Counter-Terrorism and Sentencing Bill

Further written evidence submitted by Dr Kyriakos N. Kotsoglou, Senior Lecturer in Law (Criminal Evidence), Northumbria University, and Marion Oswald, Vice Chancellor’s Senior Fellow in Law, Northumbria University (CTSB10)

Counter-Terrorism and Sentencing Bill 2019-21 (the Bill)

Supplementary w ritten evidence to the House of Commons Public Bill Committee

T PIMs and the p roposed p olygraph m easure

1. We have had the benefit of reading the transcript of the first evidence session to the Public Bill Committee.
We would respectfully disagree (for the reasons laid out in our original submission (CTSB01)) with Jonathan Hall QC’s view that polygraphs should be seen as an additional information source that is sensible to use. On the contrary, we would argue that it could be seen as irrational to rely on such a source due to its lack of validity and insufficiently reliable scientific basis.

2. Furthermore, in the context of TPIMs, the proposed inclusion of a polygraph condition creates new and potentially unforeseen risks. This is due to the proposal that the polygraph test be used to assess 'whether any variation of the specified measures is necessary for purposes connected with preventing or restricting the individual’s involvement in terrorism-related activity.' (clause 41 of the Bill). This suggests that the polygraph's assessment of truthfulness or deceptiveness of the individual subject to the TPIM (in respect of questions related to the other measures imposed) will be used to determine whether measures should be continued, expanded or discontinued. This appears to us to be highly dangerous and inappropriate (and of course would not be admissible in criminal proceedings). This provision seems to bring with it an implicit assumption that the polygraph can determine truthfulness or otherwise, and an abdication of the authorised legal officials’ decision-making prerogative to the polygraph, running contrary to the Secretary of State's duties under ss 3(3) & (4) of the Terrorism Prevention and Investigation Measures Act 2011. Neither can the polygraph detect psychological/normative concepts, nor can the polygraph operator draw evidence-based conclusions from the ‘ measurement and recording of the physiological reactions of an individual while being questioned in the course of a polygraph examination ’.

3. W e are concerned about the impact of the above approach on the fairness of the decision-making process. We would draw the Committee’s attention to a recent report on AI and National Security commissioned by GCHQ ( Babuta and Oswald, 2020: https://rusi.org/publication/occasional-papers/artificial-intelligence-and-uk-national-security-policy-considerations ) which concludes that 'None of the AI use cases identified in the research could replace human judgement. Systems that attempt to 'predict' human behaviour at the individual level are likely to be of limited value for threat assessment purposes.' We can draw an analogy from these concerns to the proposed imposition of the polygraph into sensitive decision-making processes relating to those subject to TPIMs , especially in terms of the opaqueness of the tool and the process, questions of scientific validity and the downplaying of the role of the legitimate decision-maker.

Kyriakos N. Kotsoglou and Marion Oswald
Northumbria University

30 June 2020