AustLII Home | Databases | WorldLII | Search | Feedback

University of New South Wales Faculty of Law Research Series

Faculty of Law, UNSW
You are here:  AustLII >> Databases >> University of New South Wales Faculty of Law Research Series >> 2008 >> [2008] UNSWLRS 24

Database Search | Name Search | Recent Articles | Noteup | LawCite | Author Info | Download | Help

Dixon, David --- "Interrogating Terrorist Suspects: Criminal Justice and Control Process in Three Australian Cases" [2008] UNSWLRS 24

Last Updated: 5 September 2008


David Dixon[*]


This paper examines three cases in which people suspected of terrorist activity were questioned by Australian police and security officers. They are located in the context of fundamental shifts in criminal justice processes and principles. Debates about interrogating terrorist suspects are dominated by concerns about torture. It is argued here that such concerns need to be supplemented by paying more attention to everyday questioning in the increasing mass of ‘ordinary’ terrorist cases.


The questioning of suspects takes place within a broader context of institutional practices, priorities and values. This paper will indicate that this context is undergoing fundamental change in most Western democracies. What had been generally accepted as fundamental principles of criminal justice are being compromised, devalued and even abandoned in a shift towards what is better understood as a control process with very different values and priorities. These general trends will be illustrated by reference to examples of how some people thought to be connected to terrorist activities have been interviewed by Australian authorities, and the judicial consideration of the products of such questioning. The roles of prosecutors and of government will also be considered.
Exponents of investigative interviewing need to be aware of these contextual changes. It will be argued that the current focus on interrogation of major suspects needs to be widened to take account of and make provision for the more common questioning of those on the margins.


In the twentieth century, a paradigm of criminal justice matured in Anglophone jurisdictions in which questioning of suspects had a specific place (Dixon, 2008). Suspects should be interviewed in a closely regulated period between being arrested and being charged with an offence. Such questioning should normally take place at a police station where police supervisors were responsible for ensuring access to various rights, notably of access to legal advice. The length of investigative detention was restricted by time limits (eg in England and Wales under the Police and Criminal Evidence Act 1984, a maximum of 96 hours, although regulatory hurdles ensured that most suspects were charged or released within six hours.) After being charged, a suspect could not be interviewed about that offence. These were specific expressions of a criminal justice paradigm with deep roots in liberal democratic conceptions of relations between state and citizen.
For those of us who professionally grew up seeing these arrangements as normal, it is important to appreciate their relatively recent origins. Until the mid-nineteenth century, magistrates, not police, directed criminal investigations. Until the mid-twentieth century, the propriety of police questioning suspects between arrest and charge was unclear. For a long period, such questioning attracted judicial criticism. Until legislative interventions in many jurisdictions in the late twentieth century, the legality of investigative detention was contested (Dixon, 1997).
This confined and contested conception of interviewing’s place was both product and part of a broader criminal justice paradigm. Key characteristics of this approach include:


In the last decade, a new paradigm has taken shape which I will call a ‘control process’, emphasising that its concerns are neither exclusively about what is ‘criminal’ nor ‘justice’ in the senses commonly understood in the paradigm of criminal justice. The key characteristics of this paradigm are in stark contrast to those of criminal justice:

The central concepts of liberal democratic criminal justice are devalued in the new control process. The key concern is now the minimization of risk and the security of the group. The individual is no longer the centre of attention, and so there is less emphasis on the individual’s rights and the need to prove the individual’s guilt beyond reasonable doubt through a system of due process. Flexibility of process replaces certainty of rules and procedures as a virtue. ‘Pre-emption’, compliance and efficiency are more important than individual punishment or due process: preventive detention for potential sex-offenders and others, anti-social behaviour orders, behaviour management contracts, non-association and space restriction order, and the use of bail conditions as a proactive crime control measure are just the more prominent examples. There is less interest in understanding crime’s causation than in accepting crime as normal, a choice to be controlled and insured against, in which “attempts to cure or punish appear less logical than do moves to manage crime and minimize its costs” (Zedner, 2005a: 284). The state’s responsibility for crime control is “contracted out to private providers wielding state franchises, delegated to individuals and communities, or completely over taken by the growing private security industry” (Zedner, 2005a: 284). Policing intervenes proactively, preventing and pre-empting problems rather than retrospectively solving them.
Simply to say that all this goes against basic principle is rather like complaining that a game of chess isn’t being played according to the rules of draughts. The game has changed, allowing those in government to dismiss the standard civil libertarian response to new police powers as anachronistic and irrelevant. The contrast between criminal justice and control process is exaggerated here in order to clarify the difference. It has become trite to respond to by pointing out that criminal justice has always included substantial preventive elements. The modern English policing tradition has a strong preventive commitment in the Peelite tradition. Preventive detention has been possible for those refused bail, the mentally ill, habitual offenders, those suffering from certain infectious diseases, illegal migrants, refugees, sex offenders and others. Similarly, deterrence is by nature forward-looking: but its activation depends on a crime having been committed. This illustrates the essential point that while criminal justice includes preventive and deterrent elements, these do not change the ideological core commitment to a reactive, individualistic process.
While these developments have been underway for some time, they accelerated quickly after 9/11. Parliaments are now in a constant cycle of extending anti-terrorism legislation in ways that routinely deviate from liberal democratic principles in the name of necessity. It must be acknowledged that the shift from criminal justice to control process did not happen suddenly on 9/11 any more than Islamic terrorism arrived on the world stage on that day. Rather, the response to terrorism must be seen as hastening changes which were already under way, notably in the other ‘war’ of our times, that on illegal drugs. The law enforcement response to illegal drugs (and the organized crime associated with their distribution) has entailed a long series of compromises and exceptions to basic principle, reducing and shifting the burden of proof reduced from prosecution to defence and deploying incapacitative civil and criminal penalties. The contrast between criminal justice and control process is exaggerated here in order to clarify the difference. It has become trite to respond to by pointing out that criminal justice has always included substantial preventive elements. The modern English policing tradition has a strong preventive commitment in the Peelite tradition. Preventive detention has been possible for those refused bail, the mentally ill, habitual offenders, those suffering from certain infectious diseases, illegal migrants, refugees, sex offenders and others. Similarly, deterrence is by nature forward-looking: but its activation depends on a crime having been committed. This illustrates the essential point that while criminal justice includes preventive and deterrent elements, these do not change the ideological core commitment to a reactive, individualistic process. In addition, much of the groundwork for the new paradigm’s emphasis on preventative intervention and detention was laid in responses to perceived threats from sexual predators and other dangerous risks.
However, the fundamental difference between criminal justice and control process is real and marked, and there is a shift from the former to the latter. These differences can be illustrated through analysis of the different role of interrogation in the two paradigms. In the criminal justice paradigm, police question a suspect between arrest and charge in order to obtain evidence which may be admissible subsequently in court about specific offences allegedly committed by the suspect. In the control process paradigm, the purpose of interrogation may be much broader: the focus is not the suspect’s past actions but what he or she knows about future actions by others. In response to 9/11, “the interest of investigators has shifted from obtaining viable evidence for prosecution to obtaining credible information for preventing future acts of terrorism” (Strauss, 2003: 206). In this context, what counts as success may be much less than obtaining a confession:

Interrogators find tiny bits of the truth, fragments of information, slivers of data. We enter a vast desert, hundreds of miles across, in which a few thousand puzzle pieces have been scattered. We spend weeks on a single prisoner, to extract only a single piece – if that. We collect, and then we pass the pieces on, hoping that someone above us can assemble them (Mackey, 2004: xxv).

Controls on interrogation which are designed around the potential sanction of evidence being excluded as inadmissible are of limited value when producing admissible evidence is not the main objective. Criminal justice and control process are paradigms. They are neither clearly distinct nor sequential, with one simply replacing the other. Rather it is exactly the discordant overlap which will emerge as a significant problem in some of the specific cases to be discussed below.


Perhaps to the disappointment of some and the relief of others, this chapter is not primarily concerned with torture. While only a few years ago, torture was widely (although wrongly) regarded as a historical relic, its modern use has spawned a vast literature. The disclosure of (some of) what happened in Abu Ghraib and the global prison network of the USA’s affiliates has been an extraordinary achievement by investigative journalists and human rights organizations. It provided the spur for many in the West to question and challenge the post 9/11 policies and actions of the USA and its allies. However, this concentration of attention appears problematic from two perspectives. First, some academic discussion of torture seems to share some elements of the dark, obsessional, almost pornogaphic, interest displayed in some contemporary films and TV series. Secondly, and more relevant to this chapter’s concerns, the focus on the negative means that we have been concerned with what must not be done rather than what can and should be done in questioning suspects. My particular interest is in what should be permissible in the routine, everyday questioning of those thought to be on the edges (or connected to the edges) of activities under investigation. For every high-profile suspect, there are hundreds of people who the authorities detain and question in the process of accumulating the vast banks of information on which counter-terrorism depends.
The issue of torture – its morality, legality and effectiveness – has dominated debates about interrogation related to terrorism. This has been inevitable: the fact that at the time of writing, there continues to be serious talk about whether simulated drowning should be regarded as torture is an indication of how far the compass has shifted in recent years.[2] While it may have been necessary, the debate on torture and what should not be done in interrogation dragged attention away from the practical question of what should be done. The answer to this question depends in part on the moral dilemmas founding the torture debate. But it also depends on what the interrogation is for. If the interrogation is part of a process which will or may lead to criminal law prosecution, the answer will be very different than if the interrogation is part of an intelligence-gathering exercise. Of course, in reality this distinction is very difficult to maintain. As we will see in the discussion below of Australian cases, the overlap between criminal justice and control process has proved to be very problematic
A good example of the problems is provided by a contribution to a conference on ‘Law and liberty in the war on terror’ by Neil James, who has operational, supervisory and training experience in interrogation. His paper provides a familiar critique of torture, focusing on instrumental issues of effectiveness and practicality. He seeks to distinguish torture from legitimate interrogation. Quoting from the Australian Defence Force’s Interrogators’ Handbook, (of which he was the original author), James states:

Among professional interrogators in countries abiding by the rule of law the common working definition of interrogation is ‘the systematic extraction of information from an individual, either willing or unwilling, by the use of psychological attack only’. Thus, interrogation is essentially an intellectual process not a physical one. No physical or mental pain or severe suffering is involved. The subject is convinced to cooperate by reasoning and by overcoming their will to resist (2007: 161).

Some may argue that the line between torture and interrogation is not so defined as James suggest. However, this is not my concern here. Rather, my focus is on the acceptability of this form of interrogation in a criminal justice model. Any evidence lawyer’s ears would have pricked up at this account of interrogation: unwilling extraction of information, psychological attack, and overcoming of the will to resist provide ready ammunition for an argument that a confession was involuntary, so failing to get past the first requirement of evidential admissibility.
The problem is clarified when James goes further, justifying “measures or conditions of discomfort or annoyance designed to encourage cooperation are not unreasonable ... With obvious safeguards such conditioning may, for example, legally include the strictly controlled and temporary use of measures such as isolation, sensory deprivation or sleep deprivation” (2007: 162). While it may be the case that such treatment does not amount to criminal offences, it certainly would render any confession thereby produced inadmissible under the current law of evidence in many countries.
James is concerned with more serious and difficult cases. My primary interest is rather different. An interrogation to discover a ‘ticking bomb’ will happen very rarely, if ever, despite the enormous attention this ‘scenario’ has attracted. Stepping down the scale of seriousness, few interrogators will be called on to deal with the kind of suspect for whom even techniques such as sleep deprivation could be considered as appropriate. Just as has often been the case in policing more generally, there is an unfortunate concentration on everything except the everyday and ‘normal’ practice. The vast majority of terrorism related interrogations will continue to involve low-level people on the fringes. For them, length of detention will be the critical variable.
A now standard part of critiques of torture and related techniques is that other methods are more effective, even in questioning those allegedly involved in terrorism. In particular, it is argued that interviewers with appropriate language skills, cultural knowledge and training can build rapport with suspects and thereby produce results (Gelles et al., 2006; Pearse, 2006). The problem is that such rapport is most unlikely to be established in the short periods allowed under standard criminal justice regimes. As Gelles et al. suggest, “A rapport-building (or relationship-based) approach will yield the best results in an interview/interrogation that occurs over days/weeks/months” (2006: 31). James confirms that “the time available” is crucial (2007: 161). In these approaches, it is taken for granted that the subject of interrogation must be detained for an extended period in order for interrogation to be successful. Yet this runs counter to one of the basic principles of criminal justice regimes such as PACE which were constructed on an understanding that extended detention in itself could make confessions unreliable because people would say anything (even at long term cost) to win a short term reprieve from investigative detention.
If interrogation is intended to produce confessions and admissions which are acceptable to a criminal justice paradigm, lengthy pre-charge detention is unacceptable because it undermines the voluntariness which is a precondition of evidential admissibility. If interrogation is primarily intended to produce information and actionable intelligence rather than admissible evidence, then the concerns of criminal justice will not be paramount. In addition, problems for the authorities arise when the lengthy interrogation for intelligence is over. Are the subjects of interrogation to be detained for ever (or until the ‘war on terror’ is over, if there is a difference)? Are they simply to be released and placed under administrative control orders which limit their movements and contacts? Or should an attempt be made to create an ersatz criminal justice, dressing up a militaristic control process with some trappings of legality? These are, of course, the questions which the USA has been grappling with over the detainees at Guantanamo Bay.


Much of the extensive post 9/11 literature on ‘liberty versus security’ and related changes in criminal justice has been general and abstract. This section seeks to provide a closer focus through brief analysis of three Australian cases involving alleged terrorist offences in which controversial interrogations have played crucial roles.

i. Jack Thomas

In 2003, Jack Thomas (an Australian citizen) was arrested and detained for five months in Pakistan. During this period, he was ‘interrogated numerous times by Pakistani, American and Australian officials, often whilst blindfolded, hooded and shackled’ (Lynch, 2006: 313). According to his account, which was accepted as truthful by an Australian court, he was assaulted, threatened with torture and with the rape of his wife, and offered inducements of favourable treatment in return for cooperation.
In a final session in Pakistan, Thomas was interviewed by Australian Federal Police (AFP) agents. This took place in the same room as previous interviews with security and police officials, and the AFP interviewers had both attended some of the previous sessions. The purpose of this interview was ‘to gather evidence in a form and by a process that would be admissible in an Australian court’ (Lynch, 2006: 314). It was, in other words, an attempt to bridge the gap between control process and criminal justice, between a series of interviews conducted in a security facility in a foreign country and the proceedings of an Australian court. In this interview, the AFP agents attempted to meet criminal process requirements, explaining the right to silence and emphasising that participation in the interview was voluntary. The relevant Australian law requires a suspect to be provided with access to legal advice: the AFP tried to arrange this, but Pakistani Inter-Service Intelligence officials refused (Lynch, 2006: 315). The Victorian Court of Appeal commented that the interview was ‘conducted in what can be reasonably described as a conventional fashion’ (R v Thomas [2006] VSCA 165 at para 51). Statements made by Thomas in this interview were subsequently presented as part of the case against him when (more than year later) he was arrested and charged in Australia with offences of receiving funds from and providing support to a terrorist organisation.
The crucial issue for the Australian courts was whether the final interview could be distinguished from what had preceded it so that the evidence produced could meet criminal justice standards – notably, the base requirement that a confession or admission must be made voluntarily. In a voir dire, the trial judge ruled that the admissions had been made voluntarily: Thomas was convicted and sentenced to five years’ imprisonment. Narrow legalism, sympathy to police and prosecutors, and a narrow view of reality have often been characteristics of Australian criminal trial judges. The trial judge concluded that Thomas had a choice to answer questions or not, and exercised that choice. The pressure to answer came not from the interviewers ‘either expressly or implicitly’, but from Thomas’s own assessment of the ‘risk of indefinite detention in Pakistan or of removal to the United States or Cuba’ (DPP v Thomas [2006] VSC 243 at para. 42). The judge found that this interview could be distinguished from the earlier interviews and the inducements offered in them: “There was a clear bifurcation in purpose, function and form between the ASIO interviews and the AFP interview. Mr Thomas fully understood it” (ibid. at para. 50).
The Victorian Court of Appeal rejected this artificiality and restated the accepted common law principle that an earlier inducement (which may be a threat or promise – in lawyers’ terminology, “fear of prejudice or hope of advantage”) offered by a person in authority (such as the security interviewers) can continue to affect the suspect’s voluntariness. The Court of Appeal’s conclusion was the same as any commonsense understanding of the situation – the final interview could not be divorced from those preceding it – “same place, same AFP personnel, same topics” (2006 VSCA 165 at para 84).

Obviously, the fact and circumstances of his detention, the various inducements held out and threats made to him, the prospect that he would remain detained indefinitely, can be seen to have operated upon the mind of the applicant when he decided to participate in the (final AFP) interview. Whist nothing occurred in the interview itself that could be seen to overbear the will of the applicant, there can be little doubt that he was, at that time, subject to externally-imposed pressure of a kind calculated to overbear his will and thereby restrict, in a practical sense, his available choices and the manner of their exercise. (ibid, para 92)

The AFP’s attempt to lay a patina of criminal justice over a structure of control process was rejected. Notably, the failure to provide access to legal advice could not be excused simply because the refusal was by the local authorities, not the Australian investigators. Their refusal meant that an interview conforming with Australian law could not be conducted in Pakistan (ibid., para 111).
For the future in a case like Thomas’s, the alternatives for the authorities are clear. Either there must be a much greater distinction in time, place and personnel between interrogations carried out for different purposes, or prosecutions must be based on evidence other than confessions or admissions, or an alternative to criminal justice must be deployed. As will be shown below, the Australian authorities are exploring all options. From this perspective, the trial judge’s convolutions in Thomas’s case are understandable: however flawed, they represented an attempt to maintain the relevance of a criminal justice paradigm.

ii. Mohammed Haneef

Dr Mohammed Haneef, an Indian doctor practising in Queensland, was detained following the London and Glasgow car-bomb incidents in June 2007. These marked a significant shift in concern about terrorism. The previously accepted wisdom was that the threat of terrorism was associated with alienation and anomie in ethnic minorities: yet those allegedly associated with the London and Glasgow incidents were not unemployed inner-city youth, but doctors. The shift from risk to precaution (Zedner, 2005b; 2007a) is illustrated by this. If members of a professional elite were engaged in terrorism, the attempt to identify auditable risk factors seems doomed: instead, the authorities respond as if risk is all around and that precautionary action is necessary.
Dr Haneef’s alleged connection to the British incidents was that he was second cousin to a man who died from burns suffered in the Glasgow incident and a telephone SIM card purchased by Haneef had been found with the alleged bombers’ possessions. Suspicion was increased by his attempt to board a flight from Australia to India. With at least tacit encouragement from a government facing a general election which had previously exploited security scares for political advantage, sections of the media treated Haneef as a prize capture. It was alleged, for example, that he was planning to blow up an apartment block on Queensland’s Gold Coast, a tourist area.
The ‘case’ against Haneef then spectacularly fell apart: his flight to India was to see his newborn child; the apartment bomb story was based on no more than a photograph of Haneef and his wife on a Gold Coast beach; the SIM card was found not, as initially reported, in the vehicle driven into Glasgow airport, but in Liverpool. (Haneef had passed it on when he left England so that the remaining credit would not be wasted.) As in Thomas’s case, we see the courts applying the criminal justice principles; but these rub hard against exigencies of the new control process. When a magistrate took the brave and unusual step of ordering Haneef’s release on bail, the Australian Government intervened by withdrawing his immigration visa and converting investigative detention into pre-deportation detention. The ‘orthodox process’ of law was overtaken by administrative discretion (Lynch, 2007: 228). This was done in a way (citing the national interest) which sought (ultimately unsuccessfully) to prevent judicial scrutiny of the decision. Exposure of the weakness of the prosecution case and widespread public criticism of the police followed. While the Australian government continued to mutter darkly about Haneef’s connections with terrorism, the Commonwealth Director of Public Prosecutions eventually dropped the charges against Haneef. He was released, and left for India to see his daughter for the first time (Lynch, 2007: 226-7). In a final humiliation for the Government, the Federal Court (in a decision which was scathing about the Immigration Minister’s behaviour) ruled that the visa cancellation was unlawful. At the time of writing, Dr Haneef was considering returning to resume his medial career in Australia.
Interrogations conducted by Australia Federal Police agents with Dr Haneef played a vital part in this reversal. Haneef was interviewed over several days. ‘He was the first person to be detained under new anti-terrorism powers which enable Australian police to hold a suspect without charge for an extended period of time during which questioning up to a maximum of 24 hours may occur’ (Lynch, 2007: 225). The interviews were lengthy but were carried out in accordance with criminal justice values and standards. The relevant legislation follows the standard practice of specifying a maximum period for active investigation to which is added ‘time outs’ in which the detention clock is stopped while other investigations take place, and the suspects is allowed to rest and eat. The indeterminate length of detention under a ‘time out’ system has been criticised. However, the problem of lengthy detention had rarely been so clearly demonstrated as in Haneef’s case: he was detained for 12 days before being charged with providing ‘support or resources’ to organisation involved in terrorist activity (Lynch, 2007: 225-6). This was a much longer period than those responsible for drafting the legislation had thought would occur (Lynch, 2007: 228).
The interviewers were polite and respectful, if not very well prepared. We know this because Dr Haneef’s barrister, Stephen Keim, responded to the Government, police and media misrepresentation of his client by releasing the transcript of the first of the recorded interviews to the press.[3] Apart from demonstrating Haneef’s apparently full cooperation, the transcripts revealed that aspects of the prosecutions case against Haneef were unfounded: he had, for example, a good reason for his supposedly suspicious attempted departure from Australia. It emerged that some of his actions were apparently inconsistent with guilt – for example, far from fleeing from investigators, he had made several attempts to contact the UK police (Lynch, 2007: 227). When the Immigration Minister proceeded to selectively release passages from the second interview in an attempt to justify his stance, while claiming that he was unable to release the full record on the basis that it might prejudice ongoing police investigations (Lynch 2007: 227), Haneef’s lawyers again released the full transcript. Its anodyne contents deflated the government’s attempts to justify Haneef’s treatment. As in other contexts, a comprehensive record of interview (ie not just a recording of a rehashed confession) can provide suspect as well as police with valuable resources (Dixon, 2007).

iii. Izhar Ul-Haque

In early 2003, Izhar Ul-Haque, an Australian citizen, spent three weeks at a camp in Pakistan organised by Lashkar-e-Taiba, which would subsequently be prescribed under Australian law as a terrorist organization. Six months after his return to Australia, Ul-Haque was interrogated by Australian Security Intelligence Organisation (ASIO) officers and subsequently was charged with an offence of training with a terrorist organisation. The circumstances of Ul-Haque’s interrogation emerged at his trial. In his ruling on a voir dire concerning the results of this interrogation, the trial judge was scathing about ASIO’s conduct: not only refusing to admit the interview records into evidence, he concluded that the investigating officers had committed significant criminal and civil offences in their treatment of Ul-Haque. His judgement provides a vivid picture of the contrast between criminal justice and control process.
As noted above, in the criminal justice paradigm, interrogation is designed to produce information and, if appropriate, admissible evidence about an offence committed by an individual. In Ul-Haque’s case, the function of the alleged offence was as a lever with which to put pressure on Ul-Haque to collect and provide information about a person suspected of more serious offences: the trial judge commented “It seems almost certain that the action taken against the accused by the authorities was instigated by his being connected with one Fadeen Lodhi”. (Lodhi was later convicted of terrorism offences.) Ul-Haque had a family connection with Lodhi: communication between the two following Ul-Haque’s return from Pakistan was what “excited the authorities and instigated the actions of the authorities that led to interviews here in question and his ultimately being charged” (R v Ul-Haque [2007] NSWSC 1251 at para 13). According to Ul-Haque, the officers told him that they did not wish to speak to him about his training in Pakistan: “They said, ‘No, we know about that. We’re not concerned with that’” (ibid., para. 21). An officer told the court that his colleagues questioned Ul-Haque because “we had an investigation underway and we had information which indicated (he) had information which could assist us in that investigation” (ibid., para 30). Ul-Haque claimed that an officer told him “we see you as a helper, as an informant and as a witness ... I don’t think you have done anything wrong” (ibid., para. 78).
In November 2003, ASIO officers intercepted Ul-Haque at a railway station on his way home from university, where he was studying medicine. He was not formally arrested. The officers’ and Ul-Haque’s accounts of the initial confrontation differed somewhat, but the differences were not significant. It is clear that what occurred was a very familiar example of an order lightly camouflaged as a request: as Ul-Haque commented, “the thought of choice never really occurred because I was under extreme pressure and stress” (ibid para 21. On the relationship of ‘consent’ to the exercise of police powers, see Dixon, 1997: ch.4.) As the judge concluded, “Although it is described as a request, ... his being told to accompany them to a nearby park was an instruction and was intended to be taken as such” (op. cit., para 27).
Sitting between two officers in the back of a car, he was taken to a public park and questioned about his knowledge of Lodhi. An ASIO officer told him that “he was in substantial trouble and that his full cooperation with ASIO ... would be required” (ibid., para 25). On the officers’ own account, this included “robust discussion and considerable prompting from the interviewing officers” (ibid., para 23). The interrogation was not contemporaneously recorded, and again the two sides’ accounts differed in tone. They agreed that he was given a choice of cooperation or facing the consequences of failing to do so: in the officers’ account, this meant continued investigation, while in Ul-Haque’s it meant action against him unless he collected and provided information about Lodhi (ibid., paras 20-1). To a young Muslim man aware of contemporary developments in the ‘war on terror’, the possible implications of the threat were very serious:

...when he said ... “we can do this the easy way or the hard way”, I believed that unless I kept ... answering their questions that they will use the hard way (which) to me meant ... for example that I could either be deported , I could be arrested, I could be taken to a secret location for interrogation ... and my family ... would be taken into custody (ibid., para 71)

The judge accepted Ul-Haque’s account of his fears: “the accused was given to understand that it was in his interests to co-operate, and there was an implicit threat that if he did not do so some adverse consequences might follow” (ibid., para 67).
Ul-Haque was then interviewed further in the car on the way to his parents’ house, which was being searched by some 25 ASIO and police officers, and again on the way back to the railway station where Ul-Haque’s car had been left. He then returned, accompanied by an agent, to his parents’ house, where he was interviewed for a further 2 hours and 46 minutes, ending at 3.45am. Later that day, and then again five days later, he was subjected to further questioning, on these occasions at a police station by Federal Police agents. Two weeks later (after numerous telephone contacts), the agents returned to Ul-Haque’s house. The AFP’s account was that they requested Ul-Haque to “further assist the Federal Police by undertaking covert enquiries or acting as a witness”. Ul-Haque’s account was that they threatened to prevent him from continuing to study medicine and “make life difficult” for his family (ibid., para 110)
Many people will not be surprised at how ASIO treated Ul-Haque: this is how an intelligence agency may be expected to behave. The problem arose when an attempt was made to shift from control process to criminal justice by charging him. (The real motivation appears to have been to punish him for non-cooperation with ASIO rather than for his brief involvement with Lashkar-e-Taiba.) What might have been conventional security agency practice came under intense, critical scrutiny in a criminal court. The judge was merciless in his critical dissection of ASIO practice from a criminal justice perspective. The ‘request’ to go to the park was an unlawful arrest. Ul-Haque should have been informed of his rights, taken to a police station, detained according to regulations and had his interview recorded. The failure to specify his alleged offence and to tell him instead that “he knew what he had done wrong” was “reminiscent of Kafka” (ibid., para 31), while “to conduct an extensive interview with the accused, keeping him incommunicado, under colour of the (search) warrant, was a gross breach of the powers given to the officers under the warrant” (ibid., para 44). The interviewers’ ‘prompting’ of Ul-Haque was condemned as bad interviewing practice: according to the judge, “the ASIO officers told him when they thought he was not telling the truth and told him, or suggested, what the truth was” which was a method not used by police because it is “calculated to obtain what the suspect believes the interrogator wants to hear ... It is inevitable that the truth of what is said is likely – if not certainly – to be adversely affected” (paras. 46, 102).
According to the judge, the ASIO officers’ conduct did not merely affect the admissibility of evidence from the interviews with Ul-Haque: it included a string of criminal offences – false imprisonment, kidnapping, and assault which were intentionally committed (ibid., paras. 59,61). These assessments were firmly located in a context of constitutional principle - the requirement of legal authorization for the executive to interfere with an individual’s liberty (ibid., para 60). The conduct of the ASIO officers was

grossly improper and constituted an unjustified and unlawful interference with the personal liberty of the accused (and) was a gross interference by the agents of the state with the accused’s legal rights as a citizen, rights which he has whether he be suspected of criminal conduct or not and whether he is a Muslim or not. Furthermore, the conduct was deliberately engaged in for the purpose of overbearing the accused in the hope that he would co-operate’ (ibid., pars. 62, 95).

It was made clear that the Organisation was responsible for the individual officers’ misconduct was made clear: “The impropriety ... was grave. There is no suggestion that the officers acted contrary to ASIO protocols and good reason for thinking that they did not” (ibid., para 105)
As in Haneef’s case, electronic recording assisted the defence. While the ASIO interrogations in the park, house and car were not contemporaneously recorded, that by the AFP at the police station was audio-visually recorded, allowing the judge to make an assessment of Ul-Haque’s demeanour and manner of answering questions: “The overwhelming impression that I got from watching the interview is that the accused was cowed” (ibid., para 73). This contributed to his assessment that the negative answers which Ul-Haque gave to the routine questions about threats, promises or inducements at the interview’s close were made because “he just decided he was going to be compliant and wanted the interview to end” (ibid., para 75). He regarded the questions designed to expose any threats, promises or inducements as “an empty formality” (ibid., para 92). The judge recognised the strangeness of the legal fiction that any suspect whose will has been overwhelmed by threats, promises or inducements during an interview will be able to slough off their effects and to answer the concluding questions genuinely.

iv. Three cases

These are very different cases: Thomas had, on his own account, been significantly involved with terrorist organisations and was a legitimate subject of investigation; Haneef’s distant family connection appears to have been enough to excite Australian politicians making political capital out of a ’terrorist’ drama and security agencies eager to show their worth; in the case of Ul-Haque, it appears that the investigation was conducted primarily to get Ul-Haque’s cooperation in collecting and providing information about others and, when this cooperation was refused, punishing by prosecution.. What connects them for present purposes is their illustration of the uncomfortable relationship between criminal justice and control process paradigms and the central, controversial role played by interrogation.


The criminal justice paradigm is not dead when there are responsible, independent professionals who take its principles and values seriously. Thomas’s, Ul-Haque’s and Haneef’s lawyers provide an excellent example of the long tradition of lawyers who, refusing to be intimidated, rely on the basic principles of the rule of law and natural justice. Haneef’s lawyers deserve particular comment: refusing to be swept along in the political and media hysteria about their client, they insisted on due process and took the fight to the authorities by releasing the interview records. The legal principles which provide the motivation for lawyers such as these are not complicated: a fair hearing according to law before an independent judge applying, in a criminal case, the principles of criminal justice outlined above.
As this statement indicates, judicial officers play a pivotal role. The justices of the Victorian Court of Appeal who heard Thomas’s appeal, the Queensland magistrate who granted Haneef bail and the New South Wales judge in Ul-Haque’s case can proudly take their places as members of a group identified by Dyzenhaus and Thwaites : “(t)here seems to be something like a judicial ‘coalition of the willing’ forming – judges prepared to uphold the rule of law in the face of executive claims about national security” (2007: 10). There is certainly evidence in the judgements quoted above of judges who are uneasy about current trends and who are prepared to challenge governments over aspects of them.
Some governments have expressed concern about the limitation or exclusion of legality from responses to terrorism. This frequently finds expression in calls for the rhetoric of war to be abandoned, and for countering terrorism to be seen as a task for law enforcement (Wilson, ed. 2005). What ‘law enforcement’ means in this context is unsettled, with police and security over-lapping, cooperating, and occasionally squabbling.
However, law has familiar limits as a restraint on power. Court cases are expensive and slow. Lawyers acting for the prosecution too often resemble government agents rather than independent professionals. Too many of the heroic judicial affirmations of freedom have come from judges in dissenting judgements. Even if such judges carry the majority, courts can rarely have the final say: governments and parliaments can respond to judicial decisions which they do not like by undermining them by legislative action or by turning to administrative rather than judicial means of control. They are particularly able to do so in a jurisdiction like Australia which has very limited constitutional protections of individual rights. For example, Thomas may not have been convicted, but he was made subject to an administrative control order (which the High Court found to be legal in Thomas v Mowbray [2007] HCA 33). Ironically, reliance on law may serve to push state action outside the reach of legal principle, as what was previously unacceptable is legalised. As Lynch concludes pessimistically, legal systems around the world are “undergoing a rapid re-adjustment in order to respond to the post 9/11 world” (2007: 231-2)
Courts are increasingly attacked if they give any indication of being “soft on terror” (Lynch, 2006). The judgements in the cases considered here attracted considerable criticism from popular media, although some of the quality press supported them. Of more concern is the official response: to date, there has been no recognition by the Australian Government of the judicial critiques, far less any prosecution or disciplining of officers for demonstrated misconduct. Security authorities do not see acknowledgement of legality as an appropriate response to these court rulings. Finding ways around them seems more appealing: the Australian Federal Police Commissioner declared that “Both in the UK and Australia we are testing the courts. We make no apologies for that, .. it’s part of the work police do ... and will help prevent a (terrorist) attack here”.[4] Action has been threatened against lawyers who challenge the authorities. For example, Stephen Keim, Dr Haneef’s barrister. had to defend himself against allegations that he had broken professional conduct rules in the way he released interview tapes to the media. The official response to the judge’s criticisms in the Ul-Haque case was not to take action against the police and ASIO officers, but to launch an official complaint against the judge.


The vulnerability of the criminal justice paradigm to executive action and legislative change leads to a pessimistic conclusion. It is perhaps appropriate to end by referring to the cases of the two Australians who were held at Guantanamo Bay, Mamdouh Habib and David Hicks (Sales, 2007). Habib was released without charge, apparently because US authorities did not welcome the prospect of more public scrutiny of what happened to him in Egyptian jails during a lengthy stop-off on his rendered passage to Cuba. At the time of writing, Hicks is the only sometime detainee of Guantanamo Bay who has been ‘convicted’. In a mere façade of legality, confessions produced by years of interrogation led to a plea of guilty to crimes which did not exist, in a court which was not a court, in a place which the US Government had tried to isolate from international law. This guilty plea meant that neither his military ‘commission’ nor, a fortiori a real court, examined the means by which the confessions underlying it were obtained. At the time of writing, Hicks has just been released from the Australian jail where he spent the last few months of detention, apparently broken, not by torture, but by prolonged detention often in isolation. Yet even in this case, the strength of legal principle became evident in the unlikely figure of Hicks’s American army lawyer, Major Michael Mori, who, along with a team of other lawyers (Stafford Smith, 2007) fought for his client skilfully and at very considerable personal cost. Both Habib and Hicks are now subject to indefinite security surveillance and administrative control orders.

I am grateful to my colleagues Professor Jill Hunter and Dr Andrew Lynch for their advice and assistance. My account of Thomas’s case draws heavily on Dr Lynch’s published research.


Chesney, R.M. & Goldsmith, J.L. (2008, forthcoming) ‘Terrorism and the convergence of criminal and military detention models’ Stanford Law Review 60: 1-59.

Dershowitz, A. (2006) Preemption (New York: WW Norton)

Dixon, D (1997) Law in Policing: Legal Regulation and Police Practices (Oxford: Clarendon)

--- (2007) Interrogating Images: Audio-visually Recorded Police Questioning of Suspects (with G.Travis) (Sydney: Institute of Criminology)

--- (2008, forthcoming) ‘Authorise and regulate: a comparative perspective on the rise and fall of a regulatory strategy’, in E.Cape & R.Young, eds. Regulating Policing (Oxford: Hart)

Dyzenhause, D & Thwaites, R (2007) ‘Legality and emergency – the judiciary in a time of terror’, in Lynch et al. eds 2007: 9-27

Garland, D. (2001) The Culture of Control (Oxford: Oxford University Press)

Gelles, MG, McFadden, R, Borum, R & Vossekuil, B (2006) ‘Al-Qaeda-related subjects: a law enforcement perspective’, in Williamson, ed., 2006: 23-41.

James, N (2007) ‘Torture: what is it, will it work and can be justified?’ in Lynch et al eds. 2007: 155-64

Lynch, A (2006) ‘Maximising the drama: “Jihad Jack”, the Court of Appeal and the Australian media’ Adelaide Law Review 27: 311-34.

---- (2007) ‘Achieving security, respecting rights and maintaining the rule of law’, in Lynch et al. eds. 2007: 222-33

Lynch, A., Macdonald, E., & Williams, G. eds (2007) Law and Liberty in the War on Terror (Leichhardt: Federation)

Mackey, C. with G.Miller (2004) The Interrogator’s War (London: John Murray

Pearse, J.J. (2006) ‘The interrogation of terrorist suspects: the banality of torture’, in Williamson, ed. 2006: 64-83.

Sales. L. (2007) Detainee 002: The Case of David Hicks (Melbourne: Melbourne University Press)

Stafford Smith, C. (2007) Eight O’Clock Ferry to the Windward Side: Seeking Justice in Guantanamo Bay (New York: Nation Books)

Strauss, M. (2003) ‘Torture’, New York Law School Law Review 48: 201-74.

Williamson, T., (2006) ‘Investigative interviewing and human rights the war on terrorism’, in Williamson, ed. 2006: 3-22.

Williamson, T. ed (2006) Investigative Interviewing (Cullompton: Willan).

Wilson, R.A., ed (2005) Human Rights in the ‘War on Terror’ (Cambridge: Cambridge University Press).

Zedner, L. (2005a) Criminal Justice, (Oxford: Oxford University Press).

---- (2005b) ‘Securing Liberty in the Face of Terror, Journal of Law and Society, 32: 507-33.

---- (2007a) ‘Pre-crime and post-criminology?’ Theoretical Criminology 11: 261-81.

---- (2007b) ‘Fixing the future? Precaution and the pre-emptive turn in criminal justice’ University of New South Wales Law Faculty seminar, 15 August 2007.


DPP v Thomas [2006] VSC 243

R v Thomas [2006] VSCA 165

R v Ul-Haque [2007] NSWSC 1251

Thomas v Mowbray [2007] HCA 33

[*] Dean, Faculty of Law, UNSW, Sydney 2052;
[1] For definitive discussion of these trends, see Garland 2001; Zedner, 2007a, 2007b. For a more positive view, see Dershowitz 2006.
[2] At the time of writing, the U.S. President has just vetoed a Bill which would have prohibited the CIA from using interrogation methods such as water-boarding: ‘Veto of Bill on CIA tactics affirms Bush’s legacy’ New York Times 9 March 2008. Meanwhile, Khalid Sheik Mohammed has been charged with murdering the 9/11 victims on the basis, inter alia, of confessions obtained by water-boarding.
[3] These are available on the website of The Hindu, at (viewed 23 January 2008).
[4] “Keelty attacked for ‘court testing’” The Australian 17 December 2007, p3.

AustLII: Copyright Policy | Disclaimers | Privacy Policy | Feedback