Trojan Horse: When 'facts' are contested, who decides the truth?
In December last year the British think tank Policy Exchange released a long report which claimed to expose “a concerted campaign to rewrite history around the ‘Trojan Horse’ scandal, where hard-line activists Islamised state schools in Birmingham”.
Policy Exchange presented the report as a necessary response to the podcast released by the New York Times just over a year ago which sought to re-examine the “Trojan Horse” story.
The podcast described how the letter at the centre of the scandal was initially dismissed as a hoax by officials and police. There was no attempt to identify the author of the letter, but nonetheless it prompted a number of inquiries into the allegations it made of an Islamist plot - with devastating consequences for students, teachers and governors at the schools under scrutiny.
The most significant of these was ordered by the then-education secretary Michael Gove in 2014 after details of the letter were leaked to the media. Gove appointed a former counter-terrorism police chief, Peter Clarke, to head the inquiry, which the podcast examined at length.
Gove, the podcast said, was told by officials that counter-terrorism police had decided the letter was a hoax, but “used the letter to sanction numerous high-level investigations into potential extremism in Birmingham schools anyway”.
New MEE newsletter: Jerusalem Dispatch
Sign up to get the latest insights and analysis on Israel-Palestine, alongside Turkey Unpacked and other MEE newsletters
The Policy Exchange report is co-authored by Damon L Perry, a senior research fellow at Policy Exchange and an associate fellow at King’s College London’s International Centre for the Study of Radicalisation (ICSR), and Paul Stott, Policy Exchange’s head of security and extremism.
But Gove himself provides a heavyweight endorsement in a foreword co-authored with Nick Timothy, a former adviser to Theresa May who was home secretary at the time the Trojan Horse story was making headlines. They take issue with the New York Times' podcast, describing it as a “travesty… replete with errors and omissions”.
Gove remains in government as a minister for housing and communities and has been an influential figure in the Conservative Party for over two decades. He was the founding chairman of Policy Exchange when it was set up in 2002. Timothy is now a senior fellow at the think tank.
Describing themselves as having been “at the heart of the government’s response”, they suggest the podcast gave “fresh wind” to a “campaign of grievance” by Islamist activists seeking to undermine the government’s counter-extremism work and the Prevent strategy.
The “facts” of the Trojan Horse affair had been established, they wrote, by Clarke – now also a Policy Exchange senior fellow - and by other inquiries commissioned at the time by Ofsted, the schools inspectorate, the Education Funding Agency and Birmingham City Council.
The implication is of a number of independent inquiries arriving at similar conclusions. However, Ofsted and the EFA are both agencies of the Department for Education and one of the inspectors from the EFA inquiries went on to act as an adviser to Clarke, while Ian Kershaw, who conducted the inquiry for Birmingham City Council, liaised with Clarke.
Gove and Timothy wrote: “Those facts are not open to debate, for they are facts. It should not be necessary to reassert what happened in a new report like this, but thanks to campaigners and useful idiots in publications like The New York Times, it sadly is.”
The New York Times is not the only target of the Policy Exchange report.
The authors of this article are named in the title of a chapter of the report which characterises their work on this issue as a “defence” of Tahir Alam and the Park View Educational Trust.
Alam was the Park View school governor and chairman of the trust of three schools who was banned from managing schools by the Department for Education as a consequence of the Trojan Horse inquiries.
Policy Exchange describes its own report as “the first comprehensive documentation of the key events in Birmingham”, yet it examines at length John Holmwood’s 250-page academic study, co-authored with Therese O’Toole, which was published with a full chronology in December 2017.
Middle East Eye is cited several times among a number of “Islamic and left leaning sites” which the report said had given the New York Times’ podcast positive coverage.
Our central purpose however is not to address criticism of our own work, although we will refer to this again later in this article.
The purpose of the article is instead to challenge serious errors of fact and interpretation which bring into question Gove and Timothy’s assertion that the Policy Exchange report should be considered “the final word”.
The need to challenge this claim, and the authority of Peter Clarke's flawed report, has grown more urgent with the publication this month of the Independent Review of Prevent in which reviewer William Shawcross calls for the controversial counter-extremism strategy to be focused more sharply on countering "non-violent Islamist extremism".
Citing Clarke, Shawcross referred to "revelations that a group of individuals had sought to gain undue Islamist influence over several schools in Birmingham" - framing the episode alongside the murder of British soldier Lee Rigby in Woolwich in 2013 and British nationals travelling to Syria to fight with jihadist groups in the country's civil war - as indicative of an "intensification" of a threat to the UK from Islamist terrorism.
Clarke was also among signatories of a letter published by the Times newspaper on Friday which welcomed the Shawcross review and urged the government to "get on the front foot in the fight against terrorism and extremism".
'The UK's most influential think tank'
Policy Exchange has its own stake in preserving the myth of a Trojan Horse plot. As described above, several of those who were involved in shaping the official narrative have close links to the think tank.
MEE has previously reported on its outsized influence on British state policy, under both Labour and Conservative governments, towards Muslim communities.
It describes itself as “the UK’s most influential think tank” and claims credit on its website for the shift under David Cameron’s government towards a greater focus on countering extremism and the rewriting of the UK’s Contest counter-terrorism strategy and the Prevent programme within it.
In education, it links its report in 2010, which concluded that faith schools were vulnerable to extremist influences, to the creation of a unit – the Due Diligence and Counter Extremism Group – within the Department for Education “to prevent the teaching of extremist religious views in schools”. This unit would later play a key role in shaping the department’s response to the Trojan Horse allegations.
Gove too has been influential in embedding a neoconservative worldview hostile to a perceived Islamist threat at the heart of government.
He had used the term “Trojan Horse” in a title chapter in his book on the theme, Celsius 7/7, published in the aftermath of the deadly al-Qaeda attacks in London in 2005, which warned of the dangers of grassroots Islamism.
We now turn to the many errors, omissions and flaws of reasoning contained in Policy Exchange’s report, beginning with Gove and Timothy’s foreword in which they refer to “the takeover of secular, state schools”.
While there are faith-designated schools, there is no such thing as a secular state school in England. All publicly funded schools, including academies and free schools, have a statutory duty to provide a daily act of collective worship and provide compulsory religious education.
In most schools this worship will be “of a broadly Christian character”, but schools with a majority non-Christian intake (more than 90 percent of students at the so-called Trojan Horse schools were from a Muslim background; at Park View it was 98.9 percent) can apply to the appropriate authority to have a different form of worship.
Two of the schools at the centre of the affair, Park View and Oldknow, had had determinations for Islamic collective worship for over a decade.
The obligation of schools to provide a daily act of worship is itself contentious. In December last year, the Department for Education defended the principle after teachers called for it to be scrapped in light of census results showing that less than half of the population of England and Wales identified as Christian.
A department spokesperson said: “Collective worship encourages pupils to reflect on the concept of belief and the role it plays in society. Schools are able to tailor their provision to suit their pupils’ needs."
Gove and Timothy’s failure to understand this crucial point has important consequences. Together with Clarke, they call schools that are not formally designated as faith schools “secular”. This means any form of Muslim collective worship can be presented as an Islamist attempt to impose “undue religious influence".
For example, their foreword argues that “broadcasting the call to prayer across the playground” is motivated by “a hardline and politicised strand of Sunni Islam”.
But in a Muslim-majority school the call to prayer is comparable to the sound of church bells indicating service in chapel familiar to many government ministers from their own school days.
Indeed a 2012 Ofsted report on Park View, which judged the school to be outstanding, had specifically commended the “wide range of opportunities for spiritual development, for example, through the well attended voluntary Friday prayers meeting".
This mistake is especially disturbing coming from a former education secretary. Gove, in fact, had used the same language – describing the Birmingham schools as “secular" – in a statement to Parliament in June 2014.
Troublingly, the Peter Clarke report – which the Policy Exchange report insists on treating as documented, unassailable fact – made the same error.
Clarke described “a deliberate attempt to convert secular state schools into exclusive faith schools in all but name".
In the case of Clarke, a policeman rather than an educationalist, this kind of mistake, though grievous, is perhaps understandable. But it led to unfortunate consequences for the Trojan Horse schools: Clarke interpreted religious activity as evidence of a potential problem.
His lack of familiarity with the education system also led him to more serious errors. Clarke failed to mention that the Park View Educational Trust, which ran the school at the centre of the accusations, took over other schools at the suggestion of the Department for Education.
Under the academies programme overseen directly by the department, a school deemed to be failing was required to leave local authority responsibility and "convert" to become an academy within a multi-academy trust run by a successful "sponsor" school.
If there was a Trojan Horse it was, at the very least, escorted into place by Gove and his department. Policy Exchange fails to address this issue. It describes the “plot” as involving the undermining of successful head teachers - but it was the schools that supposedly enacted the takeover that had been recognised as successful.
Moreover, any "takeover" – more properly the incorporation of a school into an academy trust – necessarily involves minutes and memos from officials at the Department for Education, since all such proposals must be approved by the secretary of state.
These are strikingly absent from any of the official reports and from the Policy Exchange report, though some of them did come to be disclosed as part of the misconduct case brought against senior leaders at Park View Education Trust.
'Patterns' - but no proof
A further problem with the Clarke report was the decision to accept anonymous statements from witnesses. This was potentially against natural justice because it meant that those criticised in the report were unable to respond to allegations against them.
Many of these allegations were carried across the different reports by Ofsted and the Education Funding Agency and also leaked to the press.
Clarke’s approach was to present these allegations as suggestive of “a pattern”. He acknowledged that Park View School “disputed most, if not all,” of the allegations and that “specific patterns of behaviour observed were not present in all of the schools”.
This approach meant that none of the allegations made by Clarke would be tested until they reached a court of law. They were hearsay at best, published by an investigator with an inadequate understanding of how schools work, and in particular their interface with government and religion.
These allegations, then, are not - as Gove and Timothy describe them – “facts”. Still less are they “facts that are not open to debate”. One of the merits of the New York Times’s investigation was that it demonstrated the unsubstantiated nature of this hearsay evidence.
Take the Policy Exchange account of one of the most notorious examples of alleged extremism at Park View school: the claim that a jihadist video was shown to pupils.
Accounts of this episode in documents and media stories at the time were muddled and vague.
Policy Exchange relies on oral evidence provided by Kershaw - who had carried out Birmingham City Council's investigation - during his appearance alongside Clarke as witnesses at a parliamentary inquiry into “Extremism in schools” in September 2014.
Policy Exchange reports that Kershaw “told the Education Select Committee that a violent extremist video that was ‘completely unacceptable’ was shown to pupils in a Birmingham state school. Asked by MPs whether this was a type of ‘violent jihadist promotional video’, he indicated it was".
Clarke had been more circumspect in his evidence, telling the committee: “There was some suggestion that that sort of film had been shown or copied by a technician within one of the schools…”
In his report, Clarke was more cautious still, describing an allegation (among a list which he acknowledges is disputed) of “IT technicians recording what appeared to be al-Qaeda terrorist videos into a DVD format” – but makes no mention of a video being shown to pupils.
In response to queries for this article, Kershaw said the incident he described to the select committee was not the same one referred to by Clarke in his report.
Kershaw told MEE: “The example I gave was of a single incident of a teacher showing a violent extremist video to students in a single lesson which was not addressed by a senior member of staff who witnessed it.”
He said evidence relating to this incident was not available to Clarke and that none of the evidence he had collected was passed to anyone beyond Birmingham City Council.
Kershaw also said he was not involved in any of the misconduct cases brought against staff at the Birmingham schools by the National College for Teaching and Leadership (NCTL), a government agency then responsible for investigating and prosecuting allegations of misconduct against teachers.
In fact, when these cases went before an NCTL panel no evidence of any alleged video incident was presented by lawyers for the Department for Education.
MEE has also seen an internal Department for Education memo in which officials say they can find no proof to back up the video allegation.
The failure by Policy Exchange to mention that the video was not pursued as one of the examples of concern in the misconduct hearings means that readers of Policy Exchange’s report are misled.
In another problematic example, the report talks about “sex segregation in subjects such as maths at Park View, and in some cases in primary schools, as reported by Clarke”.
But, as with the video episode, the evidence of sex segregation in classrooms, as reported by Clarke and other investigators, is far from clear cut.
Clarke had relied in part on the previous investigation into the Park View Educational Trust schools by the Education Funding Agency (EFA).
According to this report, there had been gender segregation in some classes at two of the trust's schools.
“Some classes observed had boys sitting at the front of the class and girls around the edges. There were also lessons where boys and girls sat at the same table but on opposite sides. We also observed lessons where boys and girls were mixed. Boys and girls are taught separately for PE lessons, which is not unusual," the report said.
It noted too that boys and girls were taught separately at Park View school in lessons such as PSHE [personal, social, health and economic education] and RE [religious education], and said teachers had been "evasive" when asked if this was the case in all year groups.
Some of these arrangements were cited by the EFA as examples of the schools' possible non-compliance with equality legislation.
Segregating boys and girls in some classes "could constitute less favourable treatment of girls", it said, as well as calling for the trust to "ensure that any inappropriate gender segregation is eliminated".
Yet the practices described in the EFA report, of boys and girls sitting at separate tables within the classroom, are common in many schools. Its argument that such classroom configurations may have disadvantaged girls is undermined by the fact that girls were outperforming boys in the schools under investigation.
Once again, none of these concerns were ultimately raised in the misconduct cases brought against the teachers.
Hearsay and news stories
Elsewhere, the Policy Exchange report describes a head teacher who it says was alleged to have put pupils in “stress positions” as a form of punishment.
But Policy Exchange gets this story wrong. It relies on a BBC news story as a source, yet the story reports that senior teachers were accused of allowing punishments to happen – not of instigating them as Policy Exchange says.
The teacher who had given the punishment was disciplined and given a final warning by the head teacher, but was not among teachers later accused of misconduct. None of this relevant information is cited in Policy Exchange’s report.
This is a problem that recurs elsewhere. Policy Exchange’s chronology mostly relies on news reports, which tend to represent a partial and sometimes hostile view.
Policy Exchange repeats a widespread error in media reporting at the time by describing Hardip Begol as “director of assessment, curriculum, qualifications and accountability at the DfE”.
This is surprising since Begol was head of the Department for Education’s Due Diligence and Counter Extremism Group - the very body that Policy Exchange says was set up as a consequence of its 2010 faith schools report.
Policy Exchange doesn’t just get Begol’s job title wrong. More importantly, it fails to report his role in seeking to modify judgements of misconduct by an NCTL panel – later overturned on appeal - against two teachers in 2016.
The judgements initially stated that the allegations against the pair were “in no way concerned with extremism”.
But this wording was amended to insert the word “violent” before extremism, following an intervention in the case by Begol, as the High Court judge who heard the appeal case noted in his judgement in favour of the teachers.
Again and again, Policy Exchange’s report resorts to the words “alleged” or “allegedly” - terms whose frequent use in a report which claims to be a “documentary record” suggest a lack of authority.
Like Clarke, Policy Exchange relies heavily on hearsay.
To take one example, it quotes from a media statement issued by the Muslim Women’s Network UK (MWNUK) in July 2014 describing an incident at a school in which “a male member of staff allegedly hacked into a girl’s mobile phone and informed her parents about its contents”.
A version of this story was first reported in May 2014 by Andrew Gilligan – now also a senior fellow at Policy Exchange – in the Telegraph newspaper.
According to Gilligan’s account, a male teacher at Park View School confiscated the phone of a 16-year-old student,
But while the story was reported in the media, no reference to it appeared in a subsequent Education Funding Agency report on the school, nor was it mentioned by either Clarke or Kershaw or in any misconduct cases.
Policy Exchange repeats this story without including this relevant information.
It also fails to note that safeguarding protocols mean that the school could not answer such allegations or what it understood to be the circumstances of the case, which it had correctly referred to the local authority safeguarding officers
Further important context is missing from the Policy Exchange report’s repeated references to the teaching of creationism at Park View schools. To substantiate this assertion, Policy Exchange links to a paragraph of the Clarke report headlined “Teaching belief as fact."
The paragraph reports: “Staff have said that creationism has been taught as fact in science lessons and in assemblies at Park View School. A member of staff at Park View reported that pupils had said: ‘I’m made of clay [...] There is no evolution. I’m made of clay because that is what Mr Hussain [the acting headteacher] told us in assemblies’.”
What the report does not say however is that Clarke’s claims about creationism were rebutted by defence lawyers at the misconduct hearings when it emerged that statements made in an assembly, such as “we are all made of clay”, were used to provide arguments against racism. There is no evidence they were made in a science class.
Why the misconduct cases collapsed
This leads us to one of the most important failings of the Policy Exchange report, namely its handling of the collapse of the misconduct cases against teachers.
The hearings brought by the NCTL were the first opportunity for teachers at the schools to challenge the allegations against them.
But, according to Policy Exchange, the case against them “collapsed due to a legal technicality”.
The hearings, it said, had been “thrown out” by the NCTL panel “after the DfE’s lawyers failed to disclose evidence requested by the defendants’ lawyers. The evidence they sought was the testimonies of the witnesses interviewed by Peter Clarke… which were provided on the condition of anonymity.”
This account leaves out relevant information. The lawyers had not simply sought the testimonies of Clarke’s anonymous witnesses, but material that had been in their possession at the time of preparing their case.
In fact, the disclosure associated with these testimonies had taken place in December 2016. This included but was not limited to material from Clarke. The case finally collapsed in May 2017 after the prosecuting lawyers handed a note to the panel stating they had not properly informed it about the material they had used in the construction of witness statements.
This matters because the case did not collapse on a “technicality” but because the panel had been misled systematically over a number of months.
It was not only that Clarke failed to corroborate witness statements, he also failed to consider them against the relevant guidelines from the Department for Education.
These statements to Clarke were then taken by lawyers acting for the Department for Education in the prosecution of the misconduct cases and re-drafted as witness statements which were claimed to be independent of what was provided to Clarke.
In the process – as in the case of the supposed teaching of creationism – witnesses had their own evidence claims misrepresented and were easily challenged on it as one of the defending barristers explained.
Policy Exchange describes the episode as involving a failure to disclose material to the defence, but omits the scathing conclusion of the panel that in the particular circumstances relating to this case, there has been "an abuse of the process which is of such seriousness that it offends the Panel's sense of justice and propriety”.
Adding to the confusion, the Policy Exchange report consistently misrepresents religious guidelines in British schools – an issue of overwhelming importance because religion lies at the heart of the entire Trojan Horse controversy.
For example, when discussing Tahir Alam’s approach to religion in schools, as articulated by him in a guidance document he co-authored for the Muslim Council of Britain in 2007, they accuse him of “obscuring Muslim diversity and encouraging the perception of Muslims as a distinct and homogeneous group deserving of unique rights in the public space".
This is wrong. The Equalities Act 2010 gives specific rights to religion as a protected characteristic. This is not unique to Muslims, but applies to all religious beliefs (and none). Schools are required to take the views of parents and local bodies representing religious communities into account in developing sensitive areas of the curriculum.
This failure to understand religious guidelines for schools leads Policy Exchange to evade an essential issue: what are the accommodations due to all religious beliefs, and how should they apply to the variety of Muslim beliefs?
For example, they refer to gender segregation in physical education lessons and in sex and relationship education as being potentially problematic, but Department for Education guidelines at the time clearly allow this, and many schools follow these practices (and not just those catering for the religious views of parents and children).
The Policy Exchange report also misrepresents the Prevent duty, under which providers
The report devotes a chapter to challenging the view that “the Trojan Horse affair was the pretext used by an Islamophobic government to justify the introduction of the Prevent duty”.
It said: “This assertion was put forward both in the recent New York Times/Serial podcast and by various activist media outlets. The Government has neither responded to nor challenged this completely erroneous charge.”
The charge is not erroneous, as can be demonstrated by examining the 2015 Counter Extremism Strategy, which established the Prevent duty. It cited the risks of “entryism”, and gave the Trojan Horse affair as its main example. Policy Exchange makes no reference to this document.
Yet Policy Exchange’s argument that “schools were identified as an area of concern prior to Trojan Horse”, which cites a report by an extremism task force set up by David Cameron after the British soldier Lee Rigby's murder in 2013, does not tell the whole story.
In its first iteration until 2011, the Prevent strategy had been largely directed at “integration” and, in fact, had been incorporated into the duty of schools to “promote community cohesion” in 2007. This was subject to Ofsted inspection.
A report for the Department for Education in 2011 set out how most schools addressed Prevent through activities associated with community cohesion.
However, this duty to promote community cohesion was removed from Ofsted’s inspection remit in 2012 under what Gove called a new, more rigorous regime. It seems he was unaware of its relation to Prevent.
In 2014, following the Trojan Horse affair, the duty to promote community cohesion was itself replaced by a new duty to promote fundamental British values.
It is interesting that Clarke (and Policy Exchange) fails to discuss community cohesion as a duty on schools and its relation to Prevent, except to suggest that concerns about community cohesion were a reason why the local authority failed to act on complaints against some governors and teachers.
Fallacy of composition
Policy Exchange’s flawed presentation of Prevent derives in part from a wider methodological problem that crops up in its report: the fallacy of composition.
This involves identifying a group of authors and critics and characterising the whole group by the views of some of their number. For instance, Policy Exchange argues that critics propose that the Prevent duty derives from the Trojan Horse affair.
The report says: “It is incorrect to suggest that the concept of a statutory Prevent duty arose solely because of Trojan Horse; the idea had been in existence well before the various enquiries and reports in 2014.”
However, most critics are identifying an evolution in Prevent from a concern with integration in its earliest inception, to a declared interest in challenging “extremist ideology” and not just violent extremism after 2011, though this shift was not fully implemented until 2015, after the Trojan Horse affair.
Another example of fallacy of composition occurs when Policy Exchange describes a letter to the Guardian newspaper critical of Clarke’s report and the Department for Education, which it said had been written by “Islamist individuals and organisations, along with their supporters elsewhere”.
Many of the signatories were mainstream educationalists. They did not sign the report because they were “supporters” of Islamism but because they were concerned about accuracy and high standards in public discourse.
The same fallacy of composition helps to account for Policy Exchange’s repeated use of the epithet “activist” to undermine the authority of the New York Times’ podcast, or that of anyone else who challenges its own narrative.
Here’s an example: “The podcast was mainly embraced by activists and their allies already sympathetic to Alam and the teachers involved in the affair, rather than the mainstream media."
One dictionary definition of the word activist is “a person who works to bring about political or social changes by campaigning in public or working for an organisation”.
Policy Exchange is therefore suggesting that academics who disagree with it are sacrificing scholarly standards of impartiality and accuracy to make a partisan argument.
Policy Exchange mounts the same charge against journalists, including those at the New York Times. In this context it is worth noting that a similar term - “activist lawyers” - has been used pejoratively by ministers to discredit legal challengers to government decisions.
This allegation against professional academics might carry more weight if Policy Exchange was fastidious and impartial in its own approach.
As it is, Policy Exchange would scarcely be in a position to complain if critics were to suggest that it has many of the credentials of what it calls an “activist” organisation - all the more so because the think tank and its fellows have a record of making mistakes in its reports about Muslims.
Some of Policy Exchange’s attacks are cheap. Consider this dismissive comment by Policy Exchange on Hamza Syed and Brian Reed, the journalists who present and report the New York Times’ podcast:
“But it is part of the route Syed and Reed appear to have taken as researchers: disprove the validity of particular letters, and they can then declare the subsequent ‘national panic’ to be without substance, without needing to investigate events inside the schools themselves.”
The above remark ignores the nature of the podcast genre. The substance of a possible injustice has to be established (as the New York Times did over two episodes) before the narrative can get going. In any case, the podcast then engages in considerable depth over two further episodes with the various allegations made about the schools.
Independent investigation
The Policy Exchange report cites parallels drawn by the authors of this article between the Trojan Horse affair and the cover-up that followed the 1989 Hillsborough disaster in which 97 Liverpool supporters died in a crush at the Sheffield stadium, accusing Oborne in doing so of “losing any sense of proportion”.
We draw no comparison between the events themselves but rather suggest that the process of setting up a similar body to the Hillsborough Incident Panel would be a useful first step towards establishing the truth of the Trojan Horse affair.
In the case of Hillsborough, the panel’s work was instrumental in exposing a police campaign to blame supporters for the disaster and highlighting the failings of the police and other bodies.
It led to a new inquest that in 2016 overturned an original verdict of accidental death and ruled the supporters had been unlawfully killed, in a process welcomed by the families of the victims as a long overdue correction of a miscarriage of justice.
An independent investigation into the Trojan Horse affair is now urgently needed, as Policy Exchange’s report has inadvertently shown. The claim made by Gove and Timothy that it should be considered “the final word” is false, arrogant, and flies in the face of the facts.
Additional reporting by Simon Hooper. MEE illustration by Mohamad Elaasar.
Middle East Eye delivers independent and unrivalled coverage and analysis of the Middle East, North Africa and beyond. To learn more about republishing this content and the associated fees, please fill out this form. More about MEE can be found here.