33. In some jurisdictions, plea deals are openly used to win convictions in cases where evidence is weak and may not suffice to convict the defendant at trial.48 There are numerous US cases in which prosecutors have entered into plea deals with full knowledge that a defendant is factually innocent. The so-called Alford plea,49 in which a defendant accepts a conviction without admitting the factual basis of the charges, demonstrates that courts are willing to accept the constitutionality of convictions explicitly accepted for pragmatic purposes, rather than in acceptance of true responsibility for the charges.50 This type of functional plea deal was used, for example, in the case of the socalled “West Memphis Three”.51 In that case, three young men were wrongfully convicted following a full trial for the rape and murder of three eight year old children. After years of litigation, an appeals court remanded the case to the trial court for further evidentiary hearings that doubtless would have led to exoneration. Given the fragile health of one of the defendants, all three agreed to accept Alford pleas to sentences of time served rather than remain detained during further lengthy proceedings to prove their innocence.52 34.  Cooperating witnesses, while clearly key to investigations of complex and organised crime, can also threaten the quality of investigations. Participants in the Washington roundtable expressed concern about the reliability of testimony by witnesses who are incentivised with discounted sentences and immunity for certain charges.53 Indeed, 45.9% of death row exonerations in the US since the 1970s involved convictions based on testimony by cooperating witnesses, according to the Center on Wrongful Convictions.54 Beyond concerns about cooperating witnesses providing untrue, incomplete or misleading evidence, over-reliance on cooperation agreements can also have a corrosive effect on the ethical and procedural compliance of prosecutors and investigators.55 35.  However, establishing evidence of poor practice in relation to investigations is particularly difficult in the context of trial waiver systems, which do not produce an extensive evidential and procedural record in the way that trials do, and which severely limit the availability of appeals or other post-conviction review or relief. Therefore concerns about the effect of plea bargaining on the quality and reliability of prosecutions remain speculative – pointing again to the lack of transparency associated with trial waiver systems as opposed to open trials. 48. See Turner, J. I., “Prosecutors and Bargaining in Weak Cases: A Comparative View” (Aug. 31 2011), from The Prosecutor in Transnational Perspective, Luna, E. and Wade, M., eds., Oxford University Press (2012). Available at: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1862161. 49. North Carolina v. Alford, 400 U.S. 25 (1970). Alford faced the death penalty were he to be convicted at trial on a murder charge. He strategically pleaded guilty to second degree murder to avoid the possibility of a death sentence at trial, while denying responsibility for the crime as a factual matter. He later appealed against his initial guilty plea, saying he was coerced by the threat of receiving the death penalty were he convicted at trial, but the court held that even if the defendant would not have pleaded guilty “but for” the threat of the death penalty, a guilty plea remained valid as long as evidence existed to support the prosecution and the defendant pleaded guilty in order to avoid the harshness of sentencing (though note the dissenting opinion by Justice Brennan, who stated that the conviction should have been vacated because the plea was not voluntary, but was given under duress due to the threat of the death penalty, which he considered to be unconstitutional). 50. See also Blume, J. H. and Helm, R. K., “The Unexonerated: Factually Innocent Defendants Who Plead Guilty”, Cornell Law Faculty Working Papers, Paper 113 (2014). Available at: http://scholarship.law.cornell.edu/clsops_papers/113. 51. Moriarty, E., “48 Hours Mystery: Unusual Plea That Freed the ‘West Memphis 3’”, CBS News (Aug. 10 2012). Available at: http://www.cbsnews.com/news/48-hoursmystery-unusual-plea-that-freed-west-memphis-3/. 52. Shargel, G. L., “How the West Memphis Three Got Out”, The Daily Beast (Aug. 22 2011). Available at: http://www.thedailybeast.com/articles/2011/08/21/ west-memphis-three-freed-using-rare-alford-legal-plea.html. 53. See Warden, R. et al., “Jailhouse Snitch Testimony: A Policy Review”, The Justice Project (2007) for discussion of safeguards for such testimony. Available at:http:// www.pewtrusts.org/~/media/legacy/uploadedfiles/wwwpewtrustsorg/reports/death_penalty_reform/ilhouse20snitch20testimony20policy20briefpdf.pdf. 54. “The Snitch System”, Center on Wrongful Convictions, Northwestern University School of Law (2005), p. 3. Available at: https://www.aclu.org/other/center-wrongfulconvictions-snitch-system?redirect=center-wrongful-convictions-snitch-system. 55. See Baer, M. H., “Cooperation’s Cost”, Washington University Law Review, Vol. 88, Iss. 4 (2011), p. 931: ‘Prosecutors and agents will become less vigilant and cease monitoring each other when they rely on cooperation as a means of encouraging defendants to refrain from filing motions to suppress illegally obtained evidence. The failure to follow procedural rules, moreover, may undermine law enforcement’s legitimacy, force prosecutors to enter into agreements with suboptimal cooperators, and increase the overall likelihood of unchecked corruption and abuse within law enforcement agencies, all of which may lead to an increase in criminal conduct and a decrease in the prosecution of guilty actors.’ Available at: http://openscholarship.wustl.edu/law_lawreview/vol88/iss4/3. 16.