regime for women in pre-RCMA marriages, but may not do so for second and further wives where section 7 (6)  The finding that the second marriage will be out of community of property also poses implementation problems. One would expect an ante nuptial contract regulating the second marriage. 65 This is unlikely to be the case. Nor does it seem that the accrual system could be automatically applied, given its questionable viability in a polygamous setting. 66 So imagine that a husband and a wife are married in terms of customary law after the Act's inception (and therefore in community of property). The husband obtains the consent of the first wife for the second marriage, but does not approach the court in terms of section 7(6). The husband then dies intestate with an estate valued at R400 000. How will the estate devolve?
Given the SCA judgment, the second marriage in this scenario is likely to be out of community of property. If one applies the Intestate Succession Act, 67 the first wife must get half the estate (R200 000), and then each surviving spouse must inherit R125 000 or a child's share, whichever is the greater. 68 It is assumed that in this scenario then, the first and second wife will each receive R100 000, and any children from both marriages will get nothing. There is no space in this scenario (as there is 65 Rautenbach and Matthee 2013 http://www.givengain.com/cgi-bin/giga.cgi?cmd=cause_ dir_news_item&cause_id=2137&news_id=115467&cat_id=1584#.UmfJrUpBvGg 9. 66 Heaton South African Family Law 213 indicates that the application of the accrual system is problematic in the context of discussing the matrimonial property systems that the s 7(6) contract can provide for. 67 Intestate Succession Act 81 of 1987. 68 in a divorce setting) for a court to "make any equitable order that it deems just" 69 given the duration, support etc of each wife. One could argue that an unfair position could be ameliorated by the fact that the children are able to claim maintenance against the estate, and that one (or both) wives may use the Maintenance of Surviving Spouses Act for a claim for maintenance. 70 Any claim under the latter Act has to be settled before the estate is divided among the heirs, and shares the same order of preference as a claim of a dependent child. 71 So it may be that there is competition between the wives for the half share of the deceased's estate which could be reduced dramatically with maintenance claims. In terms of implementation, there is also the issue of when the out of community of property marriage regime commenced: it seems that it would have to be at the date of the marriage, but given jurisprudence about retrospective application in terms of section 21 of the Matrimonial Property Act, 72 this is yet another difficulty to contemplate if one is a third party creditor. 73 The practical implementation issues relate to the bigger issue: under a joint property regime such as community of property and automatic sharing, substantive equality simply cannot be attained where more than one marriage subsists at the same time.
The regime is predicated on no more than two parties to a marriage. Van Heerden and Sloth-Nielsen predict that this fact is going to lead to ongoing litigation in the customary law sphere, as the concept of equality is "blurred at the edges". 74 69 This order is possible following s 8(4)(a) of the RCMA, which allows a court to redistribute assets of any customary marriage regardless of when they were concluded and regardless of what matrimonial property system was operational.

Prospectivity versus retrospectivity: the bell tolls for some 75
It is trite that the Constitutional Court has emphasised the need for effective remedies. 76 The purpose then of a constitutional remedy is to grant effective relief to the litigants before the court and to similarly situated people. 77 Given this principle, we believe that there is an internal clash of logic inherent in the court's view that the consent of the first wife will be required for the validity of Vatsonga marriages, but only for marriages concluded after the date of publication of the judgment, ie prospectively. 78 The court expressly premises this on the inequity that may result for women who entered into a customary marriage without knowing that consent was a requirement for validity. This is then the only time that the court contemplates the position of the second "wife", and not in relation to Ms Ngenyama. The court expressed the opinion that it would be unfair "to deprive wives in a further marriage of the protection that recognition of the validity of their marriage under the Recognition Act would bring". 79 How then does this not apply to Ms Ngwenyama, we ask? Further, although the court purports to "develop" the customary validity requirements of Vatsonga marriages, the court is equally at pains to point out that "even before its development in this judgment", customary law required that the first wife be informed of her husband's impending subsequent marriage as if the requirement was there all along. The court, at para 43, makes the point that this would be the first time that the court has had to engage in the incremental development of customary law as envisaged in section 39 (2)  Another interesting issue that arises in the context of the remedy given is that there is some ambiguity as to whether the order applies to all customary marriages, or only to Vatsonga customary marriages. This has obvious implications for discerning who are the very "similarly situated people" discussed above. While the Court implies that its order applies to all customary marriages systems in its statement at para 86: "Our order makes it clear that the general requirement of consent operates only prospectively, to customary marriages entered into after this judgment has been published…", this implication is put into question by its preceding statement that the court "is not able to determine what the position in customary law systems other than XiTsonga system is". Himonga and Pope believe that the Court ruling applies uniformly to all customary law "in a sense". 81

Conclusions
In analysing the course the court charted, we applaud the judges for making it through to the other side largely unscathed. As noted by Himonga and Pope, the Court's approach represents a good attempt to close the gap in customary law jurisprudence, specifically related to determining the content of customary law. 84 However, it should be more readily acknowledged that compromises had to be made, and that "sailors" were lost along the way, to return to the metaphor with which we started. A more honest approach would have been to put aside the fiction that the judges do no more than "speak to the law", 85 and to acknowledge that a privileged position was to be accorded the first wife by the Court and not necessarily by the actual custom.
A related conclusion is that, while section 3(1) of the RCMA is laudable in its conception and its attempt to be flexible enough to adapt to living law, the Mayelane case shows how courts have had to enter the law-making arena in the family domain where the current environment (legislation is but one form) provides inadequate direction. While we can accept that this does and must happen in a constitutional democracy operating in a plural system, it is better that these matters should be subject to democratic deliberation and dealt with in a holistic way. We cannot forget that the courts have to make decisions ("law") on the narrow facts before them.
While it may be appropriate in many instances for courts to make decisions affecting more than just the parties before them (as the Constitution may require) 86  With regard to the consequences of Mayelane for equality, the resultant non-validity of the marriage of the second wife (whose husband failed to inform the first wife of his intention to marry) ironically puts the second wife as dependent on the vagaries of male behaviour to determine her standing as under any formal system of patriarchy. This is an invidious position indeed.
he argues that certain forms of minimalism can be "democracy-promoting" in that it ensures that certain important decisions are made by democratically accountable actors (Sunstein One Case at a Time 5). Second, he argues that -especially in a pluralistic society -courts should be making minimalist decisions when dealing with complex issues about which people feel deeply and on which there is disagreement (Sunstein One Case at a Time 5). 89 As opposed to Parliament who will have time, researchers and drafters over a long period putting legislation together. 90 See s 3.5 above. 91 Viz access to justice issues discussed above. It is not just wives who will be uncertain. Pension