© Joanne Conaghan * (published version to appear in appear in (2014) 5(4) Transnational Legal
Theory 601-521
Key words: human rights, legal reasoning, police, rape, tort,
Abstract:
This article uses Duncan Kennedy’s analysis of legal reasoning to trace the discursive dynamics in DSD & NBV v Commissioner of Police for the Metropolis , a recent English case on rape and human rights law. Against a doctrinal background historically unreceptive to imposing civil liability on the police in relation to their role in investigating and suppressing crime, the article explores how the judge in DSD & NBV managed to reach a verdict favourable to the complainant rape victims. In particular, drawing on Kennedy’s work on adjudication and legal reasoning, the article shows how the judge, Green J, ‘worked’ the legal materials, making a series of strategic moves which served to dislocate and reconstitute the core and penumbra of the relevant legal norms. As a result, a decision which might initially resemble judicial activism became, in the course of the judgement, an apparently inevitable outcome of the legal normative framework.
1 INTRODUCTION: FROM MIAMI TO THE METROPOLITAN POLICE
I first met Duncan in early 1988 at an AALS conference in Miami. I was fortunate to be friendly with
Fran Olsen who invited me to a CLS 1 party in a fancy suite in the Fontainebleau Hilton. I was young, feminist and very much predisposed to challenge the hegemonic grip of white male supremacy in legal education, critical legal education in particular. As the unofficial ‘big chief’ of the CLS movement, Duncan represented for me everything that was male/sexist about radical politics and I relished the chance to knock him off his pedestal.
Things never turn out as one imagines. The Crits 2 I discovered were a cliquey bunch and quickly concluded I was nobody worth talking to. I had resigned myself to a quiet evening perched in
* Professor of Law, University of Bristol. This article is a worked up version of a talk I gave at the School of
Law, School of Oriental and African Studies, London, as part of a conference on ‘The Past and Future of the
Legal Left: Celebrating Duncan Kennedy’s Scholarship’, on 22 May 2014. I would like to express my thanks to the organisers, Nimer Sultan and Tara Mahmoud, for inviting me to participate in this wonderful and worthy event and also to my colleague Paula Giliker for her helpful comments on an earlier draft of this article.
1 CLS stands for ‘critical legal studies’, a branch of legal scholarship adopting a progressive left-leaning approach to legal scholarship and education. CLS flourished in the US in the 1970s and 1980s, not just as a form of scholarship but also as a political movement of which Duncan Kennedy was the uncrowned king.
2 ‘Crit’ is an informal term sometimes used to describe (or self-describe) those who identify with CLS.
1

the kitchen of the fancy suite (close to the mini-bar) when Duncan homed in on me. Within moments I was the sole focus of his—disconcertingly intense—attention and over the course of the next hour he grilled me on a rich variety of topics, virtually nothing of which I now recall other than that I found the whole encounter delightful. Instead of the arrogant, closed-minded, sexist, self-absorbed creature
I expected to meet, I confronted a playful and engaging man who was incredibly clever and thoroughly diverting. Duncan has a way of making you feel that when you have his attention you are the centre of his universe. He is intensely curious about everyone he meets and, as a consequence, his interest is quite unfeigned. Of course there is also something deliberative about it. I wasn’t stupid. I quickly figured out what he was doing and inwardly applauded it, particularly in an environment when status and who-you-know seemed so important. Duncan is truly—I mean to his bones
—antihierarchical. He enters a room, reads the group dynamic within minutes and swoops down to single out whoever occupies the lowest notch on the group totem pole. At the 1988 CLS party in Miami, that was me; but over the years I have seen him do it again and again at conferences, workshops and talks.
Students in particular have benefited from this adorable Kennedy quirk while distinguished professors, keen to mingle with the great white Harvard guru, stand by unacknowledged.
And so it came to pass that while my feminist leanings led me to be rightly chary of much of the masculinity which infused CLS politics and thinking at that time, I opened my mind to Duncan, and over the years I have learned an awful lot from him. We share, I believe, a strong appreciation of the political possibilities which law and legal discourse present and we both unashamedly delight in exploring those possibilities through testing the pliability and strategic potential of legal reasoning.
Indeed, if a commonality characterises our scholarship over the years, it undoubtedly lies in our shared interest in the internal workings of law and legal processes, and our view of legal reasoning as a particular kind of activity with its own institutions, conventions and constraints. This interest in the internal workings of law is of course related to our broader shared concern with law’s political role: to master law and legal reasoning, we would agree, is to equip oneself effectively to do political battle.
I want to illustrate this shared concern by using Duncan’s scholarship to trace the discursive dynamics of a recent English case on rape and human rights law. The judgement in DSD & NBV v
Commissioner of Police for the Metropolis
3 was delivered by Mr Justice Green in the High Court early in 2014. Although a first instance decision, Green J’s judgement deserves our attention because in a single instance it has achieved what endless legal reforms, policy reviews and political exhortations have not 4 —it has called the police squarely to account for their inexcusable failures in
3 [2014] EWHC 436 (HC).
4 There have been countless reviews and reports over the last 20 years detailing the difficulties which rape victims confront in their encounters with criminal justice actors, including the police. Indeed yet another review has recently been announced – see Vikram Dodd, ‘Metropolitan Police’s handling of rape allegations to be reviewed’ The Guardian (London, 9 June 2014) < www.theguardian.com/uk-news/2014/jun/09/metropolitanpolice-handling-rape-allegations-review-scotland-yard > accessed 2 July 2014. For a detailed analysis of the problems rape victims confront in their encounters with the criminal justice system see J Temkin and B Krahe,
Sexual Assault and the Justice Gap: A Question of Attitude (Hart Publishing, 2008).
2

the context of a rape investigation. Against a historical background of entrenched judicial reluctance to impose liability in precisely these kinds of circumstances , DSD & NBV is remarkable because it bucks the doctrinal trend, opening up new avenues of strategic possibility in the context of the political struggle to end violence against women.
Duncan’s analysis comes into play in helping us understand how Green J was able to deliver what at first glance appears to be such a radical result. Drawing upon two key texts, 5 I apply
Duncan’s insights to the judgment in DSD & NBV , to offer a critical account of how the legal outcome was achieved. I begin my analysis immediately below by outlining the facts of DSD & NBV.
I then move on in Part 2 to sketch the doctrinal background to the case, focusing in particular on the development of English tort law around police liability in the context of investigating and suppressing crime. In Part 3 the focus shifts directly to Duncan’s scholarship and, in particular, to his analysis of legal reasoning. This paves the way for the substantive application of Duncan’s insights to the reasoning of Green J in Part 4. The article concludes with some brief reflections about the possible implications of DSD & NBV , both doctrinally and in terms of how we understand and deploy legal reasoning strategically to pursue progressive political ends.
First however we turn to the case itself which revolves around the actions of a serial rapist,
John Worboys, who operated unchecked in the London Metropolitan area during most of the 2000s. It is estimated that during that period Worboys assaulted well over 100 women. Two of his victims, identified as ‘DSD’ and ‘NBV’, brought claims against the Metropolitan Police Service (hereinafter
‘MPS’) for their shambolic handling of what has become widely known as ‘the black cab rapist case’ because Worboys used his role as a London cab driver to prey on women passengers. The claimants alleged that in the handling and investigation of their complaints, the MPS had breached a duty owed to them by virtue of an obligation placed on public authorities under section 6 of the Human Rights
Act 1998 (hereinafter ‘HRA’) 6 not to act in ways incompatible with the claimants’ rights under the
European Convention on Human Rights (hereinafter ‘ECHR’).
7 Specifically, the claimants argued that under Article 3 of the ECHR (the right not to be subjected to torture or inhuman and degrading treatment) the police as public authorities owed a duty to investigate to victims of particularly serious crimes perpetrated by private parties, a duty which, on the facts presented, had been breached. Green
J upheld the claims, identifying multiple systemic and operational failures which taken together
5 Duncan Kennedy, ‘Freedom and constraint in adjudication: A critical phenomenology’ and ‘A Left
Phenomenological Alternative to the Hart/Kelsen theory of legal interpretation’, both usefully reproduced in
Duncan Kennedy, Legal Reasoning: Collected Essays (The Davies Group Publishers, 2008) 11 and 153 respectively. For full details of the publication history of these two pieces, see (n 47) below.
6 According to HRA 1998, s 6(1): ‘it is unlawful for a public authority to act in a way which is incompatible with a convention right’. The overall purpose and effect of the HRA 1998 was to ‘bring rights home’ by enabling claimants to raise human rights challenges directly in the UK courts. See generally Susan Millns,
‘”Bringing rights home”: feminism and the Human Rights Act 1998’ in Susan Millns and Noel Whitty (eds),
Feminist Perspectives on Public Law (Cavendish Publishing, 1999) 181.
7 Convention for the Protection of Human Rights and Fundamental Freedoms as amended by Protocol No. 11 and No. 14 (‘ECHR’).
3

violated the claimants’ Article 3 rights.
8 He delivered a judgement which was nothing short of scathing of the police handling of the Worboys case and in a text running to well in excess of 100 pages, much of which is devoted to laying out in detail an appalling level of police ineptitude and indifference, there unfolds a tale so distressing in its content, so disquieting in its implications, that it serves as a more powerful and persuasive indictment of police failings than any academic study or data-rich report. Regardless of the actual outcome of the case, Green J’s forensic analysis of the extent of police dereliction in the concrete context of two particular, heart-wrenching experiences stands as an authoritative public record of the facts of this investigation which no subsequent decision is likely to unsettle.
2 A QUICK DETOUR INTO TORT TERRITORY
To appreciate fully the significance of DSD & NBV , the decision needs to be set against the backdrop of broader developments in English tort law regarding police liability in the context of investigating and suppressing crime. There is a rather protracted narrative here which starts with a famous claim against the West Yorkshire police in the late 1980s, Hill v Chief Constable of West Yorkshire Police .
9
Jacqueline Hill was the last victim of the notorious ‘Yorkshire Ripper’, Peter Sutcliffe. After the latter’s apprehension, Jacqueline’s mother sued the police, arguing that they owed a duty of care to her daughter to conduct the investigation into the Ripper murders with reasonable care and skill, and that breach of that duty had led to Jacqueline’s death. In denying that a duty of care was owed, the
House of Lords held that the relationship between Jacqueline Hill and the West Yorkshire police was insufficiently proximate to give rise to a duty of care: Jacqueline was but one of a large class of potential victims at risk of harm from the Ripper. Their Lordships also took the view that policy considerations militated against recognition of a duty of care. In particular, the imposition of civil liability on the police for failing to protect members of the public from injuries inflicted by unknown criminal third parties risked ‘significantly divert[ing] police manpower and attention from their most important function, that of the suppression of crime’.
10
Litigation thereafter tested whether a closer relationship of proximity might yield recognition of a duty, but this avenue was closed in Osman v Ferguson 11 in which the Court of Appeal denied the existence of a duty where the police were clearly aware of a serious threat posed by an identifiable perpetrator towards discrete, identifiable victims. Notwithstanding McCowan LJ’s acknowledgement
8 Systemic failings included i) training failures; ii) failure to supervise investigations properly; iii) failures to collect and use intelligence sources to cross-check complaints; iv) failure to maintain victim confidence in the process and to create an environment where victims were incentivised to report their assaults; v) failure to allocate proper resources to the investigation of sexual assaults, focusing resources on easier clear-up cases.
Individual operational failings in relation to the handling of DSD and NBV’s claims were also extensively documented by Green J in DSD & NBV (n 3) [13].
9 [1989] 1 AC 53.
10 Ibid [63] (L Keith). On one reading the decision in Hill appears to turn solely on lack of proximity so that
Lord Keith’s policy musings are technically obiter. However, as we shall see, the policy arguments have since come to assume central importance in determining the existence and scope of liability in such cases.
11 [1993] 4 All ER 344.
4

that the claimant and his family were ‘exposed to a risk … over and above that of the public’, creating
‘a very close degree of proximity amounting to a special relationship’ between the claimant and the police, 12 he and his brethren determined that the policy concerns articulated in Hill precluded recognition of a duty relationship even in circumstances where proximity existed.
Thereafter the doctrinal narrative becomes increasingly fragmented and confused. As against the police, tort claims relating to harm inflicted by criminal third parties hit something of a judicial brick wall.
13 By contrast, suits against other public authorities, for example, social services, while initially encountering the obstacle of Hill , began to fare better as time went on.
14 This was partly a consequence of decisions in the European Court of Human Rights: the Osman family brought a successful claim against the UK for violating their Convention rights, arguing that the ‘blanket immunity’ enjoyed by the police with regard to their negligent acts constituted a denial of their right to a fair trial under Article 6 of the ECHR.
15 The Court in Osman also held that the State’s obligations under Article 2 (the right to life) 16 imposed a positive obligation on public authorities ‘to take preventative operational measures to protect an individual whose life is at risk from the criminal acts of another individual’.
17 In the subsequent decision of Z v UK , 18 brought by the losing claimants in X v Bedfordshire , 19 this positive public duty to provide effective protection under Article 2 was extended to Article 3 (the right to be free of torture and inhuman or degrading treatment).
20
At this point the case law stood something like this: the ECHR had intimated that the imposition of any absolute immunity protecting public authorities from negligence liability could violate Article 6; the Court had also recognised that Articles 2 and 3 created positive obligations on public authorities to protect individuals from harm inflicted by criminal third parties in some, albeit very limited, circumstances.
21 In a series of cases thereafter, the UK courts struggled with the
12 Ibid [350].
13 For a penetrating critique of the legal position here, see C McIvor, ‘Getting defensive about police negligence: the Hill immunity, the Human Rights Act and the House of Lords’ (2010) CLJ 133.
14 See eg X v Bedfordshire CC [1995] 2 AC 633 (HL) in which a claim against a public authority for failing to protect children from abusive parents was struck out. By contrast, in Barrett v Enfield [2001] AC 550, a striking out application in a claim made against a council for negligent foster placement was denied while in D v East
Berkshire Community Health NHS Trust [2004] 2 WLR 58, involving mistaken allegations of parental abuse and the institution at care proceedings, the Court of Appeal allowed the children’s claims to proceed but not those of the parents’ (a further appeal by the parents to the House of Lords also failed - [2005] 2 AC 373).
15 Osman v UK [1999] 1 FLR 193 (ECHR). ECHR, Art 6 (1) reads: ‘everyone is entitled to a fair and public hearing within a reasonable time by an independent and impartial tribunal established by law’. In Z v UK [2001]
2 FLR 612, the ECHR retreated from the view that English law applied an absolute or ‘blanket’ immunity in such cases, acknowledging it had misunderstood the role of policy in Osman . See further C Gearty, ‘ Osman unravels’ (2002) 65 MLR 87.
16 ECHR, Art 2: ‘Everyone’s right to life shall be protected by law’.
17 Osman (n 15).
18 See (n 15).
19 See (n 14).
20 ECHR, Art 2 and 3 (n 7).
21 These circumstances are spelled out in Osman (n 15): ‘It must be established … that the authorities knew or ought to have known… of the existence of a real and immediate risk to the life of an identified individual or individuals from the criminal acts of a third party and that they failed to take measures within the scope of their
5

implications of these decisions in terms of delineating the scope of tortious liability. However, in general terms one detects a broad trend during the course of the 2000s towards recognizing tort liability in a handful of circumstances but mainly where the defendant in question is a public authority other than the police.
22 In the latter context, and except in some fairly contained contexts, for example, when prisoners sustain harm while in police custody, 23 the tendency has been to keep the gate of tortious liability firmly closed while simultaneously denying that to do so is to impose any ‘blanket’ police immunity. Of particular significance here is the case of Brooks v Commissioner of Police v
Metropolis 24 in which Duwayne Brooks, a friend of the murdered black teenager, Stephen
Lawrence, 25 alleged that the police breached a duty of care owed to him by effectively failing to take reasonable steps to assess whether he was the victim of a crime and to respond appropriately. Note here that the claim is not that had the police acted differently Duwayne (and Stephen) would not have been harmed by the racist gang which attacked them, but rather that the poor investigation itself was an additional source of harm and trauma to Duwayne in relation to which a duty of care arose.
Unsurprisingly, given the previous doctrinal trend, the House of Lords denied the existence of any such duty. Citing the ‘core principle in Hill ’ and accompanying policy reasons, Lord Steyn observed:
A retreat from the principle in Hill’s case would have detrimental effects for law enforcement… By placing general duties of care on the police to victims and witnesses the police’s ability to perform their public functions in the interests of the community, fearlessly and with despatch, would be impeded.
26
There is a final wrinkle in this woeful tale which needs ironing out before we can move on. It will have been noted above that the intransigence of the English courts led some claimants to seek justice by framing their claim in human rights terms rather than in tort. With the advent of the HRA 1998, the purpose of which was directly to incorporate the human rights guaranteed under the European
Convention into English law, this strategic possibility became ever more potent, and indeed as the new century progressed considerable ink was spent speculating on how the HRA could and should change the shape and direction of tort law.
27 The issue was eventually confronted directly by the powers which judged reasonably might have been expected to avoid that risk’ (at para 116). The risk to life (Art
2) was extended to the risk of torture, inhuman and degrading treatment (Art 3) in Z v UK (n 15).
22 See (n 14).
23 See eg Reeves v Commissioner of Police for the Metropolis [2000] 1 AC 360.
24 [2005] UKHL 24.
25 Duwayne Brooks and his friend Stephen were victims of a racially motivated attack in which Stephen was killed. The poor police investigation which followed was the subject of widespread criticism and led to the setting up of a public inquiry – Home Affairs, The Stephen Lawrence Inquiry: Report of an Inquiry of Sir
William MacPherson of Cluny (Cm 4262-1, 1999) in which, inter alia, the Metropolitan police were described as
‘institutionally racist’. See generally L Bridges, ‘The Lawrence Inquiry – incompetence, corruption and institutional racism’ (1999) 26 Journal of Law & Societ y 298.
26 Brooks (n 24) [130].
27 For an overview of the relevant literature, see D Nolan, ‘Negligence and the Human Rights Act: the case for separate development’ (2013) 76 MLR 286.
6

courts in two cases heard by the House of Lords as a conjoined appeal in 2008.
28 Both cases involved investigative failings by the police, resulting, in Van Colle , in the death of a key witness and, in Smith , in serious injury to the claimant in the context of domestic violence. Both claims failed notwithstanding that they were predicated on distinct legal grounds. In Van Colle , the claimants directly invoked human rights, arguing that the circumstances met the narrow criteria articulated by the ECHR in Osman v UK .
29 Their Lordships determined otherwise. In Smith , in which, on the facts, the police level of awareness of the degree and imminence of risk to the claimant was considerably greater than in Van Colle , the claim was negligence-based and, in circumstances where, had it been possible to mount a human rights argument it might have succeeded, 30 their Lordships affirmed Hill and Brooks , denying the existence of any tortious duty of care.
31 Addressing the question of the interrelationship between tort and human rights claims, the majority of their Lordships took a firmly separatist view. Thus, unlike Lord Bingham who clearly veered towards some kind of harmony of principle, 32 Lord Hope expressed the view that ‘the common law… should be allowed to stand on its own feet, side by side with the alternative remedy’.
33 This of course was no comfort to Mr Smith whose choice of action was governed by the constraints of limitation periods; indeed the denial of justice based on this kind of procedural formalism cannot fail to elicit a sense of disquiet.
In any event and bringing the focus back to the case with which we began, we see that what
DSD and NBV confronted when they embarked upon proceedings was a stubbornly closed body of tort law, a slowly awakening human rights jurisprudence and a judicially installed divide between the two kinds of claim that rivalled the Berlin wall.
3 LEGAL REASONING AND DUNCAN’S ‘PHENOMENOLOGICAL ALTERNATIVE’
Having painted a broad picture of the legal background to DSD & NVB, I now want to offer a brief elaboration of Duncan’s scholarship with a view to applying his analytic insights to that case. The core of Duncan’s critical take rests on the assertion that law is a site for the disposal of ‘ideological stakes’ and that legal reasoning comprises a set of strategic operations with political implications and distributive effects. This argument is set out in greatest depth and sophistication in A Critique of
28 Van Colle v Chief Constable of Hertfordshire Police; Smith v Chief Constable of Sussex Police [2008] UKHL
50.
29 See (n 21).
30 Mr Smith was unable to make a claim under the HRA 1998 because he was outside the limitation period. See
HRA 1998, s.7(5): ‘Proceedings … must be brought before the end of (a) the period of one year beginning with the date on which the act complained of took place; or (b) such longer period as the court of tribunal considers equitable having regard to all the circumstances’. The attack on Mr Smith took place on 10 March 2003 but legal proceedings were not issued until 2 March 2006.
31 Lord Bingham dissented from this holding, arguing for a narrow exception to the Hill principle broadly in line with the Osman criteria; see (n 21).
32 Van Colle (n 28) [58].
33 Ibid [82]. See also L Brown at para 139. This separatist approach is explored and defended by Nolan (n 27).
7

Adjudication 34 but it is a theme which infuses most of his work and in my view 35 constitutes his most potent contribution to critical jurisprudence. This is no crude, reductionist collapse of law into politics—Duncan does not argue that law is indeterminate or that judges are never constrained, nor that adjudication is a grand conspiracy by the rich and powerful to exploit and oppress the poor and disadvantaged. Rather he argues that the determinate effects of legal rules are both overstated and misunderstood; to put it another way, it is not rules per se which determine outcomes but what judges do with and to rules. Judges, Duncan argues, ‘work’ the legal materials to reach ideological outcomes they desire while simultaneously experiencing themselves as bound by the ‘texts’ upon which they draw.
36 They are then largely in denial about their role as political and ideological operators and this lack of awareness carries its own political consequences. Specifically, it reinforces the view that
‘correct’ outcomes to legal decision-making exist, placing those decisions beyond the reach of political and ideological challenges.
37
All this makes the business of harnessing law to progressive political ends somewhat hazardous. As Wendy Brown and Janet Halley argue, ‘the effects of law can be complex, multiple, and contingent’.
38 Not only can law produce unintended consequences which may even militate against the very political goals it was invoked to promote, 39 but such ‘normativizing deployments of state power’ 40 risk introducing new forms of regulatory discipline and new techniques of domination.
Duncan of course is well aware of the perils which accompany legal political activism—his caution with regard to the pursuit of left political projects is particularly evident in the context of his critique of rights.
41 This may seem a somewhat paradoxical stance to take: how can Duncan view law as an important site of political contestation and ideological stakes while simultaneously exhorting us to approach rights critically, even sceptically? It is at this point that Duncan’s left leanings collide with his postmodernist sensibilities, captured in what he describes as his ‘left/mpm’ 42 stance. On the one hand, he is openly sympathetic to left political values and aspirations—fighting injustice, promoting egalitarianism and so on; on the other hand, he expresses a strong antipathy to programmatic political projects, left leaning or otherwise, and is particularly averse to assertions of normative superiority or
34 Duncan Kennedy, A Critique of Adjudication: Fin de Siecle (Harvard University Press, 1997).
35 See further my analysis in ‘Wishful Thinking or Bad Faith: a Feminist Encounter with Duncan Kennedy’s
Critique of Adjudication’ (2001) Cardozo Law Review 721.
36 Kennedy, Critique of Adjudication (n 34), in particular ch 8.
37 Ibid.
38 Wendy Brown and Janet Halley (eds), Left Legalism/Left Critique (Duke University Press, 2002) 13.
39 For example, in the context of debate about rape reform, it is sometimes argued that in their efforts to highlight the extent to which rape myths continue to exercise a grip on the conduct and disposition of rape cases, feminist activists are actually deterring rape victims from coming forward to report their rapes. I am not aware of any data to support or deny this assertion but it does provide a useful example of the double-bind which strategic engagement with law may present.
40 Brown & Halley (n 38) 14.
41 D Kennedy, ‘The Critique of Rights in Critical Legal Studies’ in Brown and Halley (n 38) 178; see also A
Critique of Adjudication (n 34) ch 13.
42 Where ‘mpm’ stands for ‘modernism/postmodernism, see A Critique of Adjudication (n 34) 11-12.
8

‘rightness’ which, he insists, tend to characterise such programmatic efforts.
43 For Duncan law is a site, not for political reconstruction based on ‘coherent and complete (“totalizing”) descriptive and prescriptive analyses of the social order’ 44 but for ‘outward—and inward—looking ideological encounters and coalitions rather than a set of principles or a program’.
45
I take the time to explain this particular aspect of Duncan’s overall approach because if there is a point of divergence between our views, it arises here. The political choices are not as stark, I would argue, as Duncan’s analysis suggests and left political projects can and should be pursued through law (as well as through other appropriate means), albeit cautiously, tentatively and always open to the possibility of revision or indeed retreat. Sometimes we have to take responsibility for our political positions especially when the distributive stakes are great, and the ethical and justice implications of failing to act profound. Recognizing ‘the limits, paradoxes, and perils attending contemporary practices of left legalism’ 46 helps to ensure that we enter legal terrain appropriately forewarned and forearmed.
Having identified some small reservations in relation to Duncan’s overall theoretical framework (although I suspect he would hotly deny that he has ‘an overall theoretical framework’), I now want to look more closely at how Duncan characterises and deploys legal reasoning. I am going to focus on two particular texts in the Kennedy canon: Freedom and Constraint in Adjudication: a
Critical Phenomenology (hereinafter Freedom and Constraint ) and A Left Phenomenological
Alternative to the Hart/Kelsen Theory of Legal Interpretation (hereinafter A Left Phenomenological
Alternative ).
47 Both essays will inform my analysis of Green J’s judgement in DSD & NBV below.
Freedom and Constraint is an early statement of Duncan’s approach to legal reasoning. It adopts a phenomenological approach in that it purports to explore not theories of adjudication but the experience of judging: Duncan imagines himself as a particular judge in a particular court hearing a particular case in which he confronts a dilemma between what he thinks ‘the law’ requires and ‘how-
I-want-to-come-out’(or, using Duncan’s acronym, ‘HIWTCO’).
48 What reasoning processes or intellectual moves characterise the judge’s resolution of that dilemma? How does he 49 navigate beyond the apparent objectivity of the rule which stands in his way to achieve the result which he
43 Ibid, ch 14.
44 Kennedy, ‘The Critique of Rights’ (n 41) 219.
45 Ibid , 224.
46 Brown & Halley (n 38) 4.
47 Duncan Kennedy, Legal Reasoning: Collected Essays (The Davies Group Publishers, 2008) at 11 and 153 respectively. ‘Freedom and Constraint’ was first published in 1986 (36 Journal of Legal Education 518) with a shorter version appearing in A Hutchinson & P Monahan (eds), The Rule of Law: Ideal or Ideology (Carswell,
1987) 45. Various mainly foreign language versions of ‘A Left Phenomenological Alternative’ appeared before an English language version was published in the Collected Essays .
48 Ibid, Kennedy, ‘Freedom and Constraint’ 12-13.
49 I use the male pronoun here because Duncan does. After endless self-interrogation of whether to substitute the female for the male pronoun and the various consequences thereof, I decided to stick with the designation in the original text. No doubt in 1986 I would have expressed my objections to this gendered reification rather more forcefully.
9

prefers, perhaps for political reasons, perhaps because he feels it is the more just solution … but what to do about that darn rule ? The rest of Duncan’s article is a detailed account of the various strategies open to the judge in circumventing the rule, from corralling the facts so that they fall outside the rule, to delineating the rule so that it falls outside the facts, to questioning the authority of the rule and probing its strength and legitimacy. Crucially in this process the judge is able to convince himself that he is not pursuing his own agenda but testing the extent to which the law really requires the outcome he dislikes.
50 By drawing upon a range of tactics which comprise the daily bread and butter of good lawyering—comparing, distinguishing, analogizing, balancing and so forth—the judge begins to realise that there is far more room for manoeuvre than he at first realised. While he is not completely unconstrained in the choices he can make—after all, the outcome of the case depends on the legal materials he confronts—there is nevertheless more freedom to ‘work’ those materials than he perhaps thought. The judge feels simultaneously free to work towards his desired outcome while constrained by the materials with which he is working; but the degree of freedom is greatly enhanced once he recognises that those materials—the medium through which legal reasoning operates and applies—are far more malleable, far less fixed than is commonly supposed.
51 Matters come to a head, Duncan posits, once the judge recognises that so much uncertainty characterises any claim to a determinative disposition that his preferred view is as defensible as any other. At that point the judge is able to dispose of the case remaining faithful to himself and to the law which he is sworn to uphold.
52
To the modern reader perhaps, Freedom and Constraint may not elicit the kind of light-bulb effect it had on those of us who read it when it first appeared. There is plainly greater acknowledgment in contemporary legal education and scholarship of the degree of room for doctrinal manoeuvring. However, in my view, what remains important about Duncan’s critique is not that it has now—to a considerable extent—been absorbed within mainstream accounts of law and legal reasoning but that it explains with incredible lucidity how judges can and do operate politically while remaining wholly faithful to principles of legality.
53
Let us move on to Duncan’s second text, A Left Phenomenological Alternative, which is both a critique of legal positivism and a clarification of some common misperceptions held in the mainstream legal community about what precisely that critique entails. As with Freedom and
Constraint , Duncan is keen to emphasise that CLS does not commit to the view that law is merely a façade for politics or that legal rules are wholly indeterminate. Make no mistake—Duncan takes law seriously (too seriously, some have suggested) and he is particularly fascinated by the discursive power of legal reasoning.
50 Kennedy, ‘Freedom and Constraint’ (n 47) 18-19 and 60.
51 Ibid , 25.
52 Ibid, 60-65.
53 For analysis of adjudication in terms of fidelity to principles of legality, see T Bingham, ‘The Judges: Active or Passive’ (2006) 139 Proceedings of the British Academy 55. For Lady Hale’s reflections, prompted by T
Bingham’s article, on whether one can be both a judge and a feminist, see B Hale, ‘A Minority Opinion?’
(2008) 154 Proceedings of the British Academy 319.
10

A Left Phenomenological Alternative focuses on HLA Hart’s classic distinction between the
‘core’ and ‘penumbra’ of a legal rule.
54 Duncan begins as follows:
We imagine the norm that is to be interpreted as an area or space that has two parts. In Hart, there is a ‘core of certainty’ and a ‘penumbra of doubt’ … Within the Hartian core, interpretation is ‘determinate’. In the penumbra, it is a matter of discretion, or balancing of conflicting considerations, or judicial legislation, or law-making’.
55
According to Duncan, Hart and Kelsen both conceive norms broadly in terms of a core of settled content accompanied by some blurring or uncertainty around the edges. Whether or not that uncertainty will assert itself will depend on the circumstances in which the norm is invoked— sometimes the situation presented will call for no more than a straightforward application of the core content of the norm; at other times, matters may be more complex and the scope and application of the norm to the facts in question open to argument.
56 Crucially, however, and as the spatial metaphor suggests, the sphere of relative determinacy is considerably greater than the indeterminate edge; therefore while judges can and sometimes do exercise discretion, and while it would be wrong to assert that any legal norm is completely determinate (a position Hart equates with legal formalism), the broad legal positivist position is that the space for judicial discretion and choice is relatively small 57 .
Duncan has a number of things to say about this, two of which are important for purposes here. First, Duncan observes that legal positivists offer no proper account of how or why norms are plagued with indeterminacy and no satisfactory theory which explains the conditions in which indeterminacy arises or is judicially managed.
58 Secondly, and more importantly, he points out that the positivist conception of a norm as possessing a core of certainty and a penumbra of doubt which the judge must somehow tease out begs an enormous question—how in the world does the judge do so?
59 Moreover, what if, in the course of teasing out these normative parameters, the judge actually constitutes—constructs—them so that the core and the penumbra, far from being discovered , are fashioned out of legal reasoning processes.
60 This of course is Duncan’s key point—that what gets designated as core and what gets assigned to the penumbra is itself a product of ‘working’ the legal
54 H L A Hart, The Concept of Law (Clarendon Press, 1961) 119, and generally ch 7. Duncan also works in
Kelsen’s theory of legal interpretation, drawing in particular on Hans Kelsen,
Introduction to the Problems of
Legal Theory (B Paulson & S Paulson trs, OUP 1992). However, the critique speaks much more directly to
Hart’s analytical frame and for our purposes that will suffice.
55 Kennedy, ‘A Left Phenomenological Alternative’ (n 47) 154.
56 Ibid , 154-5 .
57 Ibid , 156. Hart further elaborates on his notion of judicial discretion in a ‘Postscript’ which appears at the end of the second edition of the Concept of Law (Clarendon Press, 1994) 238 at 272-276. Here he characterises judicial discretion in terms of ‘filling the gaps by exercising limited law-creating discretion’ (272).
58 Kennedy, ‘A Left Phenomenological Alternative’ (n 47) 157.
59 Ibid , 157-161.
60 Ibid.
11

materials.
61 It is important to be clear about what that means. Duncan is not saying that the core of certain content is much smaller than legal positivists acknowledge and the penumbra of doubt far larger. This is no conflict over the correct delineation of the legal interpretative space. Rather, he is arguing that what becomes framed as or assigned to the core and what becomes framed as or assigned to the penumbra is, in significant part, a result of judicial interaction with and strategic working of the relevant legal materials.
In other words:
… legal materials do or do not determine the outcome of cases only in interaction with the argumentative strategies of jurists pursuing objectives … The materials are one part of the determination but only in combination with interpretative activity.
62
Sometimes working the materials means trying to find an argument to counter the normative power of the apparently determining rule a judge confronts; but sometimes too one can unsettle the rule, cast doubt on its claim to determinacy by working the material around it. Sometimes, the way to get a result is to shift the core into the penumbra or to reposition the penumbra as part of the core, although it may be easier to do this in some cases than in others. However, acknowledging the judicial deployment of these kinds of strategic manoeuvres is a far cry from the traditional positivist account of legal reasoning in which rules operate logically and syllogistically 63 to determine outcomes— except in those rare instances where uncertainty arises requiring the exercise of judicial discretion.
This is because, for Duncan, determinacy and indeterminacy are not ‘”qualities” or “attributes” which inhere in the norm independently of the work of the interpreter’.
64
4 HOW GREEN J ‘WORKS’ THE LEGAL MATERIALS: DSD & NBV
Let’s return now to the suit against the Metropolitan Police brought by two victims of the London
‘black Cab rapist’, John Worboys. Imagine yourself in the shoes of Mr Justice Green confronted by professional failings of the most serious kind, with the most grievous of consequences and a legal doctrinal framework which, at a glance, is anything but sympathetic to the claimants. What does one do? How does one circumvent that intuitive sense of what ‘the law’ is to ensure a result consistent with HIWTCO? Of course we don’t know that Mr Justice Green did experience the dilemma of the fictional judge in Freedom and Constraint . He may have come to the issue with no particular preconceptions about the preferred result. We do know that prior to his appointment to the Bench late in 2013 he was a public law and human rights lawyer, and we do know that he looked very closely at
61 Ibid, 160.
62 Ibid , 166.
63 On the importance of the syllogism in legal reasoning, see N MacCormick, Legal Reasoning and Legal
Theory (rev edn, OUP 1994). For a critique of MacCormick’s account, see J Conaghan, Law and Gender
(Clarendon Press, 2013) ch 6, including a detailed elaboration of my own approach to legal reasoning.
64 Kennedy, ‘A Left Phenomenological Alternative’ (n 47) 160.
12

the facts and was far from happy with what he saw.
65 Certainly, were I in his shoes, I would be battling my initial apprehension that the Hill obstacle was insurmountable. So, for the sake of the exercise, let’s assume that he did also. How then did he bring off the seemingly impossible—the apparent end of the police immunity with regard to their role investigating and suppressing crime? I will argue that in the course of his judgement Green J made a series of strategic moves not unlike those detailed by Duncan in Freedom and Constraint in Adjudication . Moreover, echoing Duncan’s analysis in A Left Phenomenological Alternative , I will suggest that Green J worked the legal materials to dislocate and reconstitute the core and penumbra of the relevant legal norms so that a decision which might initially be apprehended as a pronounced form of judicial activism became, in the course of the judgement, an apparently inevitable outcome of the legal normative framework.
First, what precisely did that decision entail? We have heard that Green J accepted the claimants’ argument that by virtue of their obligations under the HRA, specifically s 6, the police owed a duty to investigate to victims of serious crimes perpetrated by private parties. This duty derived from Article 3 of the ECHR, the prohibition against torture and inhuman and degrading treatment, which, in addition to imposing a negative duty on the State to refrain from engaging in torture and so on, also carried with it a positive obligation to ensure an adequate system of legal protection for people against ill-treatment by private parties. The development of positive obligations on the State in relation to Convention rights has being going on for some decades and preceded the introduction of the HRA in the UK.
66 In the context of Article 3 it had evolved to encompass duties to protect or prevent ill-treatment as well as to investigate and enforce the law, giving rise to individual claims of rights violations where particular, fairly stringent circumstances were met. The key question before Green J was whether or not this line of Strasbourg case law supported the imposition of a positive duty to investigate in the circumstances before him.
Green J concluded that it did, that a duty to investigate was owed by the police to DSD and NBV which on the facts had been breached.
67 This was notwithstanding the arguments mounted by the defendants that (1) if a duty to investigate did exist it only applied where the police were directly implicated in the harm alleged, for example, where injuries are sustained while in police custody, and
(2) that insofar as the Strasbourg case law might suggest otherwise the cases relied upon by the claimants were ‘essentially rogue or maverick’ 68 and should not be followed. In other words, the defendants disputed the status and applicability of the Strasbourg jurisprudence in a domestic context as well as the alleged scope of any duty to investigate derived from that jurisprudence. I have to say,
65 This comes out at various points in the tone and content of his judgement, particularly in his analysis of the
Independent Police Complaints Commission (IPCC) reports on the Worboys investigation ( DSD & NBV (n 3)
[125-134]).
66 The first case to recognise a positive State obligation in the context of Art 3 was Costello-Roberts v the UK A
247-C (1003); 19 EHRR 112 . See generally Alastair Mowbray, The Development of Obligations under the
European Convention on Human Rights by the European Court of Human Rights (Hart, 2004).
67 DSD & NBV (n 3) [138].
68 Ibid [227] which includes a summary of the defendant’s main submissions.
13

Green J gave both these arguments fairly short shrift although he was very polite about it. In fact there is remarkably little direct engagement with the defendants’ arguments; rather what Green J does is to engage in a lengthy dissection of the relevant Strasbourg case law to construct a clear narrative in support of the position he adopts. But I get ahead of myself, because this is not the first strategic move that he makes but the second. Let’s take a closer look. a) Move 1: Excise the core
At the commencement of his judgment Green J announces that ‘under the common law the police do not owe a duty of care in negligence in relation to the investigation of a crime’.
69 Thereafter he gives no further consideration to tort law and conducts the discourse exclusively in terms of human rights.
Now you could say this was a bold and brilliant move; or you could say it was a decision made for him by the way in which the claimants chose to frame their arguments. Certainly it seems that the claimants’ legal representatives made a strategic decision not to attempt to unsettle what, after
Smith 70 , looked like settled law. If the unfortunate Mr Smith could not found a claim in negligence against the police, surely no one could. This assumption that the edifice of tort was impenetrable did not come without costs. For example, it provided no opportunity for a sympathetic court to draw upon
Lord Bingham’s excellent dissent in
Smith , articulating an exception within the tort framework based on the Osman criteria.
71 It invited no consideration of decisions in other common law jurisdictions where a tort claim has been recognised notwithstanding Hill . In Canada, for example, in circumstances not dissimilar to those of DSD & NBV , a duty of care was imposed upon the police in relation to their failure to conduct an adequate investigation into the activities of a serial rapist in
Toronto.
72 Similarly, in South Africa, a line of cases has evolved around a duty on the police to
69 Ibid [4].
70 See (n 28).
71 See (n 21).
71 See (n 21).
72 Doe v Metropolitan Toronto (Municipality) Commissioners of Police (1998) 160 DLR (4 th ) 269. In Doe the claim relied both upon negligence and various breaches of the Canadian Charter of Rights (s 7 & 15). With regard to the negligence claim, the applicant managed to bypass policy objections to the imposition of a duty of care by focusing on operational failures by the police in the investigation of a serial rapist and the failure to warn her that she was at particular risk. What is particularly interesting about this decision is that the allegations of negligence were successfully grounded in the sexually discriminatory attitudes of the police officers investigating the case. It was their reliance on sexist stereotypical assumptions about women (eg that they would panic and become hysterical if they knew a serial rapist was operating in their area) that led to a decision not to warn potential victims, to the serious detriment of the applicant. In addition the Court (per McFarland J) emphasised that other aspects of the investigation, including a tendency to question the credibility of victims without proper reason, as well as the low prioritisation of the investigation in terms of resources (which the judge concluded stemmed from a belief that the relatively low level of violence accompanying the rapes rendered them less serious) also contributed to the poor quality of the investigation, establishing negligence. See generally E Sheehy (ed), Sexual Assault Law in Canada:
Law, Legal Practice and Women’s Activism (University of Ottawa Press, 2012).
14

protect women from sexual violence perpetrated by third parties.
73 In both jurisdictions Hill was considered but distinguished by the relevant courts.
That the claimants chose not to avail themselves of these resources has turned out with hindsight to be a wise move; but it might not have been had things gone differently. What is bold and brilliant about Green J’s curt dismissal of tort law is not that it reflects the pleadings submitted but that it allows him virtually to ignore the policy considerations which successive judges have raised against the imposition of civil liability on the police in precisely these kinds of circumstances. At the very least it weakens their normative grip. There is no reason why a rehearsal of the usual Hill policy concerns could not have been introduced to support a finding that the duty alleged either did not exist or did not apply in these circumstances. After all, it is clear—and this is acknowledged by Green J— that in delineating the scope of the positive duties imposed by Articles 2 and 3, the Strasbourg court has recognised that policy factors militate against the application of too stringent a duty.
74 In Osman , for example the Court observed: ‘such an obligation [to protect under Article 2] must be interpreted in a way that does not impose an impossible or disproportionate obligation on the authorities’.
75 After his lengthy trawl of the Strasbourg authorities Green J reiterates these concerns, observing:
… the process of determining whether an investigation was ‘reasonable’ or ‘capable’ of leading to the apprehension, charge and conviction of a suspect is a fact sensitive exercise. It is also subject to a margin of appreciation [in terms of the choice of means to secure compliance with the State’s obligation] and to proportionality.
76
However neither proportionality nor the State’s margin of appreciation prevent Green J from recognizing a breach of duty on the facts before him. And while lip service is paid to the kind of policy considerations which are determinative in Hill / Brooks / Smith , they are not determinative here.
This is because Green J has excised the core . The arguments which were formerly central to the legal disposition of this kind of claim have been moved from centre stage to the margins of the doctrinal framework and they no longer enjoy their formal status as doctrinally determinative. Once Green J decides to sweep away the tort debris that litters earlier cases, ably aided by the insistence of the Van
Colle/ Smith
77
court that tort and human rights are entirely different kinds of legal creatures, the path before him lies empty and ready to be filled. b) Move 2: Reconstruct the core from the penumbra
73 Carmichele v Minster of Safety and Security and Minister of Justice and Constitutional Development 2001 (10)
BCLR 995 (CC) ; Ghia van Eeden v Minister of Safety and Security [2002] All SA 346 (SCA) discussed in S
Ndashe, ‘The Duty to Protect Women from Sexual Violence in South Africa’ (2004) 12 Feminist Legal Studies 213.
74 DSD & NBV (n 3) [224]-[225], [241].
75 Osman (n 15) [116], cited in DSD & NBV (n 3) [144].
76 Ibid [224].
77 See (n 28).
15

With the settled law thoroughly excised from the doctrinal frame, Green J turns to what has been hitherto the ‘Plan B’ approach to bringing claims against the police in English law. In this context, the unpropitious outcome in Van Colle might be said to do little to inspire confidence. In that case, 78 a suit brought by the family of a witness in a criminal trial who had been murdered by the person against whom he was giving testimony was held not to meet the narrow criteria laid down in Osman 79 triggering the duty on the State ‘to take preventative measures to protect an individual whose life is at risk from the criminal acts of another’.
80 Interestingly in DSD & NBV , it is not the duty to protect which is invoked. The claimants are not arguing, as for example the claimants in Hill and Smith/Van
Colle did, that the police failed to protect them from the harm perpetrated by a criminal third party.
That argument may have been open at least to the later victim, NBV, because a number of times
Green J intimates that had the police done their job properly, Worboys would have been caught sooner.
81 This was also the argument mounted by the claimant in Doe.
82 However this kind of argument brings with it complex and often unresolvable problems of causation and inevitably discriminates between victims depending upon whether their assaults occurred early or late in the relevant time cycle. The claim in DSD & NBV is more akin to that in Brooks 83 , that is, it focuses on the harm generated by the investigation itself, except while in Brooks that claim was framed
(unsuccessfully) in terms of negligence, in DSD & NBV , it takes the form of an alleged violation of human rights. More generally both the duty to protect and the duty to investigate derive from the same positive obligations imposed upon the State by the ECHR to ensure an effective system of law enforcement for the protection of individuals’ Articles 2 and 3 rights.
After the decision in MC v Bulgaria
84
, in which the Bulgarian state was held to have violated a rape victim’s Convention rights by failing to enact and apply an effective criminal justice system which protected her rights to physical and sexual autonomy under Articles 3 and 8, some feminists, myself included, speculated about the possibility of invoking the ECHR framework to challenge police practices around the investigation and handling of rape cases in the UK.
85 However, there appeared to be a considerable gulf between the kinds of systemic failings exposed by the Court in MC
78 See (n 28).
79 See (n 21).
80 Van Colle (n 28) [24] (L Bingham). The House held that the Osman duty was not triggered because it could not have been reasonably anticipated on the facts that there was a real and immediate risk to life.
81 DSD & NBV (n 3). See in particular para 76: ‘The speed with which Worboys was identified and arrested and a search conducted of his premises which generated damning inculpatory evidence is a clear indication that had a similar focus been applied to the investigation during the previous years, Mr Worboys might have been prevented from committing at the very least a significant number of the assaults he did in fact commit.’
82 See (n 72).
83 See (n 24).
84 MC v Bulgaria [2005] 40 EHRR 20.
85 J Conaghan, ‘Extending the reach of human rights to encompass victims of rape’ (2005) 13 Feminist Legal
Studies 145. See in particular my observations that ‘one cannot help but wonder at what point, if any, a criminal law regime which delivers justice only to a small minority of women who report rape within a broader context of mass non-reporting (itself linked to perceptions of the (in)effectiveness of the criminal process) risks violating the ECHR’ (at 156).
16

v Bulgaria and the concerns raised by feminists about the operation of British criminal justice. DSD &
NBV is the first case to attempt to bridge that gulf: how did Green J do it?
Well, for a start, I don’t think he is a regular reader of Feminist Legal Studies , beginning his analysis with a lament about the lack of scholarly literature to guide him. To be fair his expressed concern was the lack of any case law or literature ‘offering a systematic synthesis of Article 3 case law on the duty of the police to investigate in circumstances such as present’.
86 Whether or not such literature is there to be found is neither here nor there. What is significant about Green J’s assertion is that it identifies a gap, a doctrinal hole, which he then proceeds to fill by considering in some depth a long succession of Strasbourg cases (fourteen in all and spanning a sixteen year period from 1998 until 2012) 87 from which he extracts, in true inductive fashion, a set of thirteen principles supported by the case law (which govern the existence, scope and application of the police duty to investigate under Convention law.
88
Time will tell whether this doctrinal exegesis will survive the ravages of future litigation strategists. It is certainly extremely thorough, described by a commentator in the
Solicitors’ Journal as ‘meticulously crafted’.
89 However, for our purposes, what gives it such normative power is that it purports to map hitherto uncharted legal terrain; it stands as the official picture of the field which will define and circumscribe the form and content of debate within that field thereafter.
90 Of crucial significance here is Green J’s insistence that he is confronting a clear line of authority, a mature and settled body of case law.
91 This is important because it helps to thwart any suggestion that Green J is indiscriminately ‘following’ Strasbourg jurisprudence rather than, as the HRA enacts, merely ‘taking
86 DSD & NBV (n 3) [140].
87 Green J’s case analysis ( Ibid [142]-[210]) includes detailed consideration of the following cases: Osman (n
15); Z v UK (n 15); Edwards v UK (2002) 35 EHRR 19; Menson v UK [2003] EHRR CD220; MC v Bulgaria (n
84); Szula v UK (2007) 44 EHRR SE19; Secic v Croatia (2009) 49 EHRR 408; Ali & Ayse Duran v Turkey
Application No 42942/08 (8 th April 2008); Beganovic v Croatia Application No 46423/06 (25 th September
2009); Denis Vasilyev v Russia Application No 32704/04 (17 th December 2009); Milanovic v Serbia Application
No 44614/07 (14 th December 2010); CAS & CS v Romania Application No 26692/05 (20 th March 2012); Koky
& Others v Slovakia Application No 13624/03 (12 th June 2012); Sizarev v Ukraine Application No 17116/04
(17 th January 2013).
88 DSD & NBV (n 3) [211]-[225].
89 See ‘Judicial Review update’ (Solicitors Journal, 8 April 2014) < www.solicitorsjournal.com/public/localgovernment/judicial-review-update> accessed 6 July 2014.
90 I use the term ‘field’ as Duncan uses it in Freedom and Constraint when discussing the extent to which law and legal materials operate as a constraint upon judicial choice: ‘The constraint imposed by law is that it defines the distance that I will have to work through in legal argument if I decide to come out the way I initially thought
I wanted to. “The law” constrains in that it is an element of the situation as I initially experience it. It is the
“field” of my action’; see (n 48) 31. In DSD & NBV, Green J has literally constructed the field from an apparently empty vacuum. He has defined ‘the distance’ subsequent judges will ‘have to work through’. At the same time Green J has reconfigured the broader field of civil liability for police actions and omissions; what appears as an isolated act of normative construction is in fact a form of subtle doctrinal colonisation.
91 DSD & NBV (n 3) [13], in which Green J explains his decision ‘according to well-established case law’ and in para 211 where he states: ‘In my view when construing the HRA and when taking account of Strasbourg case law, I am bound to accord significant weight to that case law because it is coherent, well evolved and its core tenets are settled’.
17

account of’ it.
92 It also counters the defendants’ claim that the case law is not settled, that the authorities relied upon by the claimants are ‘rogue’ and ‘maverick’, 93 and that the duty to investigate rests upon a line of argument that is ‘forensically weak and insubstantial’.
94 By constructing a set of clear and coherent principles, as it seems from scratch, Green J not only ensures that his interpretation of the relevant legal materials governs but also that the primary weapon in the defendants’ discursive armoury is effectively decommissioned.
95 Once it is established that what appears to be a sea of unstable and unconnected decisions mainly taken by a court far removed from England’s shores is in fact a highly coherent, fast evolving jurisprudence, then the obligation to take account of it, not
‘slavishly’ nor ‘subserviently’ 96 for this is England after all, but with due recognition of the weight to be accorded to a ‘line of consistent case law stretching back well over a decade’, is easily established.
97
And there you have it. From the wasteland left by the evacuation of tort law comes forth a new doctrinal framework which appears to be no more than the natural and inevitable result of the legal materials the judge confronts. The old core is consigned to irrelevance; the new core is plucked from the penumbra and refashioned as settled law. What was once presented as a reason for a rule now becomes the rule itself; in Duncan’s word, a right (in this case the right to have one’s allegation of rape properly investigated) becomes real, a factoid.
98
5 CONCLUDING THOUGHTS
There are many other aspects to DSD & NBV , legal and political, which deserve attention.
99 And of course it is too early yet to assess whether the gains of this particular strategy to compel better
92 HRA 1998, s 2(1): ‘A court or tribunal determining a question which has arisen in connection with a
Convention right must take into account any (a) judgment, decision, declaration or advisory opinion of the
European Court of Human Rights’. Reflecting on the nature of the obligation to take account, the Supreme
Court observed in 2010: ‘This Court is not bound to follow every decision of the European Court of Human
Rights… of course we should usually follow a clear and constant line of decisions’ (
Manchester City Council v
Pinnock [2011] 2 AC 104 per L Neuberger, para 48). In a speech subsequently delivered in August 2014, L
Neuberger cites this paragraph and warns against being ‘too ready to treat Strasbourg court decisions as if they were determinations by a UK court whose decisions were binding on us’ (Lord Neuberger, ‘The role of judges in human rights jurisprudence: a comparison of the Australian and UK experience’ (Conference at the Supreme
Court of Victoria, Melbourne, 8 August 2014) <www.supremecourt.uk/docs/speech-140808.pdf> accessed 14
November 2014).
93 DSD & NBV (n 3) [227].
94 Ibid [140].
95 The other key argument the defendants make is that the positive duties imposed by the ECHR, Arts 2 and 3 apply only where the police are directly implicated in the harm, citing a series of domestic cases in which human rights arguments have been applied in this context. However, as Green J rightly points out, the
Strasbourg jurisprudence, even on the strictest reading extends the positive duty well beyond these confines
( Ibid [237]).
96 Ibid [227].
97 Ibid [229].
98 Kennedy, ‘The Critique of Rights’ (n 41) 185.
99 For example, DSD & NBV is being cited in public debate as a reason to oppose alleged Conservative plans to scrap the Human Rights Act; See Nicholas Watt, ‘Rape case warning over Tory plan to rewrite human rights
18

policing of crimes of sexual violence outweigh the risks. Indeed, it is far from clear that the decision will stand although, writing some months later, I am not aware of any planned appeal by the defendants and indeed the court has since announced the amount of damages to which each claimant is entitled.
100 Moreover, subsequent cases have begun to take note of DSD & NBV. In a case before the Northern Irish High Court in early May, the judge denied a striking out application made by the
Police Service of Northern Ireland (PSNI) against a claim lodged in the context of a flawed rape investigation.
101 Interestingly, the claimant in C v CC of PSNI relied both on negligence and on a breach of her human rights. Moreover, the judge decided, partly in reliance on DSD & NBV , that neither claim should be struck out. Drawing on the comments of Lady Justice Hallett in a recent Court of Appeal decision 102 as well as some of the judicial discussion in Van Colle/Smith , Gillen J emphasised that Hill could not operate as a blanket immunity and there remained the possibility of recognising exceptions in some contexts. Nor was he overly fond of the separatist thesis mounted by
Nolan and some of the judges in Smith/Van Colle . Indeed Gillen J went so far as to remark that:
At this stage I am unable to conclude that a breach of Article 3 may not at least be a contextual factor and arguably help to inform the issues at common law … In particular it may feed the argument that the facts of this case fall within the list of possible exceptions to the Hill core principle.
103
Gillen’s J’s conclusions were clearly influenced by the egregiousness of police failings and the vulnerability of the victim, C, who was mentally disabled. If this kind of litigation does nothing else it appears finally to be rousing the judiciary to face up to problems which feminists have been striving to address for some time. But for our purposes, what is so delightful about C v CC of PSNI , is that the old core, consigned to irrelevance, has reappeared - penumbral to the primary focus on human rights but worth exploring nevertheless. The greatest irony then of Green J’s curt dismissal of tort is that it can only reinsert itself into proceedings in reinvented form: what it cannot do (at least not easily, although never underestimate the strategic possibilities, especially in the hands of a skilled tactician) is continue to maintain an impenetrable doctrinal barrier between the failings of the police law’ The Guardian (London, 27 August 2014) < www.theguardian.com/law/2014/aug/27/rape-cases-torieshuman-rights-law > accessed 7 September 2014.
100 DSD and NBV v Commissioner of Police for the Metropolis [2014] EWHV 2493 (QB). Green J awarded
£22,500 and £19,000 to DSD and NBV respectively (23 July 2014).
101 C v Chief Constable of the Police Service of Northern Ireland [2014) NIQB 163.
102 Robinson v Chief Constable of West Yorkshire Police [2014] EWCA Civ 15 in which Lady Justice Hallett speculated about the scope of the ‘list of possible exceptions’ to the Hill principle (at para 49).
103 C v CC of PSNI (n 101) [38]. As this article goes to press, we continue to await the publication of the
Supreme Court judgment in Michael v Chief Constable of South Wales Police, Michael v Chief Constable of
Ghent Police , which was heard before a panel of seven in July 2014. Michael , which involved the murder of a young woman by her ex-partner after police failed to respond promptly to an emergency 999 call, alleged both negligence and breach of Ms Michael’s Art 2 rights on the part of the police. The Court of Appeal [2012]
EWCA Civ 981 upheld the decision of the trial court not to strike out the Art 2 claim but granted judgment in favour of the Chief Constables in relation to the negligence claim
19

and the consequences of those failings for people caught up in various ways in the criminal justice process.
Green J’s decision purports to say nothing about tort; but in its very silence, it says a lot. It says that human rights claims are outside the doctrinal straitjacket of Hill ; that Hill and Broo k notwithstanding, crime victims can sue the police for investigative failures; that civil liability is an appropriate means by which to call the police to account for their actions. By saying these things, the judgment in DSD & NBV subtly undermines much of the rationale supporting judicial subscription to the Hill immunity and weakens its normative power. After all, if liability can be established via another route, the police are far from immune, and the Hill policy objectives are not achieved. This gives them less purchase and opens the way for their reassessment. In effect, Green J has pulled the rug out from under Hill ’s feet. The outcome of DSD & NBV is to make Hill look less like an ‘open and shut’ rule 104 within what Duncan would describe as an ‘impacted field’, that is a field in which the doctrinal trajectory is perceived as a long straight line from which there are few or any deviations, 105 and more like a ‘contradictory field’ in which there are ‘lots of cases on both sides’ and ‘a given policy appears to vary widely in force’.
106 It remains to be seen what future courts will do with this delicious uncertainty. In the meantime and on behalf of critical legal strategists everywhere, thank you
Mr Justice Green!
104 Kennedy, ‘Freedom and Constraint’ (n 47) 16.
105 Ibid , 44-45.
106 Ibid, 47.
20