First chapter of Onder de Tap (2006), about the tapping of telecommunications in The Netherlands, by Wim van de Pol.
The Advocate General of the Court of Appeal often assumes the role of benign authority in a hearing. He would prefer to be inviolable and tends to maintain an air of detachment. He will not have been personally involved in the criminal investigation, unlike a public prosecutor, whose job it is. As a participant in the proceedings, he leaves the getting excited to the lawyers, who grasp at whatever straws remain available to their client on appeal. Not so the Den Bosch Advocate General, M. Kolkert at the Baybasin trial late in the afternoon of 6 May 2002. He repeatedly interrupted Baybasin’s lawyer, Pieter Bakker-Schut, while he was presenting his arguments, by slamming his fist down on the table in front of him. ‘I object to my integrity being questioned in this way, time and again’, he shouted. He had actually hit the nail on the head, because that was indeed exactly what happened.
Kolkert had just listened to a five-hour counsel’s speech from Baybasin’s other lawyer, Adèle van der Plas, who had appeared to tear to shreds his earlier assertion that any talk of telephone tap manipulation was nonsense. Kolkert and his Public Prosecution Service had been driven hard onto the defensive, which is another unusual position for an Advocate General to find himself in. Acting on Kolkert’s instructions, an officer of the national public prosecutor’s office, Taco Stein, produced a five-page official report to persuade the Court of Appeal that all was above board in the practice of phone tapping in the Netherlands. But the lawyers’ pleas, which were based on investigations carried out by telecommunication experts, went into remarkable detail and convinced the Court of Appeal of the theoretical possibility of outsiders tampering with telephone taps. The Dutch wiretapping room is apparently not as secure as it could be.
Nonetheless, Kolkert ultimately triumphed in the ruling given on 30 July 2002, in which Baybasin received a life sentence for complicity in murder, inciting liquidation and involvement in the heroin trade. The Court of Appeal stated that while manipulation may be possible, actual tampering had not been demonstrated in the Baybasin case, in which the Court of Appeal relied on the opinion of the Netherlands Forensic Institute.
The ruling was to the visible despair of the lawyers, who said that tampering with the taps was plain to see in the Baybasin case. Van der Plas was later to remark in the ‘Zembla’ television current affairs programme that the wiretapping vicissitudes had left her so disillusioned with the Dutch legal system that she had considered giving up her practice.[1] Both the Supreme Court and the European Court of Human Rights rejected Baybasin’s appeal. But Van der Plas will not rest before she has fought the Baybasin case to the last gasp. Her hope is now based on the steady flow of new facts that have come to light since the Court of Appeal ruling in 2002, which may eventually persuade the Supreme Court to send the case back to the Court of Appeal.
Kurdish leader
Huseyin Baybasin (born on 25 June 1956) is a Kurdish businessman and a member of one of the most prominent families in south-east Turkey. He was conscripted into military service as a youth, and later served in a special army unit. By his own account, he performed a variety of clandestine assignments for the Turkish state, such as acting as an informant and heroin trader. Baybasin was arrested in Istanbul in 1976 with 11 kilos of heroin in his possession.
His activities for the Turkish state came to an end in 1984 after he was imprisoned in England. When he was finally freed, following Turkish government intervention, he refused to resume his old network of informants. After his release in 1989 he became increasingly involved in Kurdish affairs, and helped found the Kurdish parliament in exile in Brussels.
Baybasin emphatically denies ever having performed clandestine activities for the Kurdish resistance. He claims that he and other influential Kurds had urged PKK leader Öcalan to abandon the armed struggle.
Baybasin’s public profile in Turkey had risen steadily, which was partly why he had become a state enemy of the same calibre as PKK leader Öcalan in the eyes of the security services and the popular press. The Turkish state considered Baybasin a deadly enemy from the early 1990s, and all the more so when he started to make statements in the media about his joint involvement with the Turkish state in the drugs trade. A former Turkish secret service employee said in a statement to Baybasin’s lawyers that his bosses had been attempting since the mid 1980s to ‘remove’ Kurdish businessmen and criminals who supported the PKK. A whole list of Kurdish babas (underworld bosses) were also actually killed. Baybasin, himself once a Turkish state informant, was also in line to be murdered, according to this ex-secret agent. An important reason in Baybasin’s case was his direct contacts with the PKK. This source also mentions Baybasin’s statements in the media on Turkish state involvement in the drugs trade. [2]
Statements on the links between organized crime and the Turkish government were an absolute taboo until 1996, when the Susurluk incident upset matters. The closeness of relations between the Turkish worlds of organized crime and officialdom were revealed in the aftermath of a road accident in Susurluk. One of the crashed cars contained a rather motley bunch of people. Underworld boss Abdullah Çatli, a leading ruling party politician, the Istanbul police chief and a top model were travelling together in a limousine. ‘Susurluk’ marked the point at which the close-knit interests of the Turkish state and organized crime stopped being a taboo in Turkey. Indeed, they became the opposite, as one revelation followed another of relationships between the drugs mafia and the Turkish government.
But Baybasin was taking an enormous risk when he spoke about Turkish corruption in the media long before the Susurluk incident. The Turkish secret service started to deal with Baybasin with a series criminal prosecutions, and he was said to have been tortured repeatedly while in detention. His lawyer, Adèle van der Plas, says he narrowly escaped several murder attempts by paramilitary commandos in the 1980s and 1990s. Baybasin finally disappeared from the scene in Turkey to re-emerge as an asylum seeker in the Netherlands in the 1990s.
Turkey requested Baybasin’s extradition in 1995 as a member of the PKK. This was the start of a series of legal proceedings that are still ongoing. The then Minister of Justice, Winnie Sorgdrager, decided to comply with Turkey’s wishes, against which Baybasin appealed, arguing that he would be treated badly in Turkey because he was a political refugee. Baybasin won in interim injunction proceedings on 28 October 1998, and was given leave to await the outcome of his asylum application in the Netherlands. The president of the District Court in The Hague disallowed Baybasin’s extradition because he had been tortured previously in Turkey, and probably would be again following extradition, in view of his links with the Kurdish resistance.
6000 taps
At the time of his trial, Baybasin was treated as a lethal state enemy. The Rotterdam District Court and the court bunker in Amsterdam Osdorp were on maximum security. He was escorted by several bullet-proof cars carrying heavily-armed personnel. At times, he was moved around by helicopter. The Ministry of Justice considered Baybasin to be one of the most dangerous criminals ever to have been tried in the Netherlands. Although he has now been placed under a less stringent regime, Baybasin was detained between 1997 and 2002 in the high-security prison in Vught, where between ten and twenty criminals who are considered likely to attempt an escape are subjected to the strictest prison regime in the Netherlands. The regime is especially tough psychologically. Physical contact with visitors is out of the question. On leaving the cell, the prisoner first has to put his hands through a hatch to allow handcuffs to be put on. On returning to the cell, for example after sport or exercise, there is a complete search, including an anal inspection, sometimes several times a day. Baybasin was the prisoner who had been detained longest in the high-security prison, having been there for five years. He has since been transferred to an institution where the regime is less harsh.
The massive case file consists almost entirely of telephone taps. Some 6000 of Baybasin’s telephone calls were monitored between September 1997 and March 1998. The calls were in English, Turkish and in some extremely hard-to-understand Kurdish dialects. The Court of Appeal based its conclusion that Baybasin had ordered up to two murders almost entirely on the basis of those telephone calls. Baybasin denies these allegations, and insists that the telephone calls were tampered with. Baybasin’s and his lawyers’ great frustration is that they were never given the opportunity to allow experts finally to determine whether their suspicions were justified. The Public Prosecution Service says that all the commotion over phone tapping during the Baybasin case is unfounded, because manipulation had never been demonstrated.
The crux of the matter for Baybasin is that the Den Bosch Court of Appeal refused to instruct the Public Prosecution Service to allow external experts to examine the original optical discs on which the calls were recorded.[3] Baybasin’s lawyers lost several proceedings aimed at gaining access to the original intercepted material.[4]
As far as the Public Prosecution Service and the Court of Appeal are concerned, Baybasin is an extremely dangerous and ruthless murderer who has people liquidated from a distance through his criminal organization. They point, for example, to Baybasin’s obscure use of language in the calls and the vague hints that would appear to refer to a successful liquidation in an Istanbul tea garden. For instance, Baybasin was said to have phoned from the Netherlands to inquire about the success of the tea garden murder (which actually took place). The ‘thing’ in the garden was reported to be ‘done’ and Baybasin expressed satisfaction.
Another matter was concerned with a murder that was never committed. The Court of Appeal concluded that Baybasin wanted to extract revenge for the murder of someone in ‘that hot country’ (Spain) on one of the murderer’s brothers, who lived in the United States. The target is a Kentucky civil engineering professor with no connection whatsoever with the murder in Spain. Baybasin phoned someone in the United States, and the language used in that call was also vague. At a certain point, Baybasin asked his contact in the United States to do something for him, the precise nature of which was unclear. One of Baybasin’s remarks was: ‘If the people there want to buy several cars at the same time, they will have to do so, but only if there is no alternative.’ The Court of Appeal’s interpretation was: Baybasin was prepared for other people who might be in the same room as the liquidation target to die if there was no other option. The file contains precious little evidence for this liquidation order other than this excerpt from the telephone tap. It was nevertheless enough for the Den Bosch Court of Appeal to pass sentence.
Baybasin’s numerous calls are characterized by vague use of language. At no time do the participants openly discuss heroin or liquidations. This is consistent with the vague language used on the phone by drugs criminals in countless criminal files. Many criminals are sentenced to long prison terms, often rightly, on just this kind of call. The telephone taps in cases of this kind usually play only a secondary role, alongside, for example, witness statements, observations, or bank statements. The unique feature of the Baybasin case is the prominent role of the telephone taps in the evidence. In some matters, the telephone taps form the only evidence, and in others almost the only evidence, for drugs transports and liquidations that took place very far away. The file on which Baybasin was given a life sentence contains almost no other evidence.
The credibility of Baybasin’s alleged involvement in the tea garden murder of a certain Öge is now the subject of much discussion in Turkey itself. The ‘tea garden case’ boils down to an allegation that Baybasin ordered the murder in phone calls with two accomplices. The Turkish judicial authorities originally wanted to convict both these accomplices, Korkut and Yafuz, of the murder. The public prosecutor in the Yafuz case himself requested the court early in 2004 not to proceed. The public prosecutor in the Korkut case did not even request a criminal prosecution. The Turkish judicial authorities considered that there was insufficient evidence in both cases for Korkut’s and Yafuz’ involvement in the tea garden murder. The same file was used in the Netherlands to imprison Baybasin for life as the person who gave instructions to Yafuz and Korkut.
means of exerting pressure
The fight against serious Turkish crime has traditionally been a task for the North East Netherlands Interregional Crime Squad in Zwolle.[5] Its work mainly involves damming the flow of heroin that Kurdish and Turkish organizations send to west Europe. Contacts between officers of the Turkish and Dutch police and judicial authorities have been excellent since the mid 1990s. The relationship is close both formally and informally, with Turkish officers often lavishly entertained in the Netherlands, complete with visits to Amsterdam’s red-light district.[6] In return, Dutch police and judicial authority staff were received in Turkey by the Turkish authorities, and Dutch people would sometimes come into contact with parts of the Turkish state apparatus that took human rights less than seriously.
After the Susurluk incident mentioned above, it gradually became clearer that the boundaries between the police, underworld and secret services in Turkey were not always well defined. Relations appeared to exist between Tansu Ciller’s governing party, the Grey Wolves, the Turkish police and the underworld. The contacts between the Dutch police and the Turkish authorities were also substantially compromised by the Susurluk incident because of the presence in the car of Istanbul’s chief of police, not to mention the fact that the illustrious and criminal company was on its way to a hotel for a meeting with the Minister of the Interior, Mehmet Agar. Agar had been an important point of contact for the Dutch police in his earlier positions. Later in this chapter we show how Agar intervened directly in the Baybasin case.
It is noteworthy that no one in the Netherlands publicly questioned these links, although a Ministry of Justice report did criticize the informal contacts between Dutch and Turkish officers. The official channel for cooperation with Turkey is international legal aid, in which the Minister of Justice has ultimate responsibility, with actual coordination by the relevant public prosecutor. The Office of International Legal Assistance in Criminal Matters is meant formally to play a key role between the ministry and the Public Prosecution Service, but it far from always actually does so. The investigation conducted by the judicial authorities revealed that in practice the contacts with the Turkish judicial authorities were of a largely personal nature. The public prosecutor and the police conducted matters directly with Turkey.[7] This approach avoided the sluggish contact through the Office of International Legal Assistance in Criminal Matters. Informal contacts are faster and more efficient: investigators compiled their own requests for mutual legal assistance.
Not that the Netherlands is the only European country where the police rub along with the various Turkish police and intelligence services. The list of attendees at a conference on the Turkish drugs trade that the Dutch police organized in Rotterdam in 2001 contains many German, Belgian and British officers alongside members of the Turkish judiciary.[8] Another significant point is that only one of the interpreters on the list, who is of Turkish origin, is apparently so important that he figures between dozens of international investigating officers and magistrates. He works for the Amsterdam police. An extraordinary interpreter with a seemingly special position in the Dutch investigative authorities. In view of all the contacts between the Netherlands and Turkey, his would be an excellent position for an informant in Turkish service.
A public prosecutor who played an important role in Turkish drugs investigations, Klunder, was a witness in the Breda District Court at the Baybasin hearing, and gave evidence that he was aware that one of the four police interpreters also acted as an informant for the Turkish secret service. In other words, it is established that at the time of the Baybasin investigation, the Turkish secret service had infiltrated the Dutch police. This has also been confirmed by Anton Schalks, who was a liaison officer in Istanbul for the Central Criminal Investigation Department in the late 1990s. He says in a statement to the examining magistrate that one of the interpreters ‘was a member of the Baybasin investigating team’ and he passed on ‘solicited and unsolicited information from Dutch investigations’ to the Turkish authorities. It is apparent that this was indeed a serious breach of police integrity because the National Criminal Intelligence Division and the IJsselland chief of police ordered an investigation.[9]
The Baybasin case very clearly illustrates that haggling was going on informally behind the scenes by Dutch officials at all levels. When Baybasin’s extradition failed, the intimate contact between Turkey and the Netherlands was very helpful to the Turks. It was then important to the Turks to have the prestigious state enemy put behind bars. There are crystal clear signs that Turkish officers used their informal influence within the Dutch authorities. If Baybasin was not to be moved to Turkey, then he would just have to be criminally convicted in the Netherlands. Anonymous information started to arrive at the North East Netherlands Criminal Intelligence Unit in 1996, suggesting that Baybasin was involved in large-scale international heroin trading and liquidations. This information is the basis on which the Dutch Public Prosecution Service started a preliminary inquiry into Baybasin in 1997. The Turkish authorities were delighted with Turkish press reports that the Netherlands was criminally prosecuting Baybasin. Hadn’t they always said that inside Baybasin there was a dangerous drugs dealer trying to get out?
It is speculation and impossible to prove, because the information of the criminal intelligence services is by its nature highly secret. Nonetheless, it is extremely plausible that, given the good contacts between the Turkish and Dutch police, much criminal information on Baybasin found its way clandestinely from Turkey to the Netherlands. This information then subsequently emerged anonymously at the criminal intelligence service of the North East Netherlands Interregional Crime Squad, where it was booked as coming from ‘an informant who reported to the Criminal Investigation Unit’. In fact, however, the information was Turkish and probably arrived unlawfully (i.e. informally) at the Criminal Investigation Unit in the Netherlands from the Turkish police or intelligence services. Besides the origin of the information, its reliability is also questionable. It is at any rate impossible to corroborate.
What is certainly not a matter of speculation is that the Turkish and Dutch authorities haggled over Baybasin. The clues that this is the case are extremely strong. An internal telephone memo from the Immigration and Naturalization Service (IND) of July 1997 already mentioned negotiations about Baybasin.[10] The memo was dated several months before the start of the preliminary inquiry into Baybasin. It would therefore appear that Baybasin’s asylum status depended in part on negotiations between the Dutch state (the Ministry of Justice) and Turkey. This is surprising, because Baybasin’s extradition proceedings were still sub judice at the time. At any rate, the document states that the Netherlands was using the Baybasin case ‘as a means of exerting pressure with a view to getting something out of the Turkish authorities in another case’.
The national police secret service worker mentioned above says that the service had been looking for Baybasin abroad since 1993, and finally found him in the Netherlands. On two occasions, a ‘team of three people’ had been on the point of liquidating Baybasin in the Netherlands. The liquidations did not happen because the team was not in the right position and ‘did not dare to take him out’. The Turks then shifted tactics to contacting ‘certain people in authority in the Netherlands’.
‘The intention was to detain him [Baybasin] in prison for a long period at least’ (…). I met E.G., the procurator general of the Istanbul State Security Court, in the summer of 1997 over a meal in the Hanedan restaurant in Besiktas. A friend, M.Y., was also present. During the conversation, the procurator general [of the Istanbul State Security Court –WvdP) commented that ‘the bomb would be exploding very soon, Huseyin Baybasin would regret ever being born, he would be taken out’. (…)’[11]
This ex-officer’s statement also mentioned the associated collaboration with Dutch people:
The same procurator general said during visits to him in September and October 1997: ‘Baybasin will be gone very soon now, that Agar is a very brave man, no one can do anything against him. Look at how he made a deal with the Netherlands procurator general. What was his name, Hilering, or something like that? [this could be a reference to public prosecutor Hillenaar -WvdP] That man gives us information day and night; he is even less patient than we are’, (…) ‘Baybasin is not dead but he will die in prison, we will play a game for him that he will never forget’.
Incidentally, although Hillenaar was not a procurator general, he was the public prosecutor who conducted the preliminary inquiry into Baybasin in September 1997. This source says that the Dutch public prosecutor acted as a conscientious informant for a Turkish intelligence service, at any rate providing ‘information day and night’.
The ‘mighty’ Agar mentioned can refer to none other than Mehmet Agar, who in 1997 was Minister of the Interior and member of the party of the Grey Wolves, the True Path Party DYP, which was then in power. Agar played a central role in Turkey in the contacts between the police, secret service and the drugs trade. It transpired later that those involved in the 1996 accident in Susurluk were on their way to a hotel where Mehmet Agar was waiting for them. Prior to being appointed minister, Agar’s employers included the secret service, and he was once the highest chief of police in Turkey. Agar in his capacity as minister also appeared to have personally issued a firearms permit and a diplomatic pass to Abdullah Çatli, the Mafia boss who died in Susurluk. According to a statement from Mehmet Eymür, a former senior officer of the Turkish secret service MIT, Mehmet Agar also arranged for 80 kilos of heroin to be smuggled into west Europe. This statement suggests that it looked very much like it was this Mehmet Agar who had personal contact with Dutch officers in the Baybasin case.
The statement further indicates that Turkey and the Netherlands had agreed as early as the summer of 1997 to prosecute Baybasin criminally, independently of the extradition procedure. This is also an indication that there were possibly unacceptable contacts between Dutch and Turkish officers during the Baybasin investigation.[12]
The Dutch police liaison officer in Istanbul, Anton Schalks, was to have nothing to do with investigating Baybasin, despite his position as contact person between the Dutch and Turkish police. The head of the investigation team had told him personally that ‘it was because the investigation concerned was politically extremely sensitive.’ Indeed, the exchange of information on Baybasin between the Dutch and the Turkish police apparently did not go through the usual channels.[13] It would appear that other channels were more appropriate for Baybasin.
All the wrangling about the telephone tap reports that Baybasin objected to, was played against this shady background of secret service activities.
irregularities in the taps
What precisely were the points that Baybasin, his lawyers and some experts and journalists took offence to in the tap file? We list below the curious aspects of the tap reports as they gradually emerged. This list ultimately led to the latest signal analysis of 2004, which shows that the account of the taps in the file cannot be correct, and the telephone calls had indeed been tampered with.
In the first instance it was Baybasin himself whose doubts about the degree of truthfulness of the tap reports were raised when he read his file in the remand prison in 1999. His lawyer at the time, David Moszkowicz, was given access to several calls recorded on cassette tapes. The Public Prosecution Service said that the calls had been copied from the computers of the wiretapping room, actually from the optical discs, the permanent storage media of the tap system. It is important to realize at this point that all the discussion surrounding the audio recordings was actually concerned with a couple of cassette tapes with only 140 out of an estimated 6000 telephone calls. The thousands of other calls were never made available, and therefore never heard by Baybasin or his lawyers. It is also significant that the evidence presented by the police consisted almost entirely of tap reports, the majority of which being summaries in Dutch made by police interpreters.[14]
Baybasin and Moszkowicz drew the District Court’s attention to frequent audible clicks on the tapes. The Netherlands Forensic Institute was duly instructed to examine the tapes for manipulation, but was unable to find anything improper. The Breda District Court sentenced Baybasin to twenty years’ imprisonment, the maximum term for murder, and inciting liquidation.
The next step as far as the telephone taps were concerned came only in 2002. Baybasin, then in the final phase of his appeal, had engaged the lawyers Bakker-Schut and Van der Plas. The trial was almost at the stage of public prosecutor’s demand and counsel’s plea. Bakker-Schut and Van der Plas succeeded at the last moment in finding witnesses and experts capable of investigating the taped telephone calls independently of the Netherlands Forensic Institute. This was unusual, because the world of engineers who work for the investigative services is as tight as a drum. Anyone telling tales about the systems of his employer or client would be putting his livelihood at risk. Martien Kuylman and Hans van de Ven nonetheless accepted Bakker-Schut’s and Van der Plas’ request to conduct additional investigations. Kuylman had been closely involved in the development of the tap system for the Amsterdam-Amstelland police region and was later head of security at UPC. Van de Ven had been head of signal analysis at the Military Intelligence and Security Service and had worked for some considerable time for companies that develop tap systems. Both had moved on to work independently as consultants.
Kuylman and Van de Ven observed in a number of reports that the tape recordings exhibited serious irregularities.[15] For instance, there were inexplicable audible mechanical switching clicks in the calls. These clicks sometimes coincided with sudden silences in the frequency pattern, resembling the signal-free interval that occurs when starting a tape recorder in recording mode. The experts also discovered differences in the frequencies of the tones within a single call. Finally, tones could be heard in GSM calls that can be generated only by exchanges for land-line telephony.
Like Baybasin, the experts had been given access to a number of cassette tapes onto which calls had been copied from permanent storage (an optical disc). They immediately stated emphatically that further analysis of the master material was necessary before any definitive statement could be made on the cause of the irregularities. Only then would they be able to determine whether the calls had been manipulated or not. To this day, there has been no further investigation because the Public Prosecution Service has never been willing to make the original material (storage media) available. And neither the Den Bosch Court of Appeal nor the Court of Appeal in The Hague in interim injunction proceedings honoured Baybasin’s request to order the Public Prosecution Service to do so.
There were other irregularities, besides the technical questions, that pointed to manipulation of the tap file, Bakker-Schut and Van der Plas claimed. For instance, a Dutch summary of tapped telephone calls on 1 December 1997 implies that Baybasin phoned the Turkish lawyer Necmettin. The summary states that Necmettin was in Istanbul at the time.[16] The lawyer involved wrote a letter to say that he remembered the call with Baybasin clearly, but he dated the call in the period 1993-1994, when Baybasin was still in Turkey. Necmettin explained that there was no further contact between him and Baybasin after that call. So, who had made a mistake? Was the Turkish lawyer’s memory unreliable? Or could it be that the Dutch summary of the tapped telephone calls in the file was wrong?
We can see in the Baybasin file where Baybasin was on 1 December 1997 by looking at the list of the locations where he stayed during the trial. On the date in 1997 on which the tap file has Baybasin talking on the phone to Necmettin, the list of locations says that Baybasin was in Amsterdam. However, both Necmettin and Baybasin say they were in Istanbul when that call was made.
The solution to this puzzle was provided when an independent interpreter retranslated the call. It then transpired that certain fragments in the first translation, which indicated that Baybasin must have been in Istanbul during the call, were missing. There was no way of avoiding it: the people who compiled the first translation of Baybasin’s telephone call from 1993-1994 wanted to postdate it to 1997.
There are other examples of ‘carelessness’ in the tap file that raise doubts about the correctness of the telephone tap reports or the timing of the telephone calls. It should be noted that the inconsistencies concerned are not technical details that suggest that something might be wrong, but simply a question of logic. What, then, were the technical clues that the tap reports were unreliable?
Netherlands Forensic Institute expert Broeders
The Netherlands Forensic Institute investigated thirty telephone calls in the first investigation into irregularities in the telephone taps (in 2000).[17] The team of three investigators led by A.P.A. Broeders reached a clear main conclusion:
‘The investigation carried out into the nature of the irregularities mentioned by the defence in the investigated recordings does not support the view that these recordings had in any way been manipulated.’
However, Broeders did express a surprising reservation in the first additional remark after this conclusion, stating that his investigation could not rule out manipulation. The investigation method used by Broeders was based on an ‘auditory survey’ of the material, which amounts to listening to the calls through headphones. Broeders stated that this (auditory, or headphone) method was often unable to detect splicing of a tape. Broeders also stated that the original material, in other words the wiretapping room optical disc, should also be investigated.
After the investigation, the Public Prosecution Service consistently took the Netherlands Forensic Institute (Broeders) line to defend against allegations regarding the tap file. The Public Prosecution Service argued every time that there was no need to investigate the optical disc because the Netherlands Forensic Institute had found no signs of manipulation. They therefore ignored Broeders’ comment that the optical disc would have to be investigated in order to be sure. Broeders stuck firmly to his conclusions from 2000 in his later witness statement before the Court of Appeal in 2002 and in his report of the same year, even after a thorough study of Kuylman’s and Van de Ven’s opinions.
It is important to look carefully at Broeders’ reasoning, because both the Breda District Court judgment on the tap reports and the Den Bosch Court of Appeal ruling rely totally and utterly on Broeders’ views and expertise. The Court of Appeal avoided any substantial observation on the anomalies, clicks and other technical inconsistencies. Faced with the two contradictory opinions - the one from Broeders and the one from Van de Ven and co. - the Court of Appeal ultimately opted to give one of them the status of being the truth. The main argument was that the reliability of Broeders and his Netherlands Forensic Laboratory was beyond all doubt, from which it apparently follows that the views of external experts can be completely ignored.
What, then, were Broeders’ reasons for not assuming any manipulation? He responded to Kuylman’s and Van de Ven’s findings in a letter dated 10 June 2002 to the Court of Appeal in Den Bosch. He wrote that the Netherlands Forensic Institute had not set out to demonstrate the possibility of manipulation, but was looking for evidence of actual manipulation. Broeders points out that his very first report says he takes it as read that audio recordings are open to manipulation, in view of the capabilities of digital technology.
The idea underlying Broeders’ reasoning is that it is impossible to establish any manipulation with his auditory (headphone) method without an investigation of the original optical disc. Therefore, without that investigation, no-one can reach a conclusion on any manipulation that might be inaudible through headphones. Broeders later states that his aim in the investigation was to ‘see if any signs of actual manipulation could be found’. He found none. But, according to his own reasoning, that is not enough to conclude that there had been no tampering. To be sure, original material would have to be investigated, but that was not one of his recommendations.
Broeders then dismisses the dozens of anomalies put forward by the defence, saying that all the clicks, other voices, delays and signal interruptions were either perfectly normal or simple to explain. The essence is that Broeders (wearing headphones) simply observes that clicks are ‘normal’ in telephone calls, and do not call for further investigation. A specific example is:
‘There is a click in the middle of a sentence, but the sentence is not interrupted. The click does not support an allegation of manipulation.’
Another example is:
‘The audible clicks do not support an allegation of manipulation of the recording’
Broeders, with scientific pretensions, thus expresses an opinion on mechanical-sounding clicks that occur throughout the recordings. He neither offers any explanation for them nor recommends further investigation. For a forensic scientist, who after all has the task of helping to clarify reality, this attitude is hardly ambitious. In fact he rules that the clicks are ‘normal’ without giving the slightest hint of what might have caused these extraordinary sounds. Broeders does provide explanations for some of the calls, such as for the differences in background noise in call 140. Kuylman and Van de Ven observed considerable (inexplicable) noise level differences in this call. The tone of the conversation and the words used by the speakers tend to reinforce the suggestion that the call in fact consists of two different calls spliced together. After the clicks, both parties immediately call out ‘hello’ and sound as if the connection has just been established.
Telecommunication expert Kuylman investigated this call. He observed that something is wrong in the substantial signal to noise ratio variations in the more than 45 minutes of the call. Kuylman says that there is so little noise in parts of the call that there is actually no question of it being a recording of any kind of intercepted telephone call, and definitely not a GSM call, as the file claims.
Kuylman states: ‘…in view of the sound quality, there are other more obvious explanations, for example that the recording was made directly from a microphone.’ This explanation could fit in with what Baybasin has said about this call from the outset: that it was constructed from one telephone call and from another call to the same person while in a car somewhere in the Netherlands.[18] An expert from the United States (at the request of lawyer Van der Plas) says about this call that it must have been compiled from multiple calls.[19]
After filtering the signal, the call clearly comprises two parts. The first part has rattling noises of a car in the background, whereas the second part takes place in a more hollow space, with voices audible in the background, which are absent in the other part. [20] The American telecommunication expert, Dickey, concurs with this conclusion: ‘… several anomalies associated with [call] 140 are consistent with alteration. Indications of alteration by deletion and/or data assembly are present’. [21] In other words, this was a case of splicing, which is a conclusion that Kuylman and Dickey say can be established without further investigation of the original material. They say that the transition in the frequency pattern (the change in the noise) within a single call can be explained only by manipulation. In response to the opinions of these experts, Broeders’ explanation for the difference in background noise was surprisingly simple. He just suggests that the speaker might have moved to a different room, without addressing any of Kuylman’s questions, in particular whether part of the call could possibly have been a telephone call at all. Broeders suffices by stating that there is no evidence of manipulation in this call, ignoring the observed noise level differences. And with these few simple words from Broeders, the Den Bosch Court of Appeal expressed satisfaction with call 140.
Van de Ven testified that tone pulses that occur in another call are normally impossible in GSM technology. They suggested that the call was made with old analogue technology. Broeders confirms as much in his 2002 letter, but writes later that the tones can be explained if the call had been routed through an analogue (old) exchange. What he did not add is that the call was alleged to have been made in 1997, and the last analogue exchange in the Netherlands had already been switched off in 1995. At any rate, therefore, this call could not have been made in the Netherlands, contrary to what the tap file says. Broeders is otherwise silent on this matter.
There is no doubt that Broeders’ most striking explanation is concerned with the clicks in call 11. Seven clicks are audible in a space of three seconds, fifteen minutes into that call. Broeders says these noises are no sign of manipulation whatsoever. What is more, he sees no reason at all to investigate these noises further. Broeders supports his view with a simple explanation. The clicks were likely a consequence of ‘one of the speakers checking his mobile phone batteries’. One of the speakers (K.) remarks shortly before the clicks that his battery is running out. Broeders states that: ‘a sound is audible that indicates that K. might have shaken his phone.’ Broeders says that the idea behind this would have been to get the last remnants of charge out of the battery.[22] If this explanation were correct, the shaking must have prompted a considerable burst of energy, because the tap reports show that the call continued for a good half an hour afterwards. However, not one specialist in the field of mobile telephony can be found who supports Broeders’ view that shaking a phone can coax the last remnants of charge out of a battery. Apparently none of the judges at the district court and Court of Appeal were aware of this fact. Indeed, they swallowed this explanation wholesale, based on the authority of Broeders and the Netherlands Forensic Institute. The Court of Appeal even wished not to cast any doubt on the expertise of the expert witness from the Netherlands Forensic Institute.
The Court of Appeal appears not to have asked several justifiably critical questions raised by Broeders’ simple explanation.
• Is it possible that shaking a mobile phone battery can cause more charge to flow (as shaking a gas lighter might cause more gas to flow)?
• If so, might this extra charge enable a call to continue for another half hour?
• What kind of mobile phone might make this possible?
• What exactly did Broeders mean by ‘checking the battery’ of a mobile phone?
At the last moment, Baybasin’s lawyers presented the Court of Appeal with a final report from Van de Ven, which offered an entirely different explanation for the clicks. This report showed a spectrum analysis of this call that exhibits clear signs of splicing at the position on the tape where technical rattles (clicks) occur. The frequency graph shows brief silences during the three seconds concerned. As irony would have it, the scientific literature contains an article from 1993 that discusses precisely this kind of click: periods of silence that occur when deleting and overwriting tapes by starting the recorder. The expert that wrote this report is Broeders. He describes in this article signal interruptions of between two hundred and six hundred milliseconds when overwriting tapes.[23] This is exactly what Van de Ven observed in the frequency pattern of the disputed call.
It is astonishing that Broeders did not recognize this same pattern when studying the material, and did not confirm that it was identical to the example in his 1993 article, and more astonishing that the Baybasin report lacks any reference to his earlier study. After all, it would seem obvious at least to examine the correctness of the hypothesis that Baybasin’s calls exhibited the same characteristics as the pattern he described in 1993.[24]
Broeders’ expertise
Baybasin’s lawyers put forward the argument that Broeders of the Netherlands Forensic Institute lacked expertise. He had based his conclusions only on listening to the material through headphones, while, they argued, in signal analysis practice, this ‘auditory’ investigation method is no more than a starting point for further investigation: further investigation that goes beyond analysis limited by the human ear. Broeders vigorously disputed this argument at the trial at the Den Bosch Court of Appeal. In his written response to Kuylman’s and Van de Ven’s observations to the Court of Appeal, he attaches a long list of publications that testify to his expertise. For the sake of clarity: Broeders is a linguist who approaches voice recognition and speech analysis from linguistic principles (Van de Ven is a mathematician and physicist). In a prickly response to Van der Plas’ assertion that he was not an expert in the field of digital signal analysis, Broeders even warned that he would start legal proceedings against Van der Plas. Broeders demanded that Van der Plas withdraw her assertions before the Court of Appeal.
Without directly questioning the Den Bosch Court of Appeal ruling, an interested and impartial outsider is nonetheless left with a number of questions. In particular, the question of how all the strange phenomena observed by Kuylman and Van de Ven can be explained. The discussion about Broeders’, Kuylman’s, or Van de Ven’s expertise is irrelevant.
The point is how to eliminate the unavoidable suggestion that multiple calls had been spliced together. It would at any rate appear reasonable to investigate the anomalies further. The Den Bosch Court of Appeal missed an opportunity in this regard in the last instance by deciding not to order an investigation of this kind, despite serious urging from experts called by the defence. An important conclusion of the Court of Appeal was that manipulation of the phone calls could not be ruled out.[25] Nonetheless, the Court of Appeal bases its opinion solely on the expert evidence of Broeders, who found no sign of manipulation, and not on the evidence of the other experts. The opinions of the other experts were ignored by the Court of Appeal, and the ruling does not even mention their contrary opinion. The crux of the Court of Appeal’s opinion is the authoritarian argument that the Netherlands Forensic Institute’s reliability is inviolable and that there is therefore no reason to doubt Broeders’ statement. It is then logical that sentencing followed on the basis of the intercepted calls with no need for further substantial investigation of the anomalies in the taps.
With hindsight, this was probably a mistake. The results of Van de Ven’s latest investigation of the calls inevitably suggest that the tap files in the Baybasin case cannot be consistent with the truth. This latest investigation became available only in 2004, which was too late for the appeal, and also too late for the Supreme Court, which upheld the ruling in the Baybasin case.
impossible calls
Following the Supreme Court ruling, Van der Plas and Bakker-Schut were obliged to find other, more persuasive, ways of demonstrating the manipulation than they had done before the Den Bosch Court of Appeal. They requested in interim injunction proceedings at the District Court in The Hague that the Dutch state be ordered to submit the original optical discs with the phone calls for further investigation. But they lost this case, too. They also attempted in civil proceedings, also through the Court of Appeal in The Hague, to get their hands on the original optical discs. The Court of Appeal in The Hague ruled, among other things, that Baybasin had been irrevocably sentenced by the criminal court and could not seek redress through civil law. The arguments that the investigation of the optical discs might yield new facts on which basis the Supreme Court might refer the matter back to a Court of Appeal, were also unsuccessful. The Court of Appeal ruled that the state cannot be compelled to cooperate in a request for a review of this kind.
Van der Plas’ and Baker Schut’s last chance was the appeal submitted in 2004 to the European Court of Human Rights. They put forward at that appeal new investigations by Van de Ven into the audio cassettes, using the latest signal analysis techniques. However, the European Court of Appeal turned a blind eye to that new investigation and upheld the ruling of the Dutch courts. The European judges also created an impression of wishing to avoid the core of the problem. The calls themselves and the latest investigation were not examined. The European Court of Appeal ruling was shocking in view of the result of that investigation.
Van de Ven’s conclusions suggest almost unavoidably that a series of investigated calls could never have taken place as they are presented in the file. This investigation is close to being incontrovertible evidence that the calls were indeed manipulated (although here too, scientific certainty demands studying the original optical discs). The phenomena in the calls cannot be explained other than by manipulation. In summary there are three things that do not fit:
• some calls cannot have been recorded in the Netherlands;
• some calls could not have been tapped from a telephone line (which means that the conversations were bugged ‘directly’ using a microphone);
• frequency analysis of the noise demonstrates dozens of moments in calls where a new GSM call starts. Splicing moments have thus been visibly demonstrated.
Van de Ven starts his report by addressing the relationship between his investigation and the Netherlands Forensic Institute investigation conducted by Broeders for the Den Bosch Court of Appeal. The Court of Appeal based its opinion that manipulation was not demonstrated on the basis of this report. Broeders produced his report on the basis of auditory analysis, or, in other words, careful listening. Van de Ven says that Broeders’ investigation method cannot support any statement on the strange audible phenomena on tape. Van de Ven states that an auditory survey of the audio material is a good basis for assessing whether the tapes contain unusual phenomena, after which signal analysis would form the basis of a statement on whether manipulation actually took place.
Van de Ven used computer programs that enable statements to be made on the audible telecommunication signals. In an investigation of this kind he looks mainly at the signal properties of the various tones and how they are prescribed in international protocols, for example.[26]
Van de Ven’s conclusions are as follows.
1. An AGC pattern occurs in five calls in the middle of what the file says is a continuous call.[27] This is a pattern in the frequencies of the signal that always occurs shortly before a GSM call starts, appearing as noise just before the communication path is established. If this AGC pattern is perceptible in the middle of a ‘continuous’ call, it can mean only that a new call started. This observation therefore establishes that two calls were spliced together. It is technically impossible for the recording to have been of a single call.
2. A 450Hz or 400Hz dial tone is audible in eleven calls. The dial tone in the Netherlands is 425Hz, which means that it is impossible for these calls to have been recorded in the Netherlands as the file claims. Frequencies of 450Hz and 400Hz are in use in Turkey and Britain, respectively.
3. It has been established for fourteen of the calls that they could not have been telephone calls at all, but were intercepted directly, for example with a microphone in a room. The calls have a frequency range extending from below 300Hz to over 3400Hz, which means they are outside the telephony band, which cuts off high and low frequencies. The calls cannot therefore be from any telephone call recorded in the wiretapping room, although they are registered as such in the file.
4. The signal-to-noise ratio in two calls is so good that they cannot possibly be GSM calls, as the file states, but are calls over a land line.
5. One call contains crosstalk. This phenomenon must have been caused by an analogue exchange. As of 1995, all analogue exchanges in the Netherlands had been replaced by digital ones. Contrary to what the file claims, this call could therefore never have been tapped from the Dutch telephone network in November 1997. Tone pulses can be heard in another call that could only have arrived at the wiretapping room from an analogue exchange. Therefore, neither could this call have been recorded in the Netherlands in 1997.
6. The noise drops out completely at several times in various calls, resembling the noise level of a good blank audio tape. These are strong indications of splicing moments, but this cannot be proved directly because the original audio carrier cannot be investigated.
7. The seven clicks in the call, which Netherlands Forensic Institute investigator Broeders attributed to shaking the phone, could not have been caused by shaking. Apart from the question of whether it can generate new electricity, the explanation fails because ‘shaking’ the phone would create a random signal, consisting of multiple frequencies, whereas the signal of the clicks has a systematic pattern. The clicks are paired frequencies, as are produced by pressing the various numbers on the keyboard.[28] Furthermore, an AGC pattern occurs in the middle of this call close to the clicks, which irrefutably indicates the establishment of a new GSM connection. The only possible conclusion is that ‘the call with the shaken phone’ is a composition of two calls.
Therefore, a fraction (140) of the over 6000 calls contain many inconsistencies that justify the conclusion that some calls were not tapped from the telephone network in the Netherlands. The dial tone frequency implies that some calls must have been recorded in Britain and Turkey. After years of discussion, the presence of splicing in the calls has now been established. The evidence on which a suspect was sentenced to life imprisonment in a Dutch court was deliberately tampered with. These conclusions must be fairly unsettling for Dutch criminal procedure, because the main evidence in a long drawn-out and high-profile prosecution was apparently unreliable, while the defence and journalists had been pointing out the necessity of further investigation for years. And no court considered it worth bothering to investigate precisely what the truth of all those claims might be. Even the European Court of Human Rights set aside Baybasin’s appeal with little motivation, in spite of new investigation material put forward.
Whether Baybasin may have been incorrectly, or entirely innocently, sentenced to life as a result of the far from thorough approach of these courts, is a still nagging question. How many other, uninvestigated, calls in the Baybasin case will actually exhibit the same inconsistencies? How many other ‘impossible calls’ are among the 6000 in the Baybasin file? And: if the judiciary’s monitoring of this much-used investigation technique can fail in this way, what other cases have involved tampering with tap material? But the most serious question regards the extent to which Dutch investigating officers or Dutch Public Prosecution Service staff were actively involved in tampering with Baybasin’s tap file.
‘to make him cold’
It is particularly shocking that Dutch courts have been duped by tampering with the tap file while it is very likely that the actual calls were also manipulated. But even more shocking is that in one case at least a Dutch-speaking person must have taken part in the manipulation. Was that an employee of the police or of the judicial authorities? Who was responsible and knew what was going on? It is unlikely that this will ever be investigated by a Dutch court, which makes the matter even more serious.
The call concerned is number 38, which was crucial evidence against Baybasin of attempted incitement to liquidation right up to the public prosecutor’s demand at the Court of Appeal hearing. Advocate General Kolkert even referred to it in his public prosecutor’s demand as ‘the cornerstone of the evidence’ of a liquidation contract in the United States, which in fact was never carried out. Call 38, according to the Dutch translation in the file, contains the words ‘hem koud maken, is dat alles?’, which mean ‘to take him out, is that all?’, which Baybasin answered affirmatively in that call, as the alleged client.
The original call was conducted in English by two people who do not speak Dutch, Baybasin and an unknown person. The actual English words of the unknown person’s question were ‘to make him cold, that’s all?’, at least as they are reported in the file. But experts in the English language had earlier testified to the Den Bosch Court of Appeal that the words were: ‘and all is…to make him call that’s all’. These experts also stated that the expression ‘to make him cold’ does not mean the same as ‘to take him out’ in any variant of English. Advocate General Kolkert would have been in a position to be aware of this, but, surprisingly enough, he insisted in his public prosecutor’s demand that ‘to make him cold’ means ‘to eliminate someone, or have them eliminated’. As he was speaking, Van der Plas interrupted him and drew the Court of Appeal’s attention to the fact that expert witnesses had already stated that ‘to make him cold’ did not have that meaning in English, unlike Dutch. This also went a little too far for the Court of Appeal, and Kolkert was obliged to withdraw that element from his demand. Even without this ‘cornerstone of the evidence’ the Court of Appeal found Baybasin guilty of attempted liquidation. But, with hindsight, an interesting question remains unanswered. Who came up with the idea that the words ‘to make him cold’ were on the tape? Who had embellished the report? And even more interesting: who decided that ‘koudmaken’ (to take him out) was a good translation of the passage into Dutch? At any rate, it must have been someone with a thorough knowledge of Dutch, which would suggest an interpreter, or a Dutch police officer.
The latest signal analysis by Van de Ven from 2004 makes the entire matter even worse and the questions more pressing. Because in a space of 25 seconds leading up to the disputed passage (‘and all is, to make him call that’s all?’) the signal drops out three separate times, which indicates splicing. There are a total of fourteen events of this kind in this call. If this was meant to palm Baybasin off with attempted liquidation, it was a clumsy way of doing so. Nonetheless, it almost worked. Without the investigation that Bakker-Schut and Van der Plas commissioned as a last resort, it would never have been noticed. And perhaps the Advocate General, possibly completely in good faith, albeit with limited knowledge of English, could have freely used the passage as the ‘cornerstone of the evidence’ before the Court of Appeal. If splicing took place, and that is what appears to have happened, whoever did so had a good knowledge of Dutch, because he knew the expression meaning ‘to take someone out’, and limited knowledge of English, because he did not realize that the literal translation ‘to make him cold’ was not an English idiom.
do the optical discs actually exist?
All the questions raised suggest even more new questions. For a start, there is the question of why the Public Prosecution Service never once saw reason to disprove the serious accusations of manipulation that were expressed in court and in the media. Kuylman and Van de Ven repeatedly offered to make their expertise available to the Netherlands Forensic Institute with a view to jointly investigating all doubts. Why was the matter not investigated thoroughly, and the optical discs made available for scrutiny? The Public Prosecution Service could then have triumphantly derided the legal profession and critics with an investigation that swept away all allegations! It did not happen. Under pressure from the defence, a start was made on clarifying vague areas, but a thorough investigation did not take place. And that raises even more questions, and also the suggestion that something is seriously wrong.
A following obvious question is whether in the eyes of the Public Prosecution Service there might have been an overwhelming reason for not investigating the matter thoroughly. Could this reason have been that the Public Prosecution Service was embarrassed by the negative attention being given to the telephone taps? That would be a serious error of judgment, because the investigation techniques used by the Dutch Public Prosecution Service have to be beyond reproach, affecting, as they do, the legitimacy of criminal law. Or was the reason that the Public Prosecution Service knew about the faults in the telephone tapping system in the Netherlands? Or maybe it was aware of deliberate manipulation in the Baybasin case? If that were proved, it would give rise to a crisis in the world of investigation comparable with the ‘IRT affair’. It would sink a hatchet into the legitimacy of the Dutch legal system, because if judicial authorities fail to observe the rules of the game, why should the public do so? The investigation indicates that certain calls were probably recorded in England or Turkey, and in any case not in the Netherlands. It is therefore fairly improbable that these calls have ever resided within a Dutch wiretapping system, and must have been imported from abroad. Did these calls actually ever end up on the wiretapping room optical disc? And, if so, it would appear that the system in the wiretapping room is insecure, because someone was able to tamper illegally with a tap file of a Dutch criminal investigation. If not, and the calls were stored on a tape or a CD, then was that the reason why the Public Prosecution Service was unwilling to release the optical disc. Because if it were to do so, and the calls concerned were not present, the Public Prosecution Service would be providing direct proof that the Dutch tap system is unsound, which would have enormous repercussions for all major criminal cases involving telephone taps - if other courts of justice have the courage genuinely to seek to establish the truth.
Baybasin now occupies the days of his life sentence with painting and other activities. He pins his hopes on the possible emergence of new facts surrounding his investigation that might support a request for a review at the Supreme Court. These proceedings could give rise to the case being referred back to the Court of Appeal. A long road, but it is the last straw that the Dutch legal system offers him.
[1] The broadcast (6-2-2003) can be seen on Internet on www.zembla.tv
[2] The statement of H. Topcu, in Baybasin’s criminal file
[3] The file says that all calls were recorded on optical discs in the Apeldoorn digital wiretapping room of the National Police Services Agency.
[4] An appeal to the Dutch Data Protection Authority was blocked by the Board of Procurators General.
[5] This team now operates partly under the National Investigation Service.
[6] Interview with a former employee of the North East Netherlands Interregional Crime Squad, September 2004
[7] See Hans Nelen, Els Barendse, Jaqueline van der Schaaf: Mondiaal met man en macht, WODC July 1998). This document discusses the course of events in the 4M case.
[8] List of attendees at the Rotterdam conference
[9] Statement by Schalks 8 June 1999 to the examining magistrate, 97/1056
[10] IND memo, Baybasin criminal file
[11] Statement by H. Topcu, Baybasin file
[12] The Court of Appeal completely ignored Topcu’s statement.
[13] Statement by Anton Schalks to the examining magistrate 8 June 1999 97/1056
[14] Some of the translations by the police interpreters were checked by a Kurdish language and literature professor, Van Bruinesse. Several of the translations turned out to be completely unsatisfactory.
[15] See the reports of Kuylman, Van de Ven and Dickey
[16] Call 140. The call numbers refer to the labels on the cassette tapes made available to Baybasin.
[17] Baybasin’s lawyer flagged 121 calls as suspect.
[18] Report M. Kuylman, dated 16 July 2002
[19] Report B.G. Dickey, dated 11 April 2003
[20] Report J.W.M. Van de Ven, dated 5 June 2002
[21] Report B.G. Dickey, dated 11 April 2003
[22] Report Broeders, 22 December 2000
[23] Broeders Report, 1993, Het Forensisch Onderzoek van Audio-opnamen (The Forensic Investigation of Audio Recordings), in Modus (NFI) no.1
[24] Report J.W.M. Van de Ven, 15 July 2002
[25] Den Bosch Court of Appeal ruling, July 2002
[26] Report Van de Ven 27-6-2004
[27] Automatic Gain Control (AGC)
[28] See Chapter 3 for an explanation of the frequencies (DTMF tones) produced by pressing keys.
