The Hidden Disciple: Towards a Christian Ethics of Spying

The article explores the understudied subject of the distinctly Christian ethics of human intelligence (HUMINT) and considers how a Christian intelligence officer (IO) can draw on the robust and diverse tradition of Christian ethics to make their secular vocation compatible with the ethical principles of their faith. The current intelligence ethics literature is dominated by the Just Intelligence Theory (JIT), an adaptation of the just war tradition, which offers many valuable contributions. However, I propose the enrichment of JIT by discursive ethics from within the respective Catholic and Reformed theologies. I also consider the originally non-Christian concepts of the ethics of care and the I-Thou ideal through Christian hermeneutics. I conclude that the IO's ceaseless questioning of their conduct and its sources, and the continuous striving for primarily personal relations, as opposed to institutional relations, is the essence of the ever-imperfect process of becoming a Christian IO. Such an IO is a hearing and critical disciple who does not pretend to know the mind of God, but one whose conduct is directed by their critical reflection on their Christian faith and on the institutional setting wherein they operate. Paradoxically, this presents an enormous ethical possibility for a Christian IO.

Alvarez writes bluntly that 'Pope Pius soon found himself dangerously enmeshed in espionage'. 10 Pius XII's actions and Vatican's multifaceted involvement in the world of espionage are less surprising once we consider Aaron Bateman's conclusion that the Holy See 'serves as a valued source of information for Western diplomatic services and remains a target of foreign intelligence organizations'. 11 Many Vatican-affiliated and reformed Christian organizations have also been involved in intelligence gathering and in facilitating contact between intelligence agencies and agents. Since at least the Second World War to this day, many Christian missionaries and representatives have also been spies, as Matthew Sutton convincingly demonstrates. 12 This brief historical survey indicates that the respective worlds of espionage and Christianity are in fact not two separate realms and that there may be many Christians across the world today who are affiliated with various foreign intelligence agencies in one way or another. Their names may be unknown, their faces hidden. But as Christians, they might opt not to join the procession of vain fools, as le Carré's Leamas would have them, and instead strive to make their vocation compatible with the ethical principles of their faith.

Just Intelligence Theory: Contributions and Limits
Intelligence ethics has developed into a solid academic field which, I argue, would profit from greater interdisciplinarity and diversity of thought. Intelligence ethicists have sought to develop various original ethical frameworks for HUMINT. For instance, Adam Diderichsen and Kira Ronn propose one based on policing standards of legitimacy and public consent. 13 Alternatively, Robert Frisk and Linda Johansson draw analogies between intelligence collection and the Trolley Problem. They consequently apply the preventing-allowing-doing harm (PAD) approach to derive what they call 'intelligence ethics strategies' ranging from the Borg (power) to the Viking (glory and riches) to the 10 Samurai (honour). 14 Neither Diderichsen and Ronn's call for more open intelligence activities, nor Frisk and Johansson's idea to classify intelligence activities according to PAD, introduce profound new considerations for the ethics of HUMINT on the individual level. Moreover, these approaches have yet to receive more prominent scholarly attention which is instead focused on an adaptation of Just War Theory (JWT)-Just Intelligence Theory (JIT). 15 Given JIT's prominence in the current intelligence ethics literature, I consider it in some detail. I very briefly summarize the original just war tradition to provide grounds for a subsequent more detailed discussion of JIT and its contributions and limits to guide an individual IO's conduct in HUMINT. From its Christians origins to the current human rights-centred conception, the just war tradition offers a rich body of moral, legal and political reasoning to justify and set limits on the use of armed force. 16 It sets out principles that would need to be met in order for war to be both declared and fought ethically. 17 The mainstream theory (although there is not a univocal agreement) distinguishes between the practical ethical criteria that govern the resort to war ( jus ad bellum) and the conduct of war (jus in bello). 18 As Charles Matthews observes, the tradition is both heuristic and hermeneutic, combining normative and interpretative frameworks of ethical deliberations. 19 The tradition's foundational elements are rooted in the Christian virtue of caritas, or neighbourly love, as set forth by early Christian thinkers Ambrose and Augustine. These   21 This love both brings the obligation to defend one's neighbour and imposes limits on this defence. The centrality of love has led Ramsey to focus particularly on the jus in bello principles, leaving the details of many other parts of JWT unexplored-an opportunity for later theologians such as Nigel Biggar and Oliver O'Donovan. 22 With the rise of the nation state, a secular approach to the originally Christian just war tradition has gradually developed and the tradition has evolved into a formal list of collated principles known as JWT. 23 Its post-World War II revival and redefinition has been driven by both Christian and secular thinkers, sometimes in a mutual dialogue but often rather isolated from one another. Particularly Michael Walzer's influential secular reformulation in the 1970s, and his and others' work since, have aspired to make the theory applicable to modern, lowintensity warfare, human rights interventions and counterterrorism operations, often with the involvement of non-state actors. 24  just war theorists have criticized many of JWT's key assumptions, particularly the supposed moral equivalence of combatants, the immunity of non-combatants, and the tension between the state-centred concept of public war and the individual-centred principle of discrimination. 25 The classical theory's robustness and the revisionists' focus on the individual each offer helpful insights for my later discussion about a Christian IO's moral agency. In their necessary search for an ethical intelligence framework, scholars and former practitioners have naturally turned to this rich and complex just war tradition. JIT, based on a supposed analogy between war and intelligence, has quickly become a dominant stream in intelligence studies. 26 It does not seek to rule out any action a priori; rather it strives to provide criteria against which each action can be judged. 27 The just war tradition is unlikely to bring either justice or peace, but rather to limit the frequency and the means of fighting, as Michael Walzer writes; in the same way, JIT aspires to set limits in the whole intelligence cycle of acquisition, analysis and sharing of secret information. 28 Thus, JIT's proponents have argued that, pending 'some adaptation', the just war principles have direct usefulness for intelligence operations. 29 Just war's adaptation to regulate intelligence proposes a set of criteria whose division mirrors the usual division in JWT. It is these two jus in intelligentia criteria that merit detailed consideration here as they directly concern an individual IO's conduct. But first, let us examine the central assumption behind JIT, the analogy between war and intelligence.

The State of War and a State of Intelligence
In adapting JWT for intelligence, JIT proponents such as David Omand painstakingly seek to reconcile the distinction between the exceptional state of war and standard intelligence gathering which takes place ceaselessly. 31 Omand acknowledges that while the just war conditions govern the transition from peace to war, intelligence is continuous. To make his case for a parallel between war and intelligence, Omand cites the precedent of permanent nuclear deterrence. He writes that the maintenance of a serious intelligence capability during peacetime helps maintain peace as much as nuclear deterrence does. 32 Furthermore, Omand draws not only on the international humanitarian law which applies during hostilities, but also on the more widely applicable International Human Rights Law. 33 Omand thus succeeds in somewhat broadening the utility of JWT. By narrowing the seemingly unbridgeable considers actions with simultaneously good and bad outcomes, claiming that to intend the good outcome is justifiable but to intend the bad outcome is wrong. In practice, IOs lying to friends and family is justifiable if the intention is to protect the loved ones and if the harm of the deceit is outweighed by the harm that would result from them knowing the truth. O'Donovan in particular asserts that the DDE makes it clear that there is no moral difference to the intentional character of an act whether the harm is intended as an end or as a means. gap between an exceptional state (war) and a normal state (intelligence gathering), he partially demonstrates the suitability of applying JWT to intelligence. 34 But a gap in the analogy remains. Not least because, as Ronn and Mark Phythian argue, intelligence operates continuously in a proactive and preventative manner even when no imminent threat exists. 35 Even with Biggar's understanding of threat (when one has good reason to suppose it is substantial) which is more permissive than O'Donovan's (on the verge of becoming actual), intelligence collection is more deeply preventative than just war theorists generally permit military force to be. 36 Armed conflict, O'Donovan asserts, is an 'extraordinary extension of ordinary acts of judgments'. 37 If HUMINT collection is to be regulated by the same criteria as war, it follows that HUMINT, like war, is exceptional and extraordinary. But while such characteristics may be neat and useful from an academic point of view, they do not correspond with the actual praxis of HUMINT at all. 38 JIT, perfectly suitable for extraordinary situations, is helpful but insufficient on its own for addressing the routine, daily, ethical challenges a Christian IO encounters in their professional and personal life as a spy. Therefore, Diderichsen and Ronn conclude (and I concur), that JIT is a valuable framework for highly specific activities against a known and imminent threat, but it is not a sufficient and complete framework for continuous intelligence gathering. 39 Having discussed the just war tradition, JIT, and the merits and the limits of the analogy between war and intelligence, let us now consider each of the two jus in intelligentia principles as grounded in the just war framework, and as adapted to just intelligence. 34. Interestingly, many scholars note that the post-9/11 sense of insecurity has reversed the norm, blurring the distinction between war and peace.

Discrimination
JIT requires intelligence collectors to discriminate between targets. 40 The main challenge to the principle of discrimination in JIT derives from the gap in the analogy between war and intelligence, namely the regular nature of the praxis of HUMINT. In the routinized HUMINT, the very act of collecting information about whether a target is legitimate already raises crucial ethical questions. Therefore, HUMINT weakens the central JWT distinction between knowledge (prerequisite) and the infliction of harm (action). Since action first requires knowledge, the otherwise helpful JWT discrimination criterion becomes circular when applied to HUMINT. 41 This circularity can be resolved, according to Kevin Macnish, by relying on 'pre-existing evidence … available prior to the decision'. 42 But relying on 'pre-existing evidence' only reinforces the circularity by positing that one needs intelligence-the goal of HUMINT-before engaging in HUMINT.
Discrimination means to separate the innocent from the guilty and to proportionate the means to the object. 43 But HUMINT, like other intelligence collection methods, in a way precedes discrimination. Furthermore, even once information becomes available, the distinction is blurry, as is made abundantly clear by the practical experience of HUMINT. Is the prime minister of a foreign country a legitimate target of deception and lies; is their spouse, their advisor, their friend, who all possess crucial intelligence? Is the IO's father, their friend, or child a target, so that the IO can manipulate and lie to all of them (as they do to maintain their cover)? I do not see how the principle of discrimination could, by itself or in combination with other JIT criteria, practically help a conscientious Christian IO answer these and similar ethical challenges that they encounter daily. 44 Tony Pfaff and Jeffrey Tiel try to address the circularity differently, by proposing a targeting framework based on consent, which they consider 'the manifestation of the freedom of the person'. 45 Their 'interesting way of assessing innocence', 46  to Angela Gendron, posits that individuals represent a legitimate target to the extent and to the degree to which they willingly consent to enter the world of espionage. The Pfaff/ Tiel framework thus prohibits any but the least intrusive espionage against ordinary citizens. Conversely, it permits deceit, incitement and blackmail against other intelligence professionals who have consented to 'play the game'. 47 Ross Bellaby develops the framework and summarizes it, in resemblance of the just war tradition, by claiming that Pfaff/ Tiel's use of the principle of discrimination 'distinguishes between those individuals without involvement in the threat …, and those who have made themselves a part of the threat'. 48 Pfaff and Tiel admit that Immanuel Kant, on whose ethics they build their consentbased discrimination criterion, 'saw deception as inherently wrong'. 49 But they tendentiously exclude the players of the spying game from the restriction, while (unrealistically) maintaining that the same IO who lies in the professional context cannot ever lie to their family and friends who have not consented to play the game. Thus, Pfaff and Tiel run into two contradictions, one theoretical the other practical, both of which they fail to resolve. First, theoretically, by replacing the Kantian imperative with informed consent they apply Kantian ethics selectively and inconsistently, implying that playing the spying game strips one of their personhood. Second, practically, they propose a framework which would place the IO into an unresolvable situation vis-à-vis the many non-players they encounter and engage with. 50 Pfaff and Tiel's flawed argument builds on the assumption that a state's duty to protect creates a supreme moral obligation to teleologically suspend the ethical. 51 This state/ power-centric line of reasoning leads to unconvincing arguments such as Joseph Hatfield's who claims that treason by means of espionage is just when 'the political community being betrayed fails to secure for its citizens basic human rights, economic and physical security'. 52 Focused on the state power, Hatfield, an active-duty US intelligence officer, chooses to ignore that his criteria would morally justify, and even oblige (as he himself insists), disadvantaged social and/or economic groups to spy against their own government in many a democracy.
A similar etatism also appears among Christian advocates of JIT. Consider Darrell Cole, one of the most prominent among them, who writes that spying like 'the use of 47 Hatfield grounds his state-centric moral reasoning on the equally problematic Aristotelian assumptions that (1) the supreme goal of human life is eudaimonia, which is (2) not achievable outside of a political community. It follows for Hatfield from (1) and (2) that the political community must be protected at any cost.
force takes its moral character from those who authorize it'. 53 But, as I shall argue later, a Christian IO seeking to justify their actions by appealing to their state would necessarily elevate the gifted temporal authority above the eternal Divine authority and idolize a created government instead of worshipping the Creator. In other words, the nature of the ordering authority is not the morally crucial variable. Even Biggar's robust defence of torture in extreme cases acknowledges that a systematic programme of torture is no more or less legitimate whether ordered by a supposedly democratically elected government or by an authoritarian state. 54 And one must remember that HUMINT, unlike war, is a systematic programme of continuous intelligence collection that involves routine lying and manipulation in both the professional context and outside of it. Cole himself backs down in light of this fact. He first praises O'Donovan's criticism of the Rawlsian liberalism which presents liberal society as a good to be sought above God. Then, Cole acknowledges that 'for Christians the interests of the wellordered regime can never be given a value higher than God's will'. 55 But in attempting to somehow set a state-centric moral standard for spying, Cole runs into an impasse wherein the former imperative renders the latter attempt self-defeating.

Proportionality
Second, the criterion of proportionality in intelligence, Bellaby explains, will admit only information which contributes to the just cause; this is a positive in the proportionality calculation. Conversely, harms to the target, the agent, or society in general, are to be taken as negatives. 56 Michael Quinlan argues that harm may be necessary to protect those for whom a state carries responsibility, and Omand adds that 'under certain circumstances, even the individual's fundamental rights … may have to be sacrificed so that the rights of others may be protected'. 57  On this proportionality balancing sheet, considerable benefits are to be weighed against all reasonably foreseeable costs of the action. This view is very cautiously embraced in John Paul II's 1982 message that 'peoples have a right and even a duty to protect their existence and freedom by proportionate means against an unjust aggressor'. 59 In practice, the assessment of proportionality in JIT is often closely linked to legality. Omand describes proportional intelligence collection as only that which is 'necessary to achieve lawful objective'. 60 Omand thus aims to 'give teeth and therefore meaning' 61 to the now discarded Turner test of intelligence ethics by claiming that 'there is one overall test of ethics of human intelligence activities. That is whether those approving them feel they could defend their actions before the public if the actions became public'. 62 Biggar doubts, however, that a balancing sheet is possible because it strives to compare incommensurable goods and evils. 63 (2021), to which Omand contributed significantly, operates with the principle of proportionality throughout. For instance, when deploying espionage, proportionality 'involves balancing the seriousness of the intrusion into the private or family life of the subject of the operation (or any other person who may be affected) against the need for the activity in investigative and operational terms'. Other proportionality principles in the code demand balancing the size and scope of the proposed activity against the gravity and extent of the perceived crime or harm; explaining how and why the methods to be adopted will cause the least possible intrusion on the subject and others; evidencing, as far as reasonably practicable, what other methods had been considered and why they were not implemented, or have been implemented unsuccessfully. 61. Omand and Phythian, 'Ethics and Intelligence', p. 57. 62. A CIA director under the Carter Administration, Stansfield Turner, built on Sherman Kent's famous dictum that spies ought to be 'good men, making good judgements'. Ronn cautions that the demand to assess the reaction of the public places an unreasonable burden on an IO. More problematically still, this so-called Turner test is notoriously unreliable as it lacks firm anchorage and relies on the fallible and fickle public to judge one's actions. First, the public is not a homogenous entity but rather a diverse conglomerate of diverging, and often conflicting, views. Second, countless historical cases come to mind where the majority of the public approved of evidently morally problematic actions. In other words, the rather subjective Turner test relies on purely context-dependent empirical guidance of the norms of the society one lives in. of a balancing sheet, calling it a 'false choice, masquerading as virtuous discernment of relative harm'. 64 Therefore, Stoddart cautions, proportionality involves an inherent risk of attempting to objectively measure factors that have subjectivity and value beyond any measurement. 65 Macnish warns that the coercion of one person (cost) weighed against national security (benefit) means that the person will always lose, which is a 'counter-intuitive state of affairs'. 66 Thus JIT aspires to provide a legitimation framework that, Alexander Fatic observes, 'eliminates the personal drama of decision making'. 67 But the personal drama 'should not be eliminated or reduced-it should be encouraged', 68 concludes Fatic. The personal dimension is particularly important in the practice of HUMINT with the institutionalized, instrumental handler-agent relationship at its core. The value of an agent for a handler is the function of the agent's access to the desired information. 69 For Pfaff and Tiel, as discussed above, such interpersonal instrumentality may be acceptable among players who have knowingly entered the game of espionage, while a strictly Kantian deontologist would denounce any such actively exploitative relationship as ethically indefensible. 70 Macnish explores the person-centred Kantian route and argues that HUMINT methods 'are an assault not just on a person but on personhood, on what it means to be a person (at least in Kantian terms) in terms of autonomy'. 71 The individual agent, Macnish continues, is overlooked for the greater good of state security. Crucially, Macnish observes that the negative effect of such depersonalizing assault extends beyond the individual concerned to the broader community whose autonomy is thus restricted. 72 In Macnish's critical summary, the harms to an individual involved in HUMINT 'go to the root of what we take a person to be, and they extend to the community of persons at large'. 73 Beyond Just Intelligence Theory Arthur Hulnick and Daniel Mattausch caution against excessive reliance on state-centred justifications which relieve a 'conscientious intelligence professional' 74 of their ultimate responsibility for their conduct. I argue, alongside Bryan Hehir's concluding remarks on the just war tradition, that JIT should be valued and retained, but it also needs to be expanded and enriched with other approaches. 75 The enrichment of JIT's value in HUMINT is precisely my aim in the rest of this article. Accordingly, let us approach the handler-agent relationship increasingly as one between persons whose personhood is constituted through contextual I-Thou interactions. For, as O'Donovan writes, 'in the eyes of God, the soul of a soldier is of no less value than the soul of a milkman'. 76 The proposed shift of focus to the individual echoes revisionist just war thinking which emphasizes individuals' agency and their conscience. 77 To enrich JIT, I propose the introduction of fundamentally discursive, contextual and relational ethics of care which considers how embodied human beings are tied together in a web of mutually constitutive engagements. 78 Conceptualized more broadly still as a critique of relations of power and dependence, care ethics is critical in challenging established hierarchies, structures, power relations, and general patterns of othering. 79 Offering an alternative to normative theories providing principles of conduct, the ethics of care seeks to 'disclose what is being done and the possibilities that might be available for alternative actions'. 80 The IO is therefore invited to question the established HUMINT practices, and the sources thereof; to raise selective conscientious objectionsnot to categorically reject HUMINT, but to critically examine their own conduct, institutional practices and contextual settings. 81 Fiona Robinson has convincingly argued that the critical ethics of care has profound relevance to ethical questions in international relations (IR). 82 The relevance in IR includes espionage since, as I have demonstrated, the current mainstream intelligence ethics often tend to approach spying in the context of national interest, power and legitimacy; consider different methods employed by spies through a cost and benefit analysis; and focus moral discussions on an institutional set-up and causes for action. In contrast, my approach thereafter builds on Robinson's critical challenge. I am going to argue against the primacy of a nation state for a Christian IO and in favour of questioning established structures. Further, I will invite the IO to deinstitutionalize their relationship with their agent, striving instead for a subjective I-Thou encounter where an IO's personal responsibility, grounded in God, is constitutive. Thus, I will assert that what Robinson calls 'ethical possibility' 83 lies not only in states and institutions, but primarily in the individuals who form them; hence my focus on an individual IO who, in a sense, personifies this ethical possibility. 84 Within this ethical possibility, a distinctly Christian IO would recognize what Biggar refers to (pace Grotius) as a 'universal, natural, moral law that commands the attention of consciousness even where there is no positive law and are no courts'. 85 Such a Christian IO can become what I will finally call a hidden disciple.

Espionage and the Bible
To ponder the ethics of the IO-agent relationship through a Christian lens, I first turn to several biblical texts that involve spying. From the 'honey trap' tactics of Delilah who seduces Samson to learn his secret, 86 to Joseph's deception to hide his identity from his brothers, 87 to 'agents provocateurs' who challenge Jesus' loyalty, 88 the Bible remains rather uninterested in directly commenting on the morality of espionage. 89 This lack of direct judgment also characterizes the Mosaic stories where Moses initially castigates the Israelites for their lack of faith in the Divine plan manifested in their decision to send human beings. It is also important to distinguish care ethics from the traditional human rights-focused ethics. While related, the latter tends to overemphasize human security at the expense of caring values and relations of care. For a balanced discussion of this criticism, see Michael Barns  spies, only to later dispatch spies himself to collect information about the Promised Land. 90 In yet another example, Joshua sends two spies on a reconnaissance mission from Shittim to Jericho. The prostitute Rahab hides the spies in her house and lies to representatives of the king of Jericho to protect them; they return safely and provide Joshua with crucial information. 91 Rahab's example is relevant even though the Canaanite is outside of the covenant obligation not to bear false witness, because her deceitful action is arguably condoned in the text. Rahab's story has thus become an oft-used biblical justification of espionage. Consider James Olson, a Roman Catholic with thirty years of experience in the CIA Counterintelligence Directorate, who says: 'Rahab is one of the great heroines of Israel, and if spying were inherently evil, why would she be honoured and blessed for protecting the spies? So that gives me a lot of consolation, that gives me a lot of belief that what I'm doing is morally justified'. 92 Regardless of Olson's enthusiasm, none of the stories in fact offer us a direct parallel to the IO-agent relationship. A balanced reading of the biblical texts does not provide us with a direct and definitive ethical judgement on spycraft. And yet, the craft clearly involves practices that are generally considered unethical. Accordingly, our first guide on the exploratory voyage towards a Christian ethics of spying will be a master of ethical paradoxes, Søren Kierkegaard, who argues that one can be an exemplary Christian precisely by acting unethically.

A Spy and a Knight
Kierkegaard uses the biblical story of Abraham to argue that, under certain conditions, the ethical can be teleologically suspended. He writes that by accepting God's command to sacrifice his son, Abraham 'overstepped the ethical entirely and possessed a higher telos outside of it, in relation to which he suspended the former. … He did it for God's sake because God required this proof of his faith'. 93 For Kierkegaard, Abraham is a 'knight 90. Crucially, even seemingly passive observation discussed in the texts and practised by today's spies implies a great deal of deceit to maintain cover. Compare for instance Deut.  95 The tragic hero deifies the ethical, one might say the normative, in the temporal, but the knight of faith 'gives up the finite in order to grasp the infinite'. 96 Kierkegaard's paradoxical approach to Abraham's story has been contested from multiple angles. Two of these angles are particularly relevant to the subject of my inquiry.  read in this context, 'God's command to sacrifice Isaac cannot be understood as Kierkegaard does'. 104 Rather than seeing Abraham's dilemma dialectically, Stump highlights that God's demand for Isaac and the requirements of morality are on the same side and Abraham is called to believe in the goodness of God and that by obedience, Abraham is not ending Isaac's life. Although Stump acknowledges that Kierkegaard's reading of the story is right in its description of the knight of faith, she concludes that there is no teleological suspension of the ethical. 105 Aware of two of the major problems in Kierkegaard's argument, let us proceed in our discussion and suppose that Kierkegaard's conclusion holds, and that Abraham perhaps teleologically suspended the ethical in responding hineni! (Here I am!) to the call of God. Could this ethical suspension justify unethical conduct on the part of a Christian IO? To attempt to answer, let us now turn to O'Donovan, and to his commentators, to explore the role of the state in a Christian's social and political life. For it is on behalf of their state that the IO is called to dutifully answer hineni! and behave unethically. 106

A State and God
Rather than exploring O'Donovan's complex and contentious political theology in full, I focus only on those elements directly relevant to our concern-the respective roles of the state and God in an IO's conduct. O'Donovan follows the Augustinian tradition of two cities which he considers to be 'fundamental to Christian political thought'. 107 He asserts that the authorities of the civitas terrena are hierarchically subjected to God's sovereignty in the Exaltation of Christ who 'having disarmed the powers and authorities, [he] made a public spectacle of them, triumphing over them by the cross'. 108 Christ's Resurrection, continues O'Donovan, has ever since posed a challenge to the primacy of the state, or later a nation-state for 'at the name of Jesus every knee should bow'. 109 Commenting on O'Donovan, Biggar concurs that Christians owe their primary loyalty to God, the source from which governmental authority derives. 110  further still to argue that although Christians are subject to the jurisdiction of the nation state, they may choose not to appeal to secular state institutions when they perceive a conflict with their religious commitments. 111 Hauerwas and O'Donovan agree in principle that Christians ought to choose whether their primary allegiance lies with God, or with the state authority. 112 But O'Donovan does not go as far as Hauerwas who proposes a catholicity without Leviathan, an 'alternative politics to the politics of the world'. 113 Hauerwas emphasizes the incompatibilities between the Christian polis and a liberal state, whereas O'Donovan sees the two cities as complementary rather than alternative, until Christ's Second Coming. This leads O'Donovan to explore an ideal modus vivendi for the coexistence of the two cities in time and space-although with insufficient emphasis on the importance of the non-Christian identities of a believer, assert Nicholas Wolterstorff and James McElvoy in their respective commentaries. 114 In perceiving earthly governments as derived, O'Donovan neither calls for a theocracy, nor argues for anarchical rejection. 115 He cautions that Christians are taught by Jesus' own example to pay their taxes. 116 But, for O'Donovan, the danger lies in idolizing an earthly state and seeing it as sufficient for human welfare and, implicitly, for salvation. 117 William Schweiker agrees in principle; however, he cautions that O'Donovan's argument tends 'towards a dyadic view of social reality, a homogenous religious conception'. 118 Even a milder form of the argument, digestible for Schweiker, makes O'Donovan's Christian critique of, first, the primacy of the state and, second, of the self-appointed salvific power of a state, a formidable challenge to state-centric, legalistic and normative justifications for intelligence when conducted by a Christian. 119 Unlike Kierkegaard, O'Donovan warns against radical individualism. But when the institutions of government morally fail, individual faith can be decisive. 120 A Christian IO firmly grounded in the Christian community and tradition, however imperfect and diverse that community is, and however clashing with the IO's other communal (often fluid) identities, can draw on resources to direct and question their conduct and the demands of their intelligence agency or their state. Such questioning should look in two directions, according to Biggar's reading of O'Donovan: upward to the common good, and downward to social responsibilities. 121 As such, an IO's individualism can in effect be profoundly other-centred and communal. Clearly then, strong individualism thus conceptualized would be the beginning, not the end, as the reading of Kierkegaard might lead one to believe, neither radical nor antisocial. 122 To stick to the Kierkegaardian language, the individual reaches a still higher level of existence above the tragic hero and even above the knight of faith. Humbled by their high standing in absolute relation to the absolute and driven by the Divine command to love their neighbour as they would love themselves, the conscientious Christian as if descends from the knighthood-armed but not satisfied-to contribute to the common good precisely through their individual responsible actions.
It has become evident that even if we concede with Kierkegaard that the ethical can be teleologically suspended, it cannot be done so to serve the state-an authority derived from that of God. And yet, many a Christian has spied for a state; many a and cultural impoverishment'. Crucially, for O'Donovan, authority is vested in creatures, not mediated as in a neo-Platonic account. Christian has considered the state to be so important that they might have suspended the ethical on its behalf. Prominent among them is the aforementioned theologian, pastor and spy, Dietrich Bonhoeffer, who not only theorized but also 'got his hands dirty'. 123 Let us therefore seek further elucidation in, first, the dirty hands justification in general as famously presented by Walzer, and second, in Bonhoeffer's own Christian view thereof.

Hands Dirty and Dirtier
Walzer opens his essay 'The Problem of Dirty Hands' by acknowledging what Christians, aware of the postlapsarian situation, know all too well: That to govern wholly innocently is impossible. 124 Therefore, the moral dilemma is supposedly not one of a binary choice between hands dirty and clean, but rather one of degree set in a concrete context. Only a discursive analysis of the context can provide moral substance to Hamlet's self-justification that he 'must be cruel only to be kind'. 125 Walzer's analysis explores the Machiavellian, Weberian and Catholic respective traditions of political justice, only to encounter a problem which Walzer concedes but cannot overcome within his secular framework. 126 Without an ultimate judge, writes Walzer, 'there is no one to set the stakes or maintain the values except ourselves'. 127 Commenting on Walzer, Dallas Gingles writes that Walzer's is an account of (possibly) political justice without final justice, in the Christian sense or otherwise. 128 This is why O'Donovan argues that earthly judgement must derive its authority from Divine judgement as a barrier against hands made ever dirtier by radical moral relativism and cynical, selfish individualism. Underestimating this risk, Walzer sides with Rawls in claiming, against my Christian argument, that liberal constitutional democracy is a supreme good to be valued above all else. In doing so, Walzer precisely makes the mistake against which O'Donovan warned above: he idolizes a 'temporal political order as a good to be sought above and against a permanent and the final good-God'. 129 Unlike Walzer, Bonhoeffer ties dirty hands to Christ-oriented responsibility, aimed not at achieving political justice, but at loving as Christ loves. 130 In Bonhoeffer's words, a responsible act is 'exclusively concerned about the other human being … it springs from the selfless love for the real human brother or sister'. 131 Bonhoeffer thus calls on Christians to act responsibly for the common good and flourishing of this world for which Christ died. Such Christ-centred individual responsibility concerns a decision to act, but not to judge. Bonhoeffer and Walzer agree that judgement must come from the outside; but whereas Walzer struggles to locate the judge, Bonhoeffer appeals to God. Walzer's secular, strictly political judgement is cataclysmic; Bonhoeffer's theological grounding of judgement overcomes Walzer's tragicality. For the final judgement in Christian eschatology does away with any relativism, however disputed the nature of that judgement in theological debates and across Christian traditions may be. Reflecting on his ten years of anti-Nazi resistance during which he did indeed get his hands dirty, Bonhoeffer expressed his hope that since his conduct was a bold venture of faith, 'God promises forgiveness and consolation to the man who becomes a sinner in that venture'. 132 His argument resembles the Christian conception of the love-centred just war tradition; for in the penultimate (in 'that which is before the last' 133 ) the use of force can be not only morally defensible, but even praiseworthy. 134 It is important to note that Bonhoeffer recognizes dirty hands in rare cases and as a sacrifice offered to God, not unlike in Biggar's case for torture. Therefore, the concept of dirty hands provides helpful guidance for HUMINT particularly in immediate existential danger, not unlike JIT. But an ethics that guides an IO's daily interactions with their agents and with others needs further richness still; let us seek it by going deeper into Bonhoeffer's teaching to explore his notion of discipleship.

A Spy and a Disciple
In tying one's actions to ultimate judgment and in appealing to the sovereignty of God, Bonhoeffer creates an important link between his teaching and O'Donovan's political theology. It follows for Bonhoeffer that the state, which is only God's minister, forfeits its legitimacy if it poses a barrier to discipleship-following Jesus not as a teacher, but as the Christ to whom all worldly legality is subjected. 135 The almost Hauerwasian subtone of anarchy is silenced, however, by Bonhoeffer's simultaneous appeal that the best way to be Christ's disciple is to manifest the discipleship through active life in the polis. Bonhoeffer notices that Jesus calls not himself but his disciples the salt and the light of the world. 136 Importantly, Jesus also gave disciples the power to act as they were called to. 137 Therefore, for Bonhoeffer, discipleship must be visible in action with the Incarnation providing 'the ultimate reason why the service of God cannot be divorced from the service of man'. 138 Importantly, however, discipleship can be maintained only to the extent that the world does not come between Christ and the disciple. The IO thus ought not to become an IO first, and a Christian second; on the contrary, the Christian discipleship provides practical and ethical direction to their secular calling. 139 Bonhoeffer recognizes that such discipleship is not an accomplished achievement, but rather a lifetime of strenuous effort with many lapses and occasional small victories. Indeed, Bonhoeffer notes that St Peter received the call 'Follow me' on two separate occasions separated by a whole lifetime. 140 Despite the pains and the difficulties, Bonhoeffer appeals, discipleship is not an option, 'but the Divine command to all Christians without distinction'; 141 it is an 'obligation'. 142 Bonhoeffer concurs with Kierkegaard that one becomes an individual through responding to the call of God, and that in obeying to sacrifice his son, Abraham 'becomes an individual, a lonely and solitary figure'. 143 But while Kierkegaard overemphasizes Abraham's solitary individualism, Bonhoeffer's view is closer to that of Gottlieb and Stump. He acknowledges that in receiving Isaac back, Abraham is called to live amid society, among his kith and kin. Because, for Bonhoeffer, the obedience to the divine call is manifested in loving others. 144 In other words, while the disciple acknowledges that the world 'has been built on an illusion', 145 they nevertheless manifest their faith not through solitary withdrawal, as Kierkegaard implies, but through communal activity.
Bonhoeffer in a way personified his call for discipleship wherein a Christian's moral life must be won through resistance and suffering, making one's 'whole life an answer to the question and call of God'; 146 not the Kierkegaardian solitary kind of suffering, but rather concerning himself with 'how the coming generation is to live'. 147 For Bonhoeffer realized amidst the horrors of the Second World War that the Christian is called to sympathy and action not primarily for their own sake, 'but by the suffering of his brethren, for whose sake Christ suffered'. 148 O'Donovan writes that 'to suffer well one must suffer for things worth suffering for'. 149 Martyrs are not mere dissidents whose sole glory was to refuse the order subjugating them. Rather, martyrdom, as described by O'Donovan, is a 'witness pointing to an alternative offer …, vindicated through saying yes as well as saying no'. 150 Thus, Franz Jägerstätter refused to fight in what he considered an unjust German war and willingly accepted death as a consequence. 151 Likewise, Bonhoeffer was willing to say no to the Nazi tyranny and to lay down his life to end it; and an IO ought to be willing to suffer the consequences (such as resignation) for saying no to practices required of them by their secular authority. For as Bonhoeffer writes, disciples overcome evil with good, even if it means taking up suffering-such as joblessness-upon themselves. In le Carré's words, 'a good man knows when to sacrifice himself, a bad man survives but loses his soul'. 152 Where does the critique leave our reflection on the Christian ethics of an IO-agent relationship? First, I have established that the relative silence of the Bible on the morality espionage shelves an a priori categorical prohibition thereof. Then, the discussion of Kierkegaard's suspension of the ethical reveals both that the universal ethical principle might not be always inviolable, but also that his is a particularly patriarchal reading which invites radical individualism. Considering O'Donovan's theology of the state, it has become evident that for a Christian IO, God, the giver, is hierarchically superior to an earthly authority, a gift. Thus, even if Kierkegaard's teleological suspension holds, a Christian IO cannot suspend the ethical on behalf of their state if they subjugate God in doing so. Further, I have demonstrated that Walzer's secular notion of dirty hands is a poor justification for a Christian, not least due to its absence of ultimate justice, while Bonhoeffer's Christian reading of the dirty hands is helpful primarily in extreme cases. But the IO's daily conduct could be better guided by Bonhoeffer's active, Christ-centred discipleship.
The preceding discussion has, hopefully, indicated several blind alleys and pointed to partial answers to some of the most pressing ethical questions concerning the Christian IO. At the same time, however, it opens many new challenges. In the final part of this article, I shall try to address some of those that I consider to be among the most essential. For a critical reading of Bonhoeffer reveals his necessarily limited standpoint, poignantly discussed particularly by Rachel Muers. Furthermore, O'Donovan's theology of the state opens the possibility of a twofold risk: First, the church could simply replace the state as an authority which will introduce new structural influences and establish different 148

A Hearing Disciple
Writing from a feminist perspective, Muers considers Bonhoeffer's speaking standpoint on discipleship to be partial rather than ultimate. 153 She proposes that when one affirms resurrection as a 'place to stand', 154 one is invited not only to speak, but also to hear. According to Muers, Bonhoeffer, from his standpoint, undervalues the importance of hearing to unmask falsehoods that prolong the existing hierarchies. 155 Primarily in his earlier works, Bonhoeffer indeed proposes to let the slave remain a slave to bear witness to the world's sinful condition. 156 In contrast, a true hearing relationship between an IO and their agent would reveal the person set in, but not defined by, the institutionalized context. Such unmasking requires what Muers calls relational 'hearing knowledge'. 157 Silence where words are needed and deafness when one is called to hear are caused by the twin self-centred follies of self-protection (forgetting oneself) and selfaggrandizement (forgetting others), Muers asserts. These follies can be remedied through 'inward liberation to live a responsible life before God'. 158 Crucially, through such liberation, Muers makes progress towards reconciling the conflict between the church and state, implicit in O'Donovan's theology. For a fool's liberation consists 'not in the replacement of the powerful word to which he is enslaved with another such word, but in the restoration of responsibility …, his ability to listen and to know himself as one who is listened to'. 159 Thus, responsibility does not necessarily place one in conflict between loyalty to a church and loyalty to their state; rather it demarcates one's 'place to stand' before God who hears and sees. A Christian IO cannot be the Muersian fool; neither the deaf, forgetting others, nor the mute, forgetting themselves. On the contrary, a conscientious IO will know themselves as a hearing disciple of the God who Himself hears, and thence act responsibly towards their agent.
A true hearing knowledge of the other person requires the IO to critically challenge their conduct and question the effect of structural influences and authority claims within the hierarchies wherein they operate. These questions and challenges will not be a quasi-objective view from nowhere (pace Nagel), but rather grow from the IO's Christian faith, which itself ought to be subjected to critical evaluation by the IO. It is indeed Christian hermeneutics of critical care ethics which, I shall attempt to argue, offer the IO a framework for such questioning and challenging. therefore shine before the agent, although it may remain hidden for their agency and a wider community. 170 Thus, the Christian IO can make a difference in the life of the agent and their family without becoming a loud whistle-blower; a hidden hearing and critical discipleship can achieve more than grandiose public gestures. The very existence of such a seemingly invisible hearing and critical discipleship, however hidden, 'challenges those worlds that are apparently stable and secure', 171 the norms of behaviour that serve existing authorities, the structural forces that perpetuate otherness.
Let me demonstrate the practicality of the proposed discursive approach through a discussion of Bonhoeffer's reflections on the ethics of lying. Wrestling with the morality of his routine lies and deceptions when spying against the Nazi regime, Bonhoeffer offers us a uniquely practical Christian demonstration of discursive, contextual ethics, particularly relevant for the Christian ethics of the IO-agent relationship.
Bonhoeffer's example of a student lying to his teacher illustrates the practical application of such ethics. When the student lies to conceal his father's drunkenness and protect his family's dignity, he 'inextricably links truth with the social structures and relationships within which telling takes place'. 172 Muers notes that Bonhoeffer's student-teacher example is relatively banal. So are countless contexts within which the IO lies to their acquaintances, or even people they randomly encounter. However, Muers warns, 'the banality of the examples points, paradoxically, to the seriousness with which Bonhoeffer takes the issue into consideration'. 173 Muers and Stoddart both note that Bonhoeffer distinguishes between information and truthfulness, the latter being contextual and relational. Bonhoeffer criticizes normative approaches to truth-telling through questioning the context (the publicness of the private question), and the relationships (the boy's higher loyalty to his family than to his teacher). Pondering the context ('the place at which I stand') and the relationships ('who gives me cause to speak') of his example, Bonhoeffer reaches a striking conclusion: That sometimes it is more truthful to conceal information or to withhold speech. 174 Although the child lies about his father, his lie 'contains more truth, that is to say, is more in accordance with reality, than would have been the case if the child had betrayed his father's weakness in front of the class'. 175 So, Bonhoeffer understands truthfulness as dependent on the context and the particular relation. In Stoddart's reading, truthful speech-or silence-concerns the relational reality made possible by God, not an ideal of truth. 176 Muers takes a position similar to Stoddart's and concludes her commentary on Bonhoeffer's example that 'silence or the withholding of speech is … in certain circumstances more truthful than speech'. 177 'Truthful silence' 178 which contains the 'reality as it exists in God' 179 thus opens new ethical possibilities for the IO, not only in relation to their agents, but in their broader social and personal life. These possibilities can be further directed by a Christian critical reading of Martin Buber's I-Thou relationship. While a pure I-Thou relationship between two people is utterly impossible, I argue that the attempt itself radically transforms the institutional IO-agent relationship into an inter-personal relationship wherein the hearing and critical disciple approaches the other as the imago Dei.
where 'there is no inner discord between private person and official capacity. In both we are disciples of Christ'. 203 The challenge of the supposed dichotomy gives the IO a framework to become a hearing and critical disciple who tirelessly challenges their conduct and examines the structural sources thereof, opening possibilities for reflective, often hidden actions within the institutionally constrained settings. 204 I maintain that this is the essence of the ever-imperfect process of becoming a Christian IO, one who does not pretend to know the mind of God, but one whose conduct is directed by their critical reflection on the demands of the derived authority of their state and on the ethical principles of their Christian faith. This hidden disciple shines a light on a possible route towards Christian ethics of spying, indicating how a Christian IO can indeed draw on their faith to make its ethical principles compatible with the demands of their secular vocation. Perhaps, eventually, they will even be able to prove Alex Leamas wrong. For Christians are called to outshine all the vain fools, traitors, pansies, sadists and drunkards; they are called to be hearing and critical disciples, even when they remain hidden.

Declaration of Conflicting Interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author received no financial support for the research, authorship, and/or publication of this article.