As digital technologies and algorithmic rationalities have increasingly reconfigured security practices, critical scholars have drawn attention to their performative effects on the temporality of law, notions of rights, and understandings of subjectivity. This article proposes to explore how the ‘other’ is made knowable in massive amounts of data and how the boundary between self and other is drawn algorithmically. It argues that algorithmic security practices and Big Data technologies have transformed self/other relations. Rather than the enemy or the risky abnormal, the ‘other’ is algorithmically produced as anomaly. Although anomaly has often been used interchangeably with abnormality and pathology, a brief genealogical reading of the concept shows that it works as a supplementary term, which reconfigures the dichotomies of normality/abnormality, friend/enemy, and identity/difference. By engaging with key practices of anomaly detection by intelligence and security agencies, the article analyses the materialisation of anomalies as specific spatial ‘dots’, temporal ‘spikes’, and topological ‘nodes’. We argue that anomaly is not simply indicative of more heterogeneous modes of othering in times of Big Data, but represents a mutation in the logics of security that challenge our extant analytical and critical vocabularies.

中文翻译：

---

治理他人：异常和安全的算法主题

随着数字技术和算法理性越来越多地重新配置安全实践，批判性学者已将注意力转移到它们对法律的时间性、权利概念和对主体性的理解的表演性影响上。本文提出探索如何在海量数据中使“他者”变得可知，以及如何通过算法绘制自我与他者之间的边界。它认为算法安全实践和大数据技术已经改变了自我/其他关系。“他者”不是敌人或危险的异常，而是算法产生的异常。尽管异常经常与异常和病理学互换使用，但对该概念的简要谱系解读表明它是一个补充术语，它重新配置了正常/异常、朋友/敌人和身份/差异的二分法。通过参与情报和安全机构异常检测的关键实践，本文将异常的具体化分析为特定的空间“点”、时间“尖峰”和拓扑“节点”。我们认为，异常不仅表明大数据时代更多异质的他者模式，而且代表了挑战我们现有分析和关键词汇的安全逻辑的突变。