Anmerkungen:
Nach Informationen von SSRN wurde die ursprüngliche Fassung des Dokuments April 25, 2023 erstellt
Beschreibung:
Parliamentary intelligence oversight committees are one of the classic forms of intelligence oversight, yet they are frequently regarded as inadequate. Many reasons have been proposed, such as a claim that legislators neglect oversight because of lack of sustained electoral pressure, or that political alignment between the executive and the legislative majority dampens the oversight zeal. We propose yet another hypothesis – one of regulatory capture. Oversight committee members, drawn into the secret world of intelligence and dealing regularly with its concerns, adopt the perspective of the entities they are supposed to oversee. We propose to test that hypothesis by analyzing parliamentary debates and roll call voting patterns. We expect oversight committee members to exhibit patterns differing from their similarly situated legislative colleagues – and we expect the magnitude of that effect to increase for new committee members as they gain experience. To cope with the volume of the material to be analyzed, we will use machine learning methods such as Natural Language Processing (vocabulary analysis, alignment-based clustering, sentiment analysis) and dimensionality reduction (PCA, SVM). In the present paper, we focus on the Polish case as a proof of concept, although the method described can be applied to any country