PurposeIn explaining ethno-territorial conflicts, leadership preferences have an odd status. In case studies, leadership preferences are often viewed as highly significant causes but are not usually defined and measured explicitly. In large-sample statistical studies, leadership preferences are only captured by weakly related proxy variables. This paper aims to fill this gap by developing suitable theory, which can be used consistently in both case study and statistical applications.Design/methodology/approachFormal bargaining models are used to examine the expected impact of variation in leadership preferences. Relevant leadership characteristics are then used to construct measures of variation in leadership preferences, which are applied in case studies.FindingsIn bargaining models, variation in leadership preferences is expected to have a significant impact on ethno-territorial conflict outcomes. More extreme nationalist leaders and, more conditionally, strongly power-seeking leaders, should be more likely to be willing to use force to modify the status quo – although more moderate nationalist leaderships are also willing to do so under certain conditions. In five case studies, these formally derived hypotheses receive initial empirical support.Originality/valueTheoretically and empirically, further refinement of research on variation in leadership preferences promises to add significant value. Formally, it is worth investigating the expected impact of additional preference types. Empirically, it is important to invest in measures of leadership preferences across large samples.

中文翻译：

---

领导偏好和种族谈判：理论和例证

目的在解释民族-领土冲突时，领导偏好具有奇怪的地位。在案例研究中，领导偏好通常被视为非常重要的原因，但通常没有明确定义和衡量。在大样本统计研究中，领导偏好仅由弱相关代理变量捕获。本文旨在通过开发合适的理论来填补这一空白，该理论可以在案例研究和统计应用中始终如一地使用。设计/方法/方法正式谈判模型用于检查领导偏好变化的预期影响。然后使用相关的领导特征来构建领导偏好变化的度量，并将其应用于案例研究。在讨价还价模型中，预计领导偏好的变化将对种族-领土冲突的结果产生重大影响。更极端的民族主义领导人，以及更有条件的强烈追求权力的领导人，应该更愿意使用武力来改变现状——尽管更温和的民族主义领导人在某些条件下也愿意这样做。在五个案例研究中，这些正式派生的假设得到了初步的经验支持。原创性/价值从理论上和经验上，对领导偏好变化的研究的进一步完善有望增加显着的价值。形式上，值得研究额外偏好类型的预期影响。从经验上看，重要的是投资于大样本的领导偏好测量。