To summarize the foregoing, male patrilines constituted the vertical structure within Makassar aristocratic genealogical space. Women of various origins entered into the structure with increasing frequency towards the top. The superfluous proportion of well-born sons were banished from contention. The daughters generally entered into élite marriages, creating horizontal links within the genealogical space. But these horizontal links supported rather than bound the men at the top of the hierarchy.

N.B.    These figures ignore inaugural occupants who obviously could not have inherited the post from their father. Statistics for the Makassar thrones are taken from Figure 1, and count Batara Gowa (G7) as a “greater raja”. The last three seventeenth century regents, and Gowa’s other non-territorial administrators, are not shown in Figure 1 but are documented elsewhere (Bulbeck 1992).

Thus the Makassar élite practised a type of apical demotion which depended structurally on patrilinealism and polygyny as élite privileges. To see how privileged this patrilinealism was, consider the very strong association between the status of a title (as measured by the status of the title holders) and the degree to which the title was inherited patrilineally (Table 6). Far from being a principle which included related individuals within a descent group, patrilineal descent was used by a higher status lineage specifically to exclude related men whose father belonged to a lesser lineage.

Just because a man was barred from membership within status lineages higher than his father’s, this did not guarantee him automatic membership within his father’s status lineage. On the contrary, apical demotion involves the continual reassessment of “ascribed status” depending on achievement. A man who failed to earn the required status lost his (potential) natal membership and either started a new status lineage or married into a lesser status lineage.

To take the example of the sons of a Gowa (or other Makassar) raja, any son who failed in the succession concurrently lost any direct claim on the title for his sons, and so began a new patriline. Therefore the only men included in a royal status lineage are those belonging to the patriline of rajas, including all the installed brothers.

Furthermore, any man who failed to attain a karaengship virtually guaranteed that his descendants would be banished to genealogical insignificance. So the only men included in a noble status lineage linked to the Gowa royal line are those who can trace a direct line of male Karaengs back to the son of a Gowa raja. They could also have traced an ambilineal line of ancestry into other lineages, but this would have been pointless since the Gowa royalty constituted the highest status Makassar lineage. We would also expect the Gowa nobility to exclude from their ranks any nobles directly descended from lines inferior to Gowa’s, and this expectation is confirmed by the genealogical distribution of the major noble titles (Figure 2).

Where do women fit into a genealogical system based essentially on men’s titles? For two reasons I assign women to their father’s status lineage even if the mother’s was higher. Firstly, the father’s title was the major influence on the offspring’s title irrespective of gender, and the systematic demotion observed in the opposite-sex next-of-kin comparisons held true between brothers and sisters. Secondly, the notion that women should marry at their own level or upwards implies that the husband enjoyed either equal or greater authority.

It is not even necessary to assume that daughters left their father’s status lineage upon marriage. Indeed the frequency of divorces, and the occasional instances of women marrying within their own status lineage (Bulbeck 1992), suggest that many women never did. However, the offspring were born within their father’s status lineage, either as potential members in the case of boys, or as members to be strategically married in the case of girls. This and the other points discussed above will become clearer during the description of my 17 “lineage groups” (see Map 1) and the associated status lineages.