In the absence of formal legal recognition, domestic partners (i.e. persons who cohabit outside of marriage) are required to regulate the consequences of their relationship by utilising alternative regulatory measures and remedies which are, for the most part, inadequate. The traditional justification used to differentiate between domestic partners and spouses is known by some as the “choice argument”. The choice argument is based on the rationale that persons who choose not to marry cannot claim spousal benefits. It understands choice narrowly as it only takes into account an objective legal impediment to marriage. As such, it has been the driving force behind the non-recognition of heterosexual domestic partnerships. Same-sex domestic partners have, however, been able to sidestep the choice argument considering that their sexual orientation has until recently been an objective legal impediment against marriage. According to the majority of legal commentators the enactment of the Civil Union Act 17 of 2006 has removed the objective legal impediment against same-sex marriage. As such, they argue that the choice argument should now be applied to both heterosexual and same-sex domestic partners equally. The Constitutional Court has, however, held that unless the legislature intervened the benefits accrued by same-sex domestic partners prior to the enactment of the Civil Union Act 17 of 2006 should be available to them exclusively. As the legislature has not yet done so, the legislature does not appear to view the choice argument as being equally applicable to heterosexual and same-sex couples. Taking into consideration the choice argument’s narrow understanding of choice, together with the possible unfair discrimination caused by its application, an alternative theoretical basis for the future recognition and regulation of domestic partnerships has to be found. Three possible solutions will be investigated in this study, namely, the model of contextualised choice, the function-over-form approach, and finally, the Smith model. Because of the invasive effect of the latter two approaches, the study advocates for the adoption of the model of contextualised choice. If adopted it will imply that the subjective considerations of domestic partners will be taken into account and they will be afforded a minimum degree of protection based on need. If this approach is adopted it must be determined to what extent it is supported in proposed legislation. Accordingly, it has to be investigated whether proposed legislation provides domestic partners with need-based claims while still upholding the established differences between domestic partnerships and formalised relationships. It is ultimately concluded that the proposed legislation will have the effect of blurring the differences insofar as registered domestic partnerships are concerned. The reason for this is that such a partnership comes into existence through a public expression of the partners’ commitment and, as such, does not really fall within the ambit of the definition of a domestic partnership in the narrow sense of the word. With regard to unregistered domestic partners, it is concluded that the proposed legislation goes too far in protecting unregistered partners’ proprietary rights (even if only on an ex post facto basis) as these claims are not based on need. As such, it is recommended that the proposed legislation be redrafted. If not redrafted the proposed legislation can possibly have the effect of not only infringing on the autonomy of one or both of the partners but also create a regulatory system which does not fully appreciate the differences between marriage and domestic partnerships.