England and Wales Court of Appeal (Criminal Division) Decisions

IN THE SUPREME COURT OF JUDICATURE
COURT OF APPEAL (CRIMINAL DIVISION)
ON APPEAL FROM WOOLWICH CROWN COURT
MR JUSTICE BUTTERFIELD
T20047453

		Royal Courts of Justice Strand, London, WC2A 2LL
		04/12/2008

Lord Justice Hughes :

1. We have before us four applicants who were indicted for a major terrorist conspiracy. The principal in the plot was a man called Barot, whose case had already been to this court by the time these men were sentenced: see R v Barot [2007] EWCA Crim 1119. These applicants were four of a total of seven of those of his co-conspirators who were caught and indicted. One of them, Shaffi, challenges his conviction. The other three, Jalil, Bhatti, and Ul Haq pleaded guilty as the trial was about to begin. They challenge their sentences. The applications have been referred by the Registrar. In each case, we give leave to enable them to be argued.
2. Barot, when arrested in August 2004, had been plotting terrorist attacks on major public buildings in the United States and/or this country since at least 2000. He visited the USA in autumn 2000 and again in spring 2001 to research this scheme. He produced, at some stage, detailed descriptions and plans for the destruction of the buildings of four key American financial institutions in three different cities. His proposals were plainly written for the consideration of the Al Qaeda leadership or a similar organisation; they read as professional or corporate presentations. They clinically analysed the pros and cons of different methods of destruction. They were illustrated by photographs and a video. In the end those attacks on major American targets did not take place, perhaps because in the meantime yet more terrible attacks on the USA had been mounted on 11 September 2001.
3. Barot then turned his attention to plans for similar attacks on British targets. Although he spoke of having targets in mind, the materials in due course discovered did not reveal which specific buildings they might be. They did reveal detailed proposals for several different possible methods of destruction. The most detailed by far became known as the 'gas limos' project. It involved packing a number of stretch limousines with propane gas cylinders and explosives and detonating them in the underground carparks beneath suitable buildings. Other proposals were for a 'dirty' or radioactive bomb, using many thousands of an item of domestic equipment containing some radioactive material, the sabotage of a major rail artery, perhaps in a tunnel under the river Thames, and the hi-jacking of a petrol tanker for use to ram a building. The scheme was for all or some of these various attacks to be synchronised on the same day.
4. These various US and UK proposals were found both on an encrypted disc in the UK and on a computer recovered in July 2004 from an Al Qaeda connection in Pakistan. Barot had visited Pakistan early in 2004.
5. Barot and these applicants were all arrested on 3 August 2004. There had been some surveillance prior to that, but over a comparatively short period of about six weeks beginning in mid June. The indictment originally charged, as well as a number of subsidiary charges, conspiracy to murder and conspiracy to commit public nuisance by way of explosions. Barot eventually pleaded guilty in October 2006 to the principal count. The trial of these applicants was due to begin before Butterfield J in mid April 2007. The Crown applied to add, as new count two, a count of conspiracy to commit the offence created by section 2 Explosive Substances Act 1883, that is to say to cause explosions of a nature likely to endanger life. Although for the most part not immediately, the remaining defendants apart from Shaffi pleaded guilty to that count, and their pleas were accepted by the Crown. The count of conspiracy to murder was ordered to lie on the file, not to be proceeded with in the absence of the leave of the court. Shaffi maintained his plea of not guilty and was tried upon both conspiracy counts, presented as alternatives, as well as on a subsidiary charge. He was convicted by the jury of conspiracy to murder; in those circumstances no verdict was taken on the other counts.
6. The applicants who pleaded guilty did not all do so at the same time. Bhatti entered his plea on 18 April 2007, immediately after the new count was added, having earlier signalled his intention to do so. Discussions at court proceeded over the next several days. Jalil and Ul Haq entered their pleas on 25 April, as did defendants called Feroze and Rehman, with whom we are not concerned. A final defendant called Tarmohamed was the last to plead guilty; he did so on 30 April; we are likewise not concerned with his case. The start of the trial had been delayed whilst these discussions proceeded and whilst other steps in case management were taken; accordingly the pleas of guilty were all in the end tendered and accepted, albeit after the jury was sworn, before it was put in charge. The trial of Shaffi began a few days later on 2 May and lasted about six weeks.
7. Barot had been sentenced by the same judge in November 2006 to life imprisonment with a minimum term of 40 years. By the time the remaining defendants came to plead guilty or to be tried, it was known that an appeal by Barot against the length of that sentence was due shortly to be heard in this court, under the presidency of Lord Phillips CJ. After the pleas of guilty had been entered, and during Shaffi's trial, but before any of them had been sentenced, judgment was given in the appeal of Barot on 16 May 2007: [2007] EWCA Crim 1119. His life sentence was upheld, but the minimum term was reduced from 40 to 30 years.
8. The judge sentenced these applicants to determinate sentences as follows:

Jalil: 26 years

Bhatti: 20 years

Ul Haq: 18 years

Shaffi: 15 years

As to the three other defendants not before us, Feroze was sentenced to 22 years, Tarmohamed to 20 years and Rehman to 15 years. In the case of all defendants the Judge made those sentences the custodial parts of extended sentences under section 85 Powers of Criminal Courts (Sentencing) Act 2000, and imposed in each case an extended licence period of an additional five years.

9. It is important to remember that determinate terms of the order here passed on these applicants as the custodial element of their sentences are different in kind from the minimum term component of Barot's life sentence. In his case, the minimum term of 30 years is the period which must actually be served before release on licence can even be considered. In their cases the ordinary rules as to release on licence applied. In this instance those were the rules prevailing under the Criminal Justice Act 1991, unaffected in a case of this kind by the changes now made by the Criminal Justice and Immigration Act 2008, section 26. Under such rules the applicants are eligible for consideration of licence at the half way stage, entitled to release on licence at two-thirds (which is perhaps the more likely point of release in the context of offences of this kind), and would then remain on licence until the three-quarter mark, plus the five year extension. In other words, Barot was (a) by his minimum term in roughly the position of a man with a determinate sentence of 45, or more likely 60, years and moreover (b) subject to possible detention for the rest of his life. And for anyone now convicted of an offence committed after the commencement in April 2005 of the new rules relating to release under the Criminal Justice Act 2003, a minimum term of 30 years is the equivalent of a determinate sentence of 60 years.
10. The Judge made it clear in passing sentence on those who had pleaded guilty that he attached real significance to the fact that the offence which they had admitted involved a conspiracy to cause explosions which were likely to endanger life but did not involve proof, or admission, of an intention to kill. We are sure that in the context of this case he was right to do so. The more serious count of conspiracy to murder was on the indictment and the Crown had, for good reason, not pursued it given the pleas. Of course, on other facts it might be, as this court mentioned in Barot at paragraph 59, that conviction upon a count such as count two in this case carries on the evidence, either admitted or accepted by the judge after trial or Newton hearing, the inescapable inference that the defendant in question did have an intention to kill, but that is not this case.

Jalil: departure from sentence indication ?

11. Jalil (alone) contends that in passing a sentence of 26 years the judge erred in principle in departing from a limited sentence indication which he had given prior to the pleas of guilty being entered.
12. The defendants generally had sought a Goodyear indication from the judge. He declined to give one. His principal reason was that in a multi-handed case in which each defendant had detailed submissions to make about not only his own role but his role in comparison with that of others, it was not possible to give useful prior indications of what his conclusions were likely to be about their relative positions. In that, he was, if we may say so, plainly right, and the complex competing bases of plea advanced by the various defendants both before him and now before us only serve to confirm his decision. But the judge was persuaded by the defendants to give some assistance to them where he properly could. That occurred in stages.
13. First, on Wednesday 18 April, after the new count had been added, the judge announced that a defendant who pleaded guilty to that count and whose plea was accepted would (a) not receive a life sentence but a determinate one and (b) could expect significant reduction in recognition of his plea, of the order of 15-20%. He added (c) that he would sentence on a basis which included any activity before February 2001, which was the date on which there came into force the law permitting the new count on the facts of this case; that was designed to give comfort to any defendant contemplating a plea of guilty that he would be less likely to face further prosecution, and perhaps in the US, for earlier conduct.
14. Secondly, at the further invitation of counsel, the judge returned to the question of sentence on Friday 20 April. Among other submissions, the observation had been made to him that a determinate sentence might, in some circumstances, be so long as to amount in effect to the same thing as a life sentence. The judge said that he did not have that kind of length in mind. He then added this:

"If I were unfettered, upon authority to come, I would consider on the information I presently have, that the determinate sentence would certainly not exceed the determinate sentence ultimately held to be appropriate in the case of Martin and is very likely – I am afraid I can say no more than that – but very likely to be significantly less. That is the starting point before the discount that I have already indicated for pleas of guilty."

15. Thirdly, counsel for Jalil sought yet further clarification on the following Monday, 23 April. Martin had been to this court. The trial judge's sentence of 35 years had been reduced to 28. The judge was asked to which he had been referring and he made clear that it was, as one would expect, the eventual sentence approved in this court. But he added this:

"But I make it plain of course I am bound to be bound by any subsequent observations by the Court of Appeal and the Lord Chief Justice……you understand that of course."

16. Jalil's first submission in this court is that the sentence of 26 years passed upon him departed, "no doubt inadvertently", from the indication given. 26 years, says Mr Bishop QC, could not have been predicated upon a starting point, before reduction for plea of guilty, at or below the Martin level of sentence at 28 years. Accordingly, the sentence is said to have been wrong in principle for departure from indication given.
17. We have been taken to the transcripts with some care. Both the last observation which we have quoted and the earlier careful reference to 'authority to come' were very clear acknowledgments by the judge of the then imminent hearing in this court, before the Lord Chief Justice, of Barot's appeal against sentence. Moreover, it is clear from other parts of the transcript that counsel present were plainly conscious of the judge's need to enter this caveat. There had been a number of proper discussions in open court about the possibility of any form of sentence indication. Before the judge gave the first element of his indication the possibility had clearly been contemplated by counsel that he might be unable to give any assistance, even if otherwise minded to do so, until the outcome of Barot's appeal was known. The indications that life sentences were not contemplated, and that pleas of guilty would be recognised by significant discount, were not of course such as were likely to be affected by anything said in Barot. But reference to a ceiling upon determinate sentences was likely to be affected, if in Barot this court reviewed the level of sentencing suggested in Martin. Thus it came about that the judge's second element of his indication, given at the invitation of the defendants, was preceded by a lengthy discussion. In the course of it the judge explicitly pointed out this possibility. He even had enquiries made to see if it were known when judgment in Barot might be expected, and he relayed the negative answer to counsel in open court. He said this:

"It may be, for example, that he [the Lord Chief Justice] will take the view that Martin should be revised upwards, I simply do not know and nor do I know when he is going to deliver his judgment."

A little later he said this:

"Obviously, any defendant pleading guilty at this stage, not knowing what the Lord Chief Justice will ultimately say, does take a course of some courage…"

And when eventually he was persuaded by the defendants that it was possible to give a limited indication the judge said:

"....it does occur to me that I could give, without a term of years…a general indication of the likely level of sentence consequent upon a plea of guilty 'outwith' the decision of the Lord Chief Justice in Barot. I do not know whether that would be helpful or not, but inevitably if the Lord Chief Justice took a different view I would be bound by that different view. That is the problem."

He was specifically told, by counsel for Tarmohamed who made clear that he was speaking for defendants generally, that anything he could properly say would be 'very helpful'. The crucial second part of his indication followed shortly after that exchange. Consistently with that, we also see that when, some weeks later in the course of mitigation, the point was mentioned (by Mr O'Connor QC for Bhatti), the judge responded by immediately making it crystal clear that given what had been said in Barot about Martin, he could no longer regard the latter case as a guideline.

18. We are wholly satisfied that there was nothing remotely inadvertent in the judge's departure from the provisional indication that he had given as to the significance of Martin. He took the view that Martin as a yardstick had been overtaken by what was said in Barot, which was a possibility which he had conscientiously, indeed repeatedly, made clear that he had foreseen. He was not addressed on behalf of Jalil on the basis that he was bound by what he had said to pass a sentence based on a starting point dictated by Martin, and that approach by counsel correctly recognised the reality. Jalil's first argument, which he has advanced through fresh counsel who did not appear below, therefore fails.
19. The second phase of Mr Bishop's argument is that on the correct reading of Barot there was nothing in it which ought to have caused the judge to decide that Martin no longer represented a proper guideline for sentencing in this case.
20. Martin [1999] 1 Cr App Rep (S) 477 was a case of a Provisional IRA plan to blow up unmanned electricity substations in isolated rural locations. The aim was to disrupt the national grid. The charge was of conspiracy to cause explosions likely to endanger life or cause serious damage to property. The defendants contested the trial but were convicted by the jury. The plans were well made and well advanced; execution was close at hand. The court proceeded upon the basis that the aim was not to kill, but that the defendants were reckless as to the risk of death or maiming; there were to be 37 large bombs, there would probably have been warnings which would bring bomb disposal or other emergency service teams to the locations at risk to their lives, other people might be caught in the explosions and the planned sudden disruption in electricity supply might lead directly or indirectly to injury or loss of life. Attention was drawn to the difference between a bomb in a remote rural location and one planted in an aeroplane or crowded place. The court held that for a plan to cause terrorist explosions the range of sentence from 20 to 35 years, after trial, would accommodate most but not all cases. In reaching this conclusion the court paid particular attention to the then level of time in prison actually served by those convicted of murder, and particularly of the most serious murders. At that time, such terms were set by the Home Secretary, and the most recent statement of principle which he had made (in November 1983) indicated that for the gravest murders offenders could normally expect to serve at least 20 years, and perhaps more, in prison. As we have previously recorded, the court in Martin considered that determinate sentences of 28 years, rather than 35 years, were appropriate to the case of well advanced plans, imminently to be carried out, where the aim was destruction of property and infrastructure rather than mass murder, but the risk of injury or death as a by-product could not be ignored. As the court observed (at 484), 35 years would mean 17.5 to 23 actually served, and that would be the sort of term served by the perpetrator of a murder with severely aggravating features.
21. In Barot the judge's life sentence was upheld, but the minimum term was reduced from 40 to 30 years. That was principally because, since Barot had prevented his advocate from arguing otherwise, the judge had sentenced on the basis that both the US and UK plans were ones which would, but for detection, shortly have been carried out and would have brought about the terrible massacres intended. This court embarked on an enquiry into whether Barot's proposed method of explosion was in fact as effective as he thought, and found, on the expert evidence, that it would not have generated the particularly devastating type of massive cumulative explosion desired. That led to the conclusion that there could be no certainty that the plans would have been adopted, presumably by the Al Qaeda or other command to which they were submitted. Nevertheless, if attempted, the plans would still be likely to cause unpredictable and explosive consequences, and not in remote locations, but in populous cities. The court returned, as it had done in Martin, to the relationship between sentences for such terrorist plots and the time actually served by the worst kind of murderers. It did so in the light of the enormous change made in sentencing for murder by the provisions of section 269 and Schedule 21 of the Criminal Justice Act 2003, which raise the time actually served for the worst kind of murders to 30 years or more. It upheld the life sentence, on the ground that Barot presented a threat to life and limb whose duration could not reliably be predicted. But it held (at paragraph 60) that a minimum term of 40 years, as passed by the judge, should, save in quite exceptional circumstances, be reserved for the terrorist convicted after trial, of a serious attempt to commit mass murder by a viable method. Where the offence is of conspiracy, and the acts of the defendant fall short of imminent attempt, it held that the sentence should be lower.
22. Martin was much discussed in Barot. It was agreed on all sides that the level of sentencing in Martin could not govern sentences for terrorist conspiracies where murder was the primary object, but Martin was in any event not such a conspiracy. More importantly, for present purposes, the court accepted that terrorist offences are graver today than those of which this country had experience in previous times, reaching the level of public emergency threatening the life of the nation (see the observations of the House of Lords in A v SSHD [2004] UKHL 56, [2005] 2 AC 68). It concluded that "the guidelines suggested by the court in Martin require review." That also followed, it is plain, from the relationship between sentences for terrorist conspiracies (of the Martin kind) and time actually served by the worst kind of murderers, now that the latter has been so much uplifted by Parliament.
23. We accept that the present conspiracy, so far as Jalil, Bhatti and Ul Haq were concerned, did not involve an intention to kill; that it had in common with Martin. We also accept that the stage which planning had reached was more advanced in Martin than here, as this case has been explained in Barot. On the other hand, the potential for mass injury and loss of life was vastly greater in this case, which involved projected attacks on busy buildings in urban places, and radioactivity as well as explosions, than was the risk of injury and death in Martin. Even if the gas limos project was unlikely to produce the particular type of cumulative explosion which was hoped, explosion and fire in a car park under a busy building of some kind was nevertheless the aim. Mismanaged explosions are unlikely to be other than extremely hazardous to life, to say the very least. We see this case as certainly not less serious than Martin, although these applicants were entitled to some credit for plea, which the defendants in Martin were not.
24. Barot was not the occasion to attempt to re-state the bracket for terrorist conspiracies of the Martin kind, if only because it was a more serious type of conspiracy. Nor do we attempt the exercise here; it is far preferable to have more than a single case before overall guidelines are set out. But what is clear is that Butterfield J was correct to draw from Barot the clear proposition that the bracket of sentences postulated in Martin can no longer be regarded as a valid guideline for the cases of these sentence applicants. It is enough to note that the 28 years substituted in Martin was, on the basis of release at two-thirds, just a little less than was actually served at that time by the worst kind of murderers. Whilst we do not wish to be understood to mean that the answer lies generally in any particular arithmetical relationship, it is to be observed that the sentence of 26 years passed here on Jalil pre-supposes a starting point around 30-33 after trial, and probably about 30.5; that is to be considered alongside approximately double that served nowadays by the worst kind of murderers. It follows that the submission that the sentences here passed were wrong in principle for failure to abide by Martin must also be rejected.

Jalil; other grounds

25. Jalil contends that in passing the longest sentence upon him after Barot, the judge over-promoted him in comparison to the other defendants. We do not agree. It is certainly true that in this case, as in so many others, any attempt to discern an exact hierarchy of conspirators was unreliable; each of the defendants had his own relationship with Barot, and not all of them with each other. The judge was presented with a careful written basis of plea in respect of each of the three sentence applicants before us; they had been discussed with the Crown and contained in separate sections those matters admitted by way of participation, those accepted by way of mitigation, and those asserted in mitigation in respect of which there was no agreement so that the judge was invited to reach his conclusions. So far as Jalil was concerned, the first section plainly justified the judge's conclusion that of all the defendants he was Barot's most trusted associate, even if he did not know the exact breadth or detail of his plans, nor any specific targets Barot had in mind. He was not only very regularly, if not at times almost continuously, in Barot's company when surveillance was conducted over several weeks, but he was also a computer familiar who had, or had had, on his laptop and a memory stick vital research and planning files, including the critical 'Final Presentation'. He habitually made use of internet cafes rather than any computer traceable to him, including on one occasion as far away as Swansea, and he conducted coded E-mail through an account in the name bridget_jonesdiaries, selected as improbably that of a devout muslim. He had rented a safe house for Barot and himself to use and he had had, and had made use of, access to the principal store of material found in Bhatti's garage; the judge was entitled to infer that he had worked on the plans at Bhatti's home when Barot was out of the country. We regret that we are unpersuaded by the written submission that the fact that the judge did not demur point by point to submissions in mitigation as to the inferences which could properly be drawn means that he must have accepted those submissions. Nor does it follow from the proposition that others may also have worked on the materials found at Bhatti's home that Jalil's role was other than that inferred by the judge.
26. The subsidiary submission is made that the judge unfairly distinguished between Jalil and other defendants on the basis that there was some evidence that he adhered to a radical form of ideology. But it is clear from several points in the transcript that the judge accepted the proposition that what mattered was what a defendant had done rather than why; ideology went principally to future risk rather than to participation.
27. We are conscious that the judge had the whole of the material in this substantial case, and indeed had read it with a view to trial. Nevertheless, if we were able to find any error of principle in his assessment of Jalil's part in this conspiracy, we should intervene, but we cannot. The appeal of Jalil is accordingly dismissed.

Bhatti

28. The most powerful point to be made for Bhatti was and is that he pleaded guilty earlier than anyone else, and without the security of any, even limited, indication as to sentence. That must have taken some courage. It is possible that it contributed to the willingness of others subsequently to follow the same course. The judge clearly accepted this, and we would expect that of all the defendants who pleaded guilty, Bhatti received the largest reduction in recognition of his plea. It does not, however, follow that the judge was wrong to make a deduction of about 15%, or more likely 20%, rather than more. Bhatti had been awaiting trial since August 2004. The indictment always contained a count of conspiracy to cause a public nuisance by explosions, but he had not, until about December 2006, shown any sign of contemplating any plea of guilty to anything. He had put in two substantial defence statements denying any complicity in any conspiracy. There was absolutely nothing wrong with the judge's assessment of proper reduction in recognition of the plea of guilty.
29. Nor are we persuaded that the judge erred in assessing Bhatti's role. He was no doubt a reserved, not to say shy, man. There is a report from the prison suggesting that he has distanced himself from extremist prisoners. He was, like others for that matter, under the influence of Barot, who seems to have been a man of some charisma. But his role was highly significant. His garage was the repository of a mass of detailed working materials. His own domestic computer in the house had been used to work on the plans, including when Barot was out of the country; even if, as was suggested, Bhatti had allowed others to do this rather than done it himself, he was in possession of material which would if discovered reveal the full extent of the plans. The judge was entitled to note that he was an educated, indeed technical, man; it followed that he was closely trusted. He was frequently in Barot's company. Further, he was in possession of the necessary accoutrements of at least two false identities used by Barot, and he had retained an advertisement for a job as a driver of a LGV petroleum tanker. This was the evidence, together of course with an admission by his plea to being a knowing participant in a plan to blow up buildings in the UK. The absence of sightings during the surveillance period of him visiting the safe house, which was near to his own, does not necessarily carry the inference that he had been excluded from knowledge of it; it seems to us simply neutral.
30. Bhatti's mental health is somewhat fragile. He has developed OCD, associated with some depression. The judge rightly concluded that this was debilitating and made prison the greater burden to him. We are told that he is presently housed in a prison which does not have available the treatment which he needs and would ordinarily receive if at large. We recognise the pressures on prison places, particularly for high security prisoners, but we hope that before long that deficiency can be rectified. However, the judge clearly took his health into account, and the inevitable result of admission of so serious an offence was a sentence of such length that difficulties of treatment simply have to be resolved within the prison system.
31. Mr O'Connor submits additionally that the conditions for the making of an extended sentence under section 85 of the 2000 Act were not met, at least in the case of Bhatti. Such a sentence, sadly no longer available for more recent offences, could be passed if the normal period of licence would not be adequate to prevent the commission of further offences and secure the rehabilitation of the offender. The normal period of licence applicable to Bhatti's 20 year custodial term would be from (in all probability) the two-thirds mark, viz a little under 14 years, until the three-quarter mark, viz 15 years. It seems to us that the judge had a sound basis for holding that that would not be enough to prevent the commission of further offences and secure Bhatti's rehabilitation. Persons who have been party to serious terrorist plots such as this are very likely to attract the attention of similarly minded people, and are likely to be especially vulnerable to such attention in the time following release. The judge rightly had in mind the ability of licence conditions to keep track of the released prisoner, particularly by knowing where he lives; such conditions are useful in inhibiting any further criminal activity. Even if, as Mr O'Connor submits, Bhatti's history suggests that he would be unlikely himself to initiate further criminal activity, the risks to the public of others with an interest in him, and of the vulnerability to persuasion which Mr O'Connor prays in aid, are still there. If we may say so, the extended licence period was a thoroughly sensible form of sentence.
32. For all those reasons, the challenge to Bhatti's sentence fails and his appeal is dismissed.

Ul Haq

33. Ul Haq had experience in the construction and surveying of buildings which was directly in point for this plot. He was a long-standing associate of Barot and used his training to help him plan the attacks. He was peripherally involved at the US stage, without visiting that country, and he helped Barot research the UK plans. On his behalf Mr Carter QC takes a discrete point. He contends that the judge wrongly drew an inference that Ul Haq was still actively involved in the conspiracy in August 2004, when the conclusion should have been that he had disengaged from it about a year previously.
34. The submission that Ul Haq had withdrawn was based upon the fact that the last overt act attributed to him was providing a document from a University library in July 2003. After that, the evidence affecting Ul Haq was twofold. First, on 27 June 2004, during the surveillance stage, Barot visited him at home, having telephoned him first, and the two of them went to an outdoor location to talk, plainly so as to avoid the risk of being overheard. Second, on 2 and 3 August Ul Haq was shown to have spent some time at work viewing the treatment by two different internet news sites of the arrests and searches in Pakistan, in the course of which the plans made by these conspirators were recovered. Says Mr Carter, the first is equally consistent with Barot pressing a reluctant Ul Haq for further assistance and being refused, and with the anti-surveillance techniques being Barot's alone. The second, he says, was followed by mere normal working activity, which indicates that there was no more than curiosity from past involvement. It is certainly true that Ul Haq made no attempt to leave the country, and indeed when a few days later he learned that the police wanted him, he presented himself to them. Thereafter he maintained silence in interview and gave no indication of any intention to admit anything until the days immediately preceding the trial.
35. We do not accept that the judge was bound to infer from the material before him that Ul Haq had withdrawn or disengaged himself from the conspiracy in Summer 2003. The absence of evidence is neutral rather than necessarily evidence of absence; there was no positive evidence of withdrawal and the judge was entitled to attach significance to the meeting with Barot in June 2004 which pointed the other way. As to the viewing of the news items, what the judge inferred from it was that Ul Haq knew the significance of the Pakistani arrests, which in turn meant that he was privy to the taking of the plans to Pakistan, and thus a trusted member of the conspiracy. That conclusion was properly open to the judge. It is true that the Crown had suggested additional significance in the absence of automatically saved visits to such news sites on any other recent occasion since April, and it may be that a computer report submitted on behalf of Ul Haq raised the real possibility that that did not carry the implication contended for because the computer periodically wiped itself clean of visited websites. But the judge did not refer to that contention in sentencing, and, in any event, a separate part of the computer report, recording retained cookies, did suggest that the last visits to a news site had been in April.
36. In those circumstances we do not accept the proposition that the judge erred in principle, or passed a sentence on Ul Haq which was manifestly excessive. His appeal is also dismissed.

Shaffi: conviction

37. The issue in Shaffi's trial was comparatively narrow in scope. Barot had made two research or reconnaissance trips to the USA in August/November 2000 and March/April 2001. He had been accompanied on the first by Tarmohamed. On the second he was accompanied by Shaffi, though the pair seems to have been seen off at the airport by Tarmohamed and, when Shaffi fell ill part way through, he was replaced within 24 hours by Tarmohamed. There was no dispute that the trip had been made. The issue was whether it was proved that Shaffi was a knowing party to the research and reconnaissance or might have been an innocent and cosmetic travelling companion.
38. Three distinct grounds of appeal were advanced, with considerable clarity, in Shaffi's written grounds of appeal (a fourth was abandoned at the hearing):

i) that there was no jury ballot;

ii) that the judge was wrong to admit bad character evidence relating to a company of Shaffi's called Craigon Ltd; and

iii) that the Crown failed to disclose material which would have assisted Shaffi on an issue which arose under section 34 Criminal Justice and Public Order Act 1994 and to rebut a separate Crown suggestion of recent invention.

The jury ballot question

39. No less than about 90 potential jurors were summoned for this trial, which was then expected to be of seven defendants. The judge acceded to the posing of questions designed to elicit any connections or interests which might form the basis of challenge for cause. A number of potential jurors were released as a result, and others were released because their personal commitments made it impracticable for them to sit on a trial of the length anticipated. By 18 April the number had been reduced to 14. As to two of those, the defence submitted that they should be excluded on the basis of some connection to the armed services and/or the metropolitan police. The judge ruled that they were not to be excluded, but he was sympathetic to the wishes of the defence (then not confined to Shaffi) to achieve a jury without the complication of such connections if it could be done. He encouraged the Crown to stand these two jurors by if they were selected in the ballot. The Crown agreed. They were selected, and the Crown did stand them by. Shaffi was thus tried by twelve jurors against whom no possible complaint of unsuitable connection is made. What is contended is that in those circumstances there was no proper ballot because in effect there were only twelve "eligible" jurors. It has been held that a ballot requires more than twelve entrants: see R v Mulkerrings and Samson unreported 20 June 1997, 95/7501/Y4. This is thus a highly technical challenge to the trial, but if it is well founded it does not fail for that.
40. We have reached the clear conclusion that it is not well founded.
41. The first reason is that section 18 of the Juries Act 1974 provides as follows:

"(1) No judgment after verdict in any trial by jury in any court shall be stayed or reversed by reason –

(a) that the provisions of this Act about…the selection of jurors by ballot have not been complied with…

(2) Subsection 1(a) above shall not apply to any irregularity if objection is taken at, or as soon as practicable after, the time it occurred, and the irregularity is nor corrected."

42. The transcript makes it clear that no objection was taken to the suggested absence of ballot. Various defence counsel, including Mr Griffiths QC for Shaffi, took objection to various potential jurors, on the grounds of what were suggested to be unsuitable connections, but that is not an objection to the ballot. On behalf of a different defendant, Feroze, counsel submitted at the stage when the numbers had fallen to 14 that the process over the preceding day(s) under which many potential jurors had been released did not itself constitute a ballot, but that again is a quite different question. The 14 potential jurors were balloted and empanelled in the conventional way. The nearest anyone came to taking the objection which is now raised in this court was that when the judge asked counsel for the Crown whether there was any obstacle to the procedure which was in fact adopted, he remarked that the position of the defendants appeared to be that "we do not know but we are not conceding that we are not going to take it [sc the point] hereafter." No counsel had in fact taken the objection, and this statement by the judge of what he guessed (it may well be correctly) to be their position does not amount to taking it.
43. The second reason is that what took place was in fact a ballot. There were 14 potential jurors from whom twelve were selected after two were stood by. Standing by only puts the potential juror to the back of the list. If there had been a successful challenge for cause to one of the twelve, or for any other reason one of them could not sit, one of the two potential jurors under discussion would have served. The argument for Shaffi is that because the judge ought to have excluded these two potential jurors for unsuitable connections, they were therefore "ineligible" and thus could not go into a ballot. That does not follow. Even if the judge were wrong to hold that the two potential jurors under discussion could properly sit, that would not mean that there was no ballot. The question which arose in the case of the two jurors under discussion was not a question of qualification to sit; by the Juries Act everyone correctly resident, not mentally disordered and between the ages of 18 and 70 is qualified as a juror unless disqualified by Schedule 1, Part 2, on grounds connected with criminal history. The question which arose in the case of the two jurors under discussion was one of possible challenge for cause on grounds of what is technically called apparent bias (see Porter v Magill [2001] UKHL 67, [2002] 2 AC 357 and R v Abdroikov [2007] UKHL 37, [2007] 1 WLR 2679). But the time for a challenge for cause to be made is after the ballot. Section 12(3) provides:

"A challenge to a juror in any court shall be made after his name has been drawn by ballot…and before he is sworn."

44. That does not, of course, mean that the oft-followed and convenient procedure of inviting the judge, where all parties agree, to rule in anticipation upon a possible challenge for cause is wrong, or ought not to be followed. It is not always appropriate, but if the judge agrees to do so, and rules for the defendant's prospective challenge, no doubt the potential juror in question will be excused there and then and will not go into the ballot. Equally, if the judge is against the defence submission, subsequent argument in this court that the jury was in consequence not free of apparent bias is no doubt open to the defendant. In neither case is there any need to go through the rigmarole of renewed challenge and repeated ruling at the ballot stage. If objections are upheld, it is no doubt possible that the pool of jurors might be reduced to the point where a ballot is impossible. But that is not this case. Here the judge ruled against the prospective challenge for cause. A ballot followed of 14 potential jurors. The present ground of appeal does not relate to any apparent bias of the jury or any member of it. It is a technical challenge to whether there was or was not a ballot within section 11. There was.
45. In those circumstances it is not necessary to investigate the contention that the judge erred in ruling that the two potential jurors could sit and we have not done so.

Craigon Ltd

46. Shaffi gave evidence that he was not a terrorist but a businessman working in the mobile telephone industry. He relied on the fact that he had said as much on arrest. He put his character in issue by volunteering some history of drug abuse, which he contended was in the past. He adduced evidence by cross-examination that the police had found a business plan for Craigon Ltd in the search of his home.
47. Craigon Ltd had been formed in May 2002. Shaffi was its principal. The police had looked at its bank account and found that, after opening in June 2002, there had been little or no activity until January 2003 when suddenly there were credits totalling something like £30m in a short time, which money was rapidly paid out again, often in back to back transactions. The account bore the appearance of the company being used as a vehicle for fraud, perhaps such as a carousel or missing trader fraud. Shaffi had been interviewed about the use of this company account and the police suspicions that it was fraudulent had been put to him, together with questions designed to investigate whether there might be any terrorist connection.
48. The Crown made no use of this material at trial, having formed the view that there was no evidence of terrorist connection. But when Shaffi gave the evidence he did, relying in part on this company as evidence of an honest business career, it applied to adduce the evidence pursuant to section 101(1)(f) Criminal Justice Act 2003, that is to say in order to correct a false impression given by the defendant. The judge admitted it. It is contended that he was wrong to do so.
49. This decision called for the exercise of judgment by the judge. Such judgments are within the range of those particularly committed to the trial judge with the feel for the trial, its currents and undercurrents, which only he can have. If he has erred in principle, or arrived at a decision which was not properly open to him, this court will interfere, but it does not simply substitute its own judgment in the absence of much of the material: see for example R v Renda [2005] EWCA Crim 2826, [2006] 1 Cr App R 24, page 380 at paragraph 3. Mr Griffiths submits that the evidence of the bank transactions lacked sufficient particularity to be categorised as a carousel fraud and thus lacked sufficient probative force to be admissible. We do not agree. The appearance of fraud was plain, whichever variety it may have been, and neither counsel nor Shaffi could or did gainsay it as at the time of trial, although it was contended that another director had transferred the funds from another company. The judge was entitled to say that the jury would be left with a misleading impression, as a result of the defendant's evidence, if it carried away the message that he was a bona fide businessman and for that reason less likely to be guilty of association with terrorist endeavour. He directed himself correctly as to possible exclusion under section 78 Police and Criminal Evidence Act 1984. He gave the jury a careful direction about how to approach this evidence, advising caution. No complaint is, or could be, made of his treatment of the evidence.

Non-disclosure

50. For the most part Shaffi had refused to answer questions when interviewed by the police over several days. He had on one occasion put in a prepared statement, which dealt with his possession of various items none of which was relevant by the time of the trial, but otherwise he had said nothing. At trial he gave evidence that he had indeed accompanied Barot to the USA but gave an account of a trip which was, for his part at least, entirely innocent. Since all that was new, the jury had to consider section 34 Criminal Justice and Public Order Act 1994. Shaffi's case was that he had relied upon very firm advice to stay silent which came from a particular solicitor or her various representatives. There was evidence available to him that that solicitor took an invariable line with suspects that they should not answer police questions, and that she was acting for many of those arrested in this enquiry. Such had been the concern at the time of the investigation that a 'party line' was being adopted which might not recognise the possibly conflicting interests of different suspects that the senior investigating officer had taken the unusual step of addressing all the suspects, in the presence of the various representatives and in carefully considered terms, on their right to make up their own minds. In addition to that, one of her representatives who dealt with Shaffi had disagreed with the advice, and had said so at the time to one of the officers. Shaffi called that gentleman to give evidence, which was accepted by the Crown. The complaint now made is that that gentleman had met erstwhile police colleagues soon after voicing his anxiety and the Crown did not disclose a note of what had passed between them.
51. Secondly, the Crown relied against Shaffi on an emotional telephone call which he made from the police station to his father. The Crown suggested that its content amounted to a confession that the American trip had been a guilty one. The defendant had been taxed with it very shortly after the call was made. Shaffi for his part said that what he had said had been partly mis-recorded and otherwise misunderstood. He called his father to say the same. The Crown suggested to both men that the claim of misunderstanding was a recent invention. Shaffi called Professor Noibi who had visited him about two days after the telephone call; his evidence was that the claim of misunderstanding had been made to him then. The complaint now made is that Professor Noibi had spoken of this to a police liaison officer at the time of his visit, but the Crown did not disclose what he had said.
52. The duty of disclosure is, in a case with the kind of weight of paper that this one had, a peculiarly onerous one. Nevertheless, a failure to discharge it may sometimes make a conviction unsafe. The duty extends throughout the trial, and we agree that what appeared irrelevant at one stage may become relevant as the trial develops. It is, however, very doubtful that there has been any significant failure of disclosure here. There was a note of the conversation between the solicitor's representative and ex-colleagues in the police, but we are told that it contained nothing of relevance. As to Professor Noibi, there was no record of any conversation which he had with the liaison officer, so there was nothing to disclose. In any event, the defendant was able to call both gentlemen and their evidence was unchallenged. The evidence of the solicitor's representative supported the other evidence that the defendant had received very positive advice to remain silent, which advice may have been general rather than tailored to his own particular situation. The evidence of Professor Noibi effectively undermined a particular line of cross-examination previously undertaken by the Crown. If either witness had made a previous and consistent statement to the same effect, that would not have been admissible, nor would it have improved his evidence. Even if, contrary to our clear impression, there was a stage during the defendant's evidence when the Crown ought to have disclosed some information about what one or other witness had previously said, that deficiency in disclosure made in the end no difference because the defendant was able to rely on everything relevant which the two men could say.

Shaffi: conclusion

53. For these reasons, we are satisfied that Shaffi's conviction is not unsafe on the grounds argued, whether separately or cumulatively. His appeal is accordingly dismissed.