This study evaluates the hypothesis that some documented cases of long-distance sea crossing by the Late Pleistocene Homo sapiens occurred as a result of accidental drifting, rather than by intentional seafaring. For that purpose, we use an existing computer simulation framework, with some modifications, to investigate the likelihood that a planned or unplanned island colonization by a small group of individuals will persist to establish a viable population. Within the original framework, planned colonization was operationally characterized as being initiated by equal numbers of unrelated young men and women, whereas for unplanned colonization, those who migrate inadvertently were regarded in effect as a random sample of the whole population. Here, we consider a different scenario for unplanned colonization, which we believe is more relevant to sea crossing by the Late Pleistocene humans, that is, we assume that unplanned colonization occurs when members of households on watercrafts with limited voyaging capabilities are drifted away by ocean currents and washed up on a distant island. We also extend the previous analysis by considering a broader range of combinations of fertility and mortality schedules that individuals are assumed to follow. Our simulations suggest the following: (1) colonization of an island by ten or fewer unrelated young men and women can be successful within the feasible range of fertility and mortality levels; (2) in comparison, the likelihood of success for unplanned colonization is considerably smaller for the same range of fertility and mortality levels; and (3) there exists a small range of parameter combinations for which unplanned colonization has a non-negligible prospect of success even without assuming recurrent accidental drifts to the same island, and thus, the accidental colonization scenario cannot be totally excluded. In addition, we find that the minimum founding population required for successful colonization varies substantially depending on the fertility and mortality levels.

这项研究评估了以下假设：一些晚更新世的智人长途出海的案例记录发生的原因是意外漂流，而不是故意航海。为此，我们使用现有的计算机模拟框架，并进行了一些修改，以调查由一小群人进行的计划内或计划外岛屿殖民化将持续建立可行种群的可能性。在最初的框架内，计划中的殖民地在业务上的特点是由相等数目的不相关的年轻男女发起，而对于计划外的殖民地，无意中迁移的人实际上被视为整个人口的随机样本。在这里，我们考虑了计划外殖民的另一种情况，我们认为这与晚更新世人类越过海有关，也就是说，我们假设当航行能力有限的船只上的家庭成员被洋流漂走并在遥远的岛屿上被冲走时，就会发生计划外的殖民化。我们还通过考虑假设个体遵循的生育和死亡率计划的更广泛组合来扩展先前的分析。我们的模拟表明以下几点：（1）在生育率和死亡率水平的可行范围内，由十个或更少的不相关的年轻男子和妇女在一个岛屿上定植是成功的；（2）相比较而言，在相同的生育率和死亡率水平范围内，计划外定植成功的可能性要小得多；（3）参数组合的范围很小，即使不假定经常性的偶然漂泊到同一岛上，计划外的殖民化也具有不可忽略的成功前景，因此，不能完全排除偶然的殖民化情景。此外，我们发现成功定殖所需的最低创始种群在很大程度上取决于生育力和死亡率水平。