Paper - RRH Act MLR #3.rtf


 

This is the refereed author version of: 

Goodall, Kay E., Incitement to Religious Hatred: All Talk and No Substance?. 

Modern Law Review, Vol. 70, No. 1, pp. 89-113, January 2007. DOI: 

10.1111/j.1468-2230.2006.00627.x 

Available from: http://hdl.handle.net/1893/262  

 

 

The final publisher version is available from:  

http://dx.doi.org/10.1111/j.1468-2230.2006.00627.x  

http://hdl.handle.net/1893/262
http://dx.doi.org/10.1111/j.1468-2230.2006.00627.x


 1 

Incitement to Religious Hatred:  
all talk and no substance? 

 
 
The Racial and Religious Hatred Act 2006 has a frenetic history. It is the culmination of six 
attempts in Parliament in the last twelve years to make incitement to religious hatred 
unlawful.1  Each attempt has met with intense criticism.  But now that the legislation is here, 
what may it achieve?   
 
 
The background to the legislation 
 
The first version of the Racial and Religious Hatred Bill2 aimed to revise the entire existing 
offence of incitement to racial hatred, adding hatred of others on the grounds of their religious 
beliefs.3 Many thought this too wide. Lord Lester complained that the provisions were ‘using a 
steamroller to crack a nut’,4 Geoffrey Robertson QC described them as ‘unnecessary and 
clumsily framed’,5 and the Telegraph declared: ‘now you face jail for being nasty to 
Satanists’.6  
 
It was indeed wide. Incitement was to be by words, behaviour, or material which were 
‘threatening, abusive or insulting’. Intention to stir up religious hatred was not necessary. The 
offence would be committed if the words, behaviour or material were likely to be seen or 
heard by any person in whom they were likely to stir up such hatred.7  
 
During the debate, the Lords narrowed the offence in several ways, of which three in 
particular – the ‘Lester’ amendments – are significant. First, the Lords introduced a saving for 
freedom of expression, discussed later. Second, they removed the ‘likely to’ test, so the 
offence could be committed only if it could be shown that the person had intended to stir up 
hatred. Third, they altered the mode of incitement from abuse, insult or threat to simply threat.  
 
In the Commons at third reading, the government tried to replace the ‘likely to’ element with a 
new test of subjective recklessness. They were spectacularly defeated and the Lester 
amendments stayed.8  The media reported a mood of relief:9 creating an offence of incitement 

                                                      
* Thanks to Francis Bennion, Adam Tomkins, Colin Gavaghan, Anna Burnside and ATH 
Smith for their helpful comments. The opinions expressed here, and of course any mistakes, 
are mine alone. 
1 Lord Lester proposed two amendments to the Criminal Justice and Public Order Bill, June 
1994 and the (Lord) Bishop of Oxford moved an amendment in July 1994. Lord Avebury 
introduced a Religious Offences Bill in 2001. The government has tried three times: in the 
Anti-Terrorism, Crime and Security Bill in 2001; in the Serious Organised Crime and Police 
Bill, introduced in November 2004 and (successfully) in the Racial and Religious Hatred Bill, 
introduced June 2005. 
2 Bill 11, 54/1; introduced 9 June 2005. 
3 It also proposed minor changes in the wording of the existing offence: I will not discuss 
those here. 
4 Anthony Lester in Lisa Appignanensi (ed.) Free Expression is No Offence (London: 
Penguin, 2005) 234.  See also the highly critical comments later made by FAR Bennion in 
‘Gilding the Lily on Religious Hatred’ (2005) 14(1) The Commonwealth Lawyer 35. 
5 Quoted in Ruth Gledhill et al. The Times February 7 2005. 
6 Headline over report by Joshua Rozenberg, Legal Editor, The Telegraph 10 June 2005. 
7 Schedule, paragraphs 5-11, Bill 11 54/1, which would have been inserted as sections 18-23 
of the 1986 Act. 
8 HC Deb 31 January 2006 vol 442 cols 235-244. This was only the second defeat suffered by 
new Labour in the House of Commons since 1997.  
9 David Charter The Times 1 February 2006, Michael White; Tania Branigan The Guardian 1 
February 2006, Ben Russell; James Lyons; Brendan Carlin and Neil Tweedie The Telegraph 
1 February 2006.  Similar views were presented in the red-top press. 



 2 

to religious hatred was seen as unpopular and illiberal10 and many argued that the Lester 
amendments created a sharper, more effective piece of legislation which had less impact on 
freedom of expression.  Were they right?  This article will consider the many questions which 
the new Part 3A raises. It will conclude that an offence confined to requiring proof of intention 
simply will not work.   
 
 
The new offence 
 
The Act creates a new Part 3A of the Public Order Act 1986 for England and Wales.11  Like 
the law on racial hatred,12 Part 3A casts a wide net. It covers speech and actions. It includes 
displaying material, publishing it and distributing it, and sometimes even possessing it. Those 
who present or direct a play, or distribute, show or play a recording, or include a programme 
in a programme service,13 may all be caught within the offence.   
 
The prosecution must show that the words, behaviour or material were threatening, and that  
the person intended to stir up religious hatred. ‘Religious hatred’ is defined in section 29A as 
‘hatred against a group of persons defined by reference to religious belief or lack of religious 
belief’.   
 
The offence need not be committed in public. It can take place in private. A person may 
commit the offence in a private dwelling by using threatening words or behaviour or displaying 
(with the requisite intent) any written material which is threatening, if he has any reason to 
believe that this might be seen or heard by people outside. Woe betide those who live in 
modern tenements?  The offence can also be committed in an inchoate form, such as storing 
material with the intention of broadcasting it. 
 
Any offence which includes written and broadcast material will extend beyond territorial 
borders. Part 3A will mostly not apply to Scotland, but anyone being recorded or producing 
material in Scotland may commit the offence if for instance their writings appear in England or 
their material is broadcast over the border.14  
 
Given the many activities which may be caught by Part 3A, politically active people need to 
know what is now illegal. What harms did law fail to catch, and what harms will it capture 
now?  This is a difficult question to answer.  
 
 
Gaps in the law   
 

‘When the Lord Chancellor started to talk about gaps in the law, he made me 
think of a rather over-zealous dentist looking round the teeth of a patient to see 
whether any holes need stopping’.15 

 

                                                      
10 Such views were expressed throughout the parliamentary debates: see Jonathan Petre The 
Telegraph, 31 January 2006, Leader The Guardian January 31 2006, Matthew Tempest; Polly 
Toynbee The Guardian January 31 2006 as examples of the last comments before the final 
Commons debate. 
11 Racial and Religious Hatred Act 2006 s 4.  
12 Which remains unchanged. 
13 This will include a cable ‘programme service’: for a definition, see the Broadcasting Act 
1990 s 201. 
14 Although Part 3A does not extend to Scotland other than police powers, it will affect anyone 
whose legal activities in Scotland also take place in England and Wales, as where Scottish 
newspapers are circulated in England or Scottish programme material is shown in England. 
See HA Kerrigan QC and Neil Addison Joint Opinion of Counsel for the Christian Institute re 
the Impact on Scotland of the Racial and Religious Hatred Bill (2005)   
http://www.christian.org.uk/scotland/religioushatred/kerrigan&addison_opinion.pdf  
15 Lord Peyton of Yeovil HL Deb 8 November 2005 vol 439 col 554. 



 3 

One of the most compelling arguments against creating an offence of incitement to religious 
hatred was that English law adequately addresses the mischief already, or mostly does so.16  
Many felt that the problem was lack of enforcement, not lack of legislation. Some critics 
accused the government of playing to Muslim voters disenchanted by UK involvement in Iraq, 
rather than addressing a genuine problem.17 Others complained that the gap was so narrow 
that filling it caused more problems than it solved.  
 
There has been immense confusion over this point. There is certainly a powerful body of 
criminal law which can tackle behaviour displaying religious prejudice. From religiously 
aggravated abuse and violence to criminal harassment; from the Ecclesiastical Court 
Jurisdiction Act 1860 to the Malicious Communications Act 1988; from incitement to crime to 
aggravated breach of the peace,18 most incidents would have been covered by existing law; 
usually public order offences.  
 
There is ample law to cover anyone who has targeted person or property, or committed a 
breach of the peace. If the offender is motivated by or displays racial or religious hostility, this 
may also attract an enhanced sentence.19 The typical case involves words or behaviour 
directed against individuals, but the law can also capture actions with more generic targets. In 
Norwood v DPP, a man displayed a British National Party (BNP) poster beside a photograph 
of one of the twin towers in flames and a crescent and star surrounded by a prohibition sign.20 
This display did not target any specific person, but Norwood was convicted of a religiously 
aggravated offence under section 5(1)(b) of the Public Order Act 1986.21  There is also the 
common law of incitement. If one incites another to commit a criminal act, this too is an 
offence. If a preacher exhorts his flock to ‘go forth and slay the unbelievers’, he may be 
prosecuted. 
 
Yet supporters of the Bill often gave a very different impression. Members of Parliament 
offered examples of supposed gaps in the criminal law which they maintained the new 
offence would fill. On several occasions the examples were (no doubt unintentionally) 
misleading. Members listed incidents where Muslim women, including White converts, were 
spat at; where a White Muslim woman was verbally abused when she was observed wearing 
a hijab on a bus and where unnamed people incited others to fire-bomb Muslim homes. As 

                                                      
16 Baroness Kennedy HL Deb 11 October 2005 vol 437 col 247; Lord Lester of Herne Hill 11 
October 2005 vol 437 col 175. 
17 Lord Lester of Herne Hill HL Deb 25 October 2005 vol 438 col 1075; Ian Paisley, Boris 
Johnson and John Baron HC Deb 21 June 2005 vol 435 cols 718-745. 
18 While breach of the peace is not classified as a criminal offence in English law (although it 
can be a constituent of an offence), it has been classified as such for the purposes of Articles 
5 and 6 of the European Convention on Human Rights (see Steel v UK (1999) 28 EHRR 603, 
637) and it is a crime within Scots law. 
19 The relevant provisions are ss 28-32 of the Crime and Disorder Act 1998, as amended by s 
39 of the Anti-Terrorism, Crime and Security Act 2001, which add sentence enhancement to 
nine existing offences; and s 145 of the Criminal Justice Act 2003, which applies to all other 
offences.   
In Scotland, sentence enhancement for racial aggravation is governed by s 96 of the Crime 
and Disorder Act 1998, while religious aggravation is covered by s 74 of the Criminal Justice 
(Scotland) Act 2003.  The test for the equivalent offences is ‘malice and ill-will’ (see below). 
In Northern Ireland, statutory sentence enhancement for inter alia racial and religious 
aggravation was introduced by paragraph 2 of the Criminal Justice (No.2) (Northern Ireland) 
Order 2004.  
20 Norwood v DPP [2003] EWHC 1564 (Admin); 2003 WL 21491815 (QBD (Admin Ct)).  
21 s 5(1)(b) provides that a person is guilty of an offence if he inter alia ‘displays any writing, 
sign or other visible representation which is threatening, abusive or insulting, within the 
hearing or sight of a person likely to be caused harassment, alarm or distress thereby’ and 
intends (see s 6(4)) that it be threatening, abusive or insulting, or is aware that it might be. 
Norwood was convicted of the offence as religiously aggravated under s 28(1)(b) of the Crime 
and Disorder Act 1998. 



 4 

Evan Harris MP pointed out during debate, all these could have been caught already.22 As 
baffled MPs discovered, the real gap was hard to spot. Government spokespersons were 
intractable when pressed to give examples:23 one of the few helpful statements is to be found 
in Lord Falconer’s speech at second reading in the Lords.24  
 
For practical purposes,25 the sole area in which law was lacking is where a person incites 
another, not to crime, but to religious hatred.  It is not unlawful to feel hatred, so incitement to 
hatred is not in itself an offence. Incitement to racial hatred is unlawful because it has been 
made so within the Public Order Act 1986.26  People were free to incite other forms of hatred. 
This is the crux. This brought about the many attempts to legislate. It was only when someone 
incited another to hate others, or acted in a way which was likely to stir up that hatred - 
without going so far as to incite them to commit any offence - that the gap appeared.  If the 
hatred thus stirred up was deemed ‘racial’, an offence had been committed.  If the hatred 
stirred up was deemed ‘religious’, the incitement was lawful.   
 
This gap is tiny. There were only 76 prosecutions for racial hatred between 1987 and 2005.27    
In most cases it matters little whether hatred expresses itself as religious or racist because it 
will can be caught under another heading of law. The gap lies almost entirely in naming acts 
as religious hatred, rather than in prohibiting them.   
 
If not for the unfortunate way in which the jurisprudence has been interpreted, this might be 
little more than a sophist’s distinction. But precedent has determined that certain faiths are 
‘mono-ethnic’.  It is thus easier for the prosecution to establish that an attack on Jews or Sikhs 
is to be deemed ‘racial’. Judaism and Sikhism have been defined as mono-ethnic faiths in 
British law.28 
 
Demographic research in fact supports this.  Most Sikhs in Britain share Indian ancestry,29 
and Jewish identity is mostly passed on through lineage rather than by conversion. While 
Jewish ancestry is more diverse than Sikh, the sense of Jewish identity is strong. Even 
among younger British Jews, recent research found that ‘almost all participants… felt 
themselves to be members of an ethnocultural group.’30   

                                                      
22 Sadiq Khan, Emily Thornberry, Evan Harris HC Deb 21 June 2005 vol 435 col 721-737.  Dr 
Harris cited the three examples above. For relevant law, see for instance the Crime and 
Disorder Act 1998, ss 28(1)(a) and (2), as amended, and the common law offence of 
incitement. 
23 See in particular the debates in HC Deb 21 June 2005 vol 435 cols 668-762. 
24 Lord Falconer of Thoroton, HL Deb 11 October 2005 vol 437 cols 162 and 169. He 
explained that ‘[t]he problem …is where you are addressing people other than the people 
whom you are insulting or abusing with a view to stirring up hatred among them against a 
particular religious group’ and gave several examples, such as a Muslim cleric preaching 
righteous hatred of Christians to a Muslim audience. (It was not clear however why some of 
the examples were not covered by existing law.) 
25 The 2003 Report of the Select Committee on Religious Offences discussed some other 
miniscule gaps: there is an interesting summary in chapter 3 Religious Offences in England 
and Wales vol 1 HL  95-I  (2003). 
26 Part 3; ss 17-23. 
27 HL Deb 11 October 2005 vol 437 col 274.   
28 Mandla v. Dowell Lee [1983] 2 AC 548 (HL), 565. Jews and Sikhs were specifically 
identified in Mandla as groups defined by ethnic origins.  This was a civil definition based on 
the definition in s.3 of the Race Relations Act 1976, but criminal law has adopted this 
definition: s.28 of the Crime and Disorder Act 1998 replicates s.3 except that it also includes 
‘citizenship’ and clarifies that it is irrelevant for the offence that the victim is also a member of 
a religious group. 
29 In the 2001 Census, 91 per cent of British Sikhs were recorded as ‘Indian’ (and 97 per cent 
of Jews were recorded as ‘White’): National Statistics, Focus on Ethnicity and Identity, ONS 
2005, 6.  
30 J Sinclair and D Milner ‘On being Jewish’ (2005) 20(1) Journal of Adolescent Research 91,  
112. British Jews themselves tend to see Jewishness as an ethnic rather than solely a 



 5 

 
In contrast, Islam and Christianity are worldwide religions.31 So, incitement to hate Jews, 
Sikhs (or Gypsy travellers32) can fall within the racial hatred provisions, while incitement to 
hate Muslims or Christians33 it seems cannot. We will return to the inconsistencies in that 
reasoning later.  What is important for now is that it is here where the anomaly lies. It is this 
distinction which has upset many. 
 
Thus although the existing law covered almost all of the potential cases, the problem was not 
the breadth of the gap but its significance.  A brief visit to the  British National Party website 
provides an apposite example. The site is regularly purged of statements which may attract 
legal action. There is guidance for contributors from party leader Nick Griffin: ‘[a]ny 
connection between sex, particularly sex crime, and members of ethnic 'minorities' is 
dynamite’. He adds: 
 

‘Emotive words, however justified they may be, must be avoided. Truth hurts, so 
words like 'alien', 'vermin', 'gang' instead of 'group', and such like must be 
avoided. A white rapist may be described as a 'beast' or an 'animal', but a black 
one must merely be a 'criminal'.’34  

 
With this in mind, it is not difficult to decode the following statement, from an article called 
‘The Islamic Menace’:  
 

‘The hidden epidemic of molestation, abductions and rape of scores of white girls 
in northern English cities, all show the inherent tendency that the teachings of 
the paedophile Mohammed have had on some of his followers.’35   

 
It is likely that the statement is not simply anti-Muslim but racist, and will be read so by BNP 
supporters. Yet the BNP’s legal advisers seem to believe it is not unlawful (surprisingly so, 
considering the reference to ‘white’ girls). This statement is nothing more than an incitement 
to hatred, so the racial/religious distinction becomes crucial. We see this in a later comment 
by Griffin:  
 

‘We can sometimes get away with criticising Zionists, but any criticism of Jews is 
likely to be legal and political suicide.’   

 
It is because far right activists have so successfully exploited the distinction between ‘racial’ 
and ‘religious’ that Muslims in particular have demanded the new offence.  There are other 
reasons: where wider common law already covers the field, a new nominate offence can 
make it easier to monitor prevalence, label the offender and identify a criminal history if the 
person re-offends. But it is the distinction between Muslims and adherents of other faiths 

                                                                                                                                                        
religious identity: Stephen Miller, Marlena Schmool and Antony Lerman Social and Political 
Attitudes of British Jews (London: Institute for Jewish Policy Research, 1996) ch.7. 
31 In fact, as DT Goldberg has pointed out, ‘[t]here are larger numbers of Muslim people today 
who are members of European or European-derivative societies than vice versa.’ ‘Racial 
Europeanization’ (2006) 29(2) Ethnic and Racial Studies 331, 360. 
32 CRE v Dutton [1989] QB 783 (CA), 801 and 806.  Note that this only applied to a subset: 
those of Romany origin.  
The Race Relations (Northern Ireland) Order 1997 specifically defines Irish travellers as a 
racial group (see also A Racial Equality Strategy for Northern Ireland 2005-2010, OFMDFM, 
NI).  
33 There is little authority for this point as regards Muslims. In the only reported case on 
Muslims as a ‘racial group’, the civil case of Walker v Hussain [1996] ICR 291 (EAT), 
‘Muslims’ in the abstract were taken to be a religious, not a racial group, but the arguments 
are not explored in the appeal judgment.  Sebastian Poulter cites an unreported decision of 
an industrial tribunal which goes into more detail: see Poulter ‘Muslim Headscarves in School: 
Contrasting Legal Approaches in England and France’ (1997) 17(1) Oxford Journal of Legal 
Studies 43, 62-65.  
34 http://www.bnp.org.uk/articles/race_act.htm, accessed 8 February 2006. 
35 http://www.bnp.org.uk/articles/islamic_menace.htm, accessed 8 February 2006. 



 6 

which allowed far-right activists to revile minority ethnic Muslims while other religious groups 
were better protected by law. It is because the far right so viciously exploit this anomaly that 
so many campaigned so hard to introduce the new offence. 
 
If the anomaly had been trivial, and if the offence had already been clearly labelled by the 
criminal law, then new legislation might have been unnecessary. It would have done little 
other than demonstrate that governments believe that electorates have short memories, and 
that it is easier to make new law than to remind them of what exists already.  But the anomaly 
was real, and serious.   
 
 
The meaning of religious hatred 
 
The legal source of this anomaly, however, is elusive. It is widely thought that Jews are 
always covered by the racial hatred provisions, while Muslims are not.  This is wrong. The 
relevant case law has sometimes been misunderstood by commentators, and we lack 
interpretative guidance from cases taken to appeal.  As ATH Smith has observed, in public 
order cases the higher courts have been loath to give it.36  
 
The confusion appears to come from the belief that this was the conclusion reached in R v 
DPP, ex p. Merton LBC.37  Critics claim that Crown refused to bring a case against a BNP 
activist solely on the grounds that the literature he produced was anti-Muslim and hence anti-
religious, not racist.38 It is true that this was the Crown’s initial decision, but they later changed 
it, stating that the position of Muslims is uncertain and that this did not prevent prosecution. In 
fact, Merton was simply a refusal to grant leave to apply for judicial review. The Council 
wanted a declaration that Muslims were a racial group under section 17 of the Public Order 
Act 1986, separate from any criminal proceedings, but Tucker J correctly ruled that a court 
could not be asked to decide a hypothetical question.   
 
There is a genuine legal distinction, but it is a matter of degree and circumstance. If anti-
Judaic literature is deemed to have a solely religious target, Jews would be left unprotected. 
Indeed Lester and Bindman argued in 1972 that the provisions on incitement to racial hatred 
‘left members of the Jewish community without legal protection against religious as distinct 
from racial persecution’.39   
 
In practice, though, the presumption is that a reference to Jewishness in hate literature is 
racist.40 It has not been difficult to prove.  In R v Edwards,41 for instance, a cartoonist drew a 
racially stereotypical image of a Jew draining human blood. The key events which first 
demonstrated the need for legislation - Caunt, Leese and fascist activities from the 30s 
onwards - were characterised by open anti-Semitism or express links with Nazism.42  
 

                                                      
36 ATH Smith The Offences Against Public Order (London: Sweet & Maxwell, 1987) 43.   
37 R v DPP, ex p Council of London Borough of Merton [1998] EWHC Admin 1009; 
CO/3019/98 (27 October 1998).  A report is available at www.bailii.org  
38 See eg KS Dobe & SS Chhokar ‘Muslims, ethnicity and the law’ (2000) 4 International 
Journal of Discrimination and the Law 369, 373. (This excepted, this is an excellent article).  
The error may have come from a misreading of an article by Peter Jepson ‘Tackling religious 
terminology that stirs up racial hatred’ (1999) NLJ 554, 556 where he briefly refers to the 
case. 
39 Anthony Lester and Geoffrey Bindman Race and Law (Harmondsworth: Penguin, 1972) 
363.  See also Bob Hepple (1966) 29 MLR 306, 308. 
40 This was recognised in Mandla v. Dowell Lee [1983] 2 AC 548 (HL), 561, where Lord 
Fraser of Tullybelton said obiter: ‘it is inconceivable that Parliament would have legislated 
against racial discrimination intending that the protection should not apply either to Christians 
or (above all) to Jews’.    
41 R v Edwards (1983) 5 Cr App R (S) 145, 147. 
42 Caunt spoke openly of his anti-Semitism and Leese’s supporters collected him from prison 
waving swastikas.  See Anthony Lester and Geoffrey Bindman Race and Law 
(Harmondsworth: Penguin, 1972) 343-352.   



 7 

In contrast it will often be impossible to show the link between Islamophobic material and 
racist intent. It is however open to a prosecutor to show an established history of racist activity 
against British Muslims. Proving this link is complex because anti-Muslim hostility is more 
than simply anti-‘Asian’ hostility.43 Indeed, British-Asian Muslims themselves have a strong 
religious identity. Research suggests that religion is often central to how these groups 
differentiate themselves, particularly among the first generation.44 Muslim identity is not simply 
ethnic identity. 
 
But there has been a successful prosecution in Scotland of a BNP activist who produced anti-
Muslim hate literature. In 2002, BNP activist David Wilson was convicted of distributing 
leaflets containing false accusations about Muslims in Glasgow’s Pollokshields.45 The leaflets 
claimed that ‘Muslim racists’ were ‘running amok’ and attacking ‘white’ residents.  Evidence 
was given to the court that the ethnic origin of most Muslims in Pollokshields was South 
Asian.  The sheriff concluded that the terms Muslim and Pakistani were used interchangeably 
in that community. Wilson was convicted under section 19(1)(a) of the 1986 Act for stirring up 
racial hatred. He appealed, and the decision of the sheriff was affirmed.46 
 
The two types of hatred are often entangled to their roots. Even in the flagrantly racist case of 
Edwards, the Jewish man was depicted in what was claimed (falsely) to be a religious 
practice – eating human blood at the festival of Purim. The problem is not bad law: it is simply 
that there is a blurred line and a need for proof – and perhaps also a stronger will is needed 
somewhere in the criminal justice process. 
 
 
Fair comment on Islam - or racial hatred? 
 
Extreme racism denigrates people and represents them as inferior: they endure a lower social 
or economic status and become targets of hatred or contempt. It is irrational, imputing 
characteristics to a whole group. It is also xenophobic, because the group is identified by 
supposed ethnic origin.   
 
With this in mind, consider Norwood, which associated Islam with terrorist bombing. Islam is a 
mainstream world religion, not a warrior cult. It cannot therefore be a legitimate criticism of 
Islam to imply that most or all Muslims actively abet terrorism.  There is also abundant 
evidence that most British Muslims practise a moderate form of Islam, so it cannot be a 
legitimate criticism of British Muslims to associate them with terrorism.   
 
What is true of British Muslims is that few have a long British ancestry.  Ethnically, they are 
‘other’. Only 4 per cent of British Muslims were classified as ‘White British’ in the 2001 

                                                      
43 For evidence suggesting anti-Muslim discrimination in employment, separate from ethnic 
identity in the UK, see J Lindley ‘Race or religion? The impact of religion on the employment 
and earnings of Britain’s ethnic communities’ (2002) 28(3) Journal of Ethnic and Migration 
Studies 427.   
44 Peter Aspinall ‘Should a question on “religion” be asked in the 2001 British census’ (2000) 
34(5) Social Policy and Administration 584, 588-591.     
45 PF v Wilson Sh.Ct. 24 October 2002, unpublished (my thanks to the Crown Office for 
providing a copy of Sheriff Ruxton’s decision).  This was the first Scottish prosecution for a 
racial hatred offence under the 1986 Act.   
See also the civil case of Walker v Hussain, cited above at FN 32. In this case the workplace 
was held to have two distinct racial groups: a European one and one originating from the 
Indian sub-continent, nearly all of whom were Muslim. It was held that discrimination on 
grounds of Muslim holidays was indirect racial discrimination. In Scotland, it was held in 
Empower Scotland v Khan [2005] UKEAT 0012, 2005 WL 2608279 (EAT) 2nd September 
2005, that in a workplace where 90 per cent of Pakistanis were Muslims, an insult aimed at 
Muslims could be read together with the statement that ‘You Pakistanis are all the same’ and 
amount to race discrimination. 
Across the UK, 92 per cent of Pakistanis and Bangladeshis are Muslim: National Statistics, 
Focus on Ethnicity and Identity ONS (2005) 6.  
46 Wilson v PF [2005] HCJAC97, 2005 SCCR 686. 



 8 

Census, and nearly three-quarters had South Asian origins.47 Also, British Islam is typically 
seen by white British people as being Asian, because it is the South Asian (usually Pakistani) 
Muslims who represent Islamic interests in British political debate.48   
 
The bigotry displayed in Norwood has the hallmarks of racism. Race or religion might define 
exactly who falls into the hated category, but we have already seen reports of Hindus and 
Sikhs subjected to anti-Muslim abuse. Because most British Muslims do not have White 
British ancestry, it will sometimes be possible to prove that ‘Muslim’ has been used as a racist 
marker – perhaps where the defendant has a history of racist activism. Mark Norwood, for 
instance, was a regional organiser for the BNP.   
 
Because Islam is a mainstream religion, it will also sometimes be possible to distinguish 
between (a) angry criticisms of Islamic practices and (b) anti-Muslim abuse which is motivated 
by racist bigotry. It is lawful to express angry criticism of a specific extremist Muslim 
organisation, or to mount a hate-filled attack on characteristics which most British Muslims do 
have (for instance, a belief that women should cover their hair), or to denounce Islamic states 
which enforce harsh Hadd offences under Sharia law.  These are all quite different from the 
sweeping anti-Muslim hatred which depicts British Muslims as members of a warrior cult or 
(as in Griffin and Collett49) a rapist sect. 
 
Griffin and Collett defamed a group of people, most of whom are South Asian in origin and 
almost all of whom are not ‘White British’.  Had they offered fair comment on Islam - however 
intemperate – that would be different.  If an Islamic sect in the UK advocates violence, both 
the state and individuals can legitimately respond. The state can proscribe the offending 
organisation.50 The activism of a Mark Norwood could focus only on that particular sect and 
so be a legitimate critique which could – depending on the circumstances - be expressed with 
great force. At present, it is nothing of the sort.   
 
Here lies the basis for a manageable legal distinction.  The term ‘racial hatred’ has not been 
definitively classified in criminal law as excluding any hatred expressed in religious terms. The 
words are to be given their ordinary and natural meaning, and while it is harder to prove that 
hatred targeted at Muslims addresses a racial group, it can be done.  
 
 
How might law distinguish between racial and religious? 
  
Both the reason Part 3A was introduced, and the root of the furious opposition to it, is this 
separation between ‘racial’ and ‘religious’.  Many presume that ‘race’ is a clear and fixed 
quality of identity, while religion is not. There is much truth in this - and it would be invidious to 
talk up the uncertainty of the marginal case while resolutely ignoring the racial ascriptions of 
everyday experience. We are right to make law in the belief that people are singled out 
because of their ‘race’, and treated well or badly depending on the racial identity usually 
ascribed to them. Race, for most people, most of the time, is indeed clear and fixed.  
Religious affiliation, in contrast, is often less easily discerned by others and is not immutable 
(even if it is rare that people face an open choice in which faith to adopt). And religious 
affiliation has an obvious political and moral component, while someone’s skin colour and 

                                                      
47 In the 2001 Census, 74 per cent were recorded as coming from an Asian ethnic 
background (39 per cent having been born in Asia), while only 11 per cent came from any 
White ethnic group: National Statistics, Focus on Religion ONS (2004) 5.  It is not clear 
whether those from Mixed backgrounds also had some Asian or White ancestry, though the 
proportion is in any case small.   
48 Pnina Werbner ‘Theorising complex diasporas: purity and hybridity in the South Asian 
public sphere in Britain’ (2004) 30(5) Journal of Ethnic and Migration Studies 895, 898. 
49 Griffin is reported as having taken advantage of a conviction of young Asian men for rape 
by saying (as part of a long tirade) ‘These attacks are going to continue, because that is what 
the Qur'an says. The bastards that are in that gang, they are in prison so the public think it's 
all over. Well it's not. Because there's more of them.’  The Guardian January 18 2006.  
50 By order under the Terrorism Act 2000, Part 2.  



 9 

facial features may give no guidance at all about a person’s allegiances or beliefs.  It does 
make sense for law to distinguish religion and ethnicity. 
 
The need to reach decisions in particular cases does not however mean that legal personnel 
should strive to create definitions which can be packaged up and carried from case to case. It 
is not just that ‘racial’ classifications (in their narrowest sense) have no sound biological 
basis.51 Ethnic classifications too have no logical one beyond social and historical 
convenience. Instead they are a mishmash of skin colour, geographical ancestry, nationality 
and local community identity.52 Nationality is the only element which can reasonably said to 
be dominated by a co-ordinated set of rules, because its categories are created by law. There 
is no other absolute marker for distinguishing these social, not biological categories.   
 
And even if such absolute markers existed, they would not mesh with local perceptions. 
Hostile racial ascriptions are highly contextual. At different times and in different places, a 
person may ‘pass as white’ or be singled out as alien. Meanwhile those who are always seen 
as alien may be ‘one of us’ or ‘one of them’ depending on whether their community is seen as 
integrated into British cultures. ‘Race’ may be mostly clear and fixed but racist hostility is 
volatile and hard to predict.   
 
Religion, too, needs to be understood both as faith and as social category – as a way of 
marking out groups.  In Scotland and Northern Ireland, sectarian conflict is between two 
groups who use religion mostly as a social category: it is not a battle of faith between devout 
believers.53 Yet conflict over faiths persists: the most recent report of the US Office of the 
Coordinator for Counterterrorism lists 37 ‘foreign terrorist organisations’, of which at least 22 
are expressly religious.54 
 
The problem for law is that (as we saw above) British Muslims and South Asians are often 
seen by others as interchangeable. Islam has a significant public identity in Britain and so we 
cannot simply assume that anti-Muslim hostility is racist.  But ‘Muslim’ as a term of abuse in 
Britain is often shorthand for other-than-white. Tariq Modood et al argue that ‘there is a 
palpable racialization of Muslims going on in Britain’55 – in other words, Muslim is often used 
to mean someone who looks ‘Asian’.  
 
This needs to be recognised in criminal law. Lawyers’ perceptions, or prosecutorial decisions 
- or both - have become riddled with irrelevant and ahistorical distinctions taken from civil 
discrimination. Criminal actions are not so easy to subdivide.  
 
A civil case may well consider a pattern of discrimination, while criminal prosecutions typically 
centre on a single incident. Yet social conventions in the UK now discourage explicit racist 
speech. This makes it hugely problematic to identify any single act as racist and to find 
evidence to justify that decision. Emo Gotsbachner has demonstrated how racism in everyday 
discourse is often expressed by omission and emphasis rather than by unambiguously 
prejudiced speech - often one can judge whether a statement is racist only in the context of 

                                                      
51 While the relative incidence of genes varies among different human populations, they do 
not tend to vary in a way which indicates precisely demarcated groups.  ‘Race’ as a means of 
genetic classification, without more, is rejected by the overwhelming majority of modern 
biologists. There is a wealth of writing on this, but for short and authoritative recent 
statements, see issue 5(8) of Nature Reviews Genetics (2004). 
52 Again, there has been a torrent of material on this topic.  The best work on ethnicity comes 
from social scientists: two intelligent introductions are John Solomos and Les Back Racism 
and Society (Basingstoke: Palgrave Macmillan, 1996) and Kenan Malik The Meaning of Race 
(Basingstoke: Macmillan, 1996). 
53 See L Philip Barnes ‘Was the Northern Ireland Conflict religious?’ (2005) 20(1) Journal of 
Contemporary Religion 55, 59 ff. Barnes points out, for instance, that those most directly 
involved in the violence in Northern Ireland were not the most religious, and that the main 
churches did not sanction the violence. 
54 US Department of State, Country Reports on Terrorism 2004 (DSP, 2005) 92-111. 
55 Tariq Modood, Richard Berthoud and James Nazroo ‘”Race”, Racism and Ethnicity’ (2002) 
36(2) Sociology 419, 422. 



 10 

the whole exchange.56  Civil law is better placed to do this and so it makes sense to 
distinguish the ‘racial’ and the ‘religious’ in civil discrimination law. The concept of racial intent 
in criminal law however rests on a particularly shaky house of cards.  A ‘contextual meaning’ 
approach to incitement cases, judicially developed, is long overdue.   
 
Successive judgments in the aggravation case law have led to more contextual 
interpretations. In the particular circumstances of a case, words such as ‘African’,57 ‘bloody 
foreigners’58 and ‘immigrant’59 have all been enough to convict for an offence racially 
aggravated.  Most recently in R v Rogers, the Court of Appeal held: 
 

‘Hostility demonstrated to foreigners because they are foreign can be just as 
objectionable as hostility based on a more limited racial characteristic. All who 
are black form a racial group, defined by reference to colour, within section 
28(4), as do all who are white. This demonstrates the width of the concept of 
racial group in this context. It is no great extension of the concept to embrace 
within a single racial group all who are foreign.’60 

 
It would not be a huge leap to alter the law so that the use of referent words as a surrogate for 
race expressly fell within incitement to racial hatred. Lord Lester twice proposed an 
amendment which would have achieved this.61 
 
If this worked, it would capture racial hatred directed at minority ethnic Muslims – the core 
problem which the government declared it wished to tackle.  If so, then the sole gap left would 
be incitement to hatred directed at white Muslims.  And even there, White people victimised 
(for instance) because of their relationships with Asian Muslims are covered by the racially 
and religiously aggravated offences.62 While incitement to hatred against them, couched in 
religious terms, could become a widespread problem, there is not enough evidence to justify 
introducing an ambiguous new law for this alone. 
 
So is this the solution?  Unfortunately, it may not be. The problem is that there are two 
phenomena we associate with ‘religious hatred’: 
 
• incitement to hatred motivated by racism disguised as antipathy towards a religion 
• incitement to hatred genuinely motivated by religious or anti-religious belief. 
 
The first is what Part 3A was designed to combat; the second is what it is far more likely to 
catch.  Drawing on John Brewer’s work on sectarianism,63 we can take this analysis a little 
further.   There are three relatively common problems in the UK: 
 
• religion used as a surrogate for racism (as in the opportunist change of language in 
far-right literature) 

                                                      
56 Emo Gotsbachner ‘Xenophobic normality: the discriminatory impact of habitualised 
discourse dynamics’ (2001) 12(6) Discourse and Society 12(6) 729, 731. 
57 R v White [2001] EWCA Crim 1756 at [17-19]; 1 WLR 1352: the word was said to have a 
‘racial connotation.’   
58 DPP v M [2004] [2004] EWHC Admin 1453 at [31]; 1 WLR 2758.  The hostility was in the 
context of that case directed at a member of a group of people defined ‘by reference to race’.  
See also R v Rogers [2005] EWCA Crim 2863 at [20]; [2006] 1 Cr App R 14. 
59 Att Gen’s Reference No.4 of 2004 [2005] EWCA Crim 889 at [24]; [2005] 1 WLR 2810. 
60 R v Rogers [2005] EWCA Crim 2863 at [20]; [2006] 1 WLR 962 at 968.  The case is under 
appeal to the House of Lords. 
61 This was first proposed on 16 March 2005 for the Serious Organised Crime and Police Bill 
and the text can be found at  
http://www.parliament.the-stationery-office.co.uk/pa/ld200405/ldbills/024/amend/ml024-
ire.htm.   
62 Crime and Disorder Act 1998, ss 28(1)(a) and (2), as amended.  
63 JD Brewer ‘Sectarianism and racism, and their parallels and differences’ (1992) 15(3) 
Ethnic and Racial Studies 352, 358. 



 11 

• religion acting as a social marker for other conflicts, economic or social (as in most 
Protestant/Catholic sectarianism in the UK)64 
• religion as a source of conflict in its own right (as in the aggressive activities of 
extremist Muslims and Christians against other faiths). 
 
We will see in the last section how difficult it is to design clear legislation which captures the 
first two problems. Until religious extremists adapt their rhetoric, it may be primarily the third 
group which is vulnerable to Part 3A.  Such activists are just as deserving of punishment (and 
indeed may have links to the far right themselves) but any appearance of bias will be 
politically destructive. 
 
 
The ‘ordinary meaning’ of ‘hatred’ 
 

‘I would like to incite people to hate bigotry, and I am worried about provisions 
that say that I cannot go round inciting people in that way. That incitement - 
which, as we have heard, involves loathing and intense dislike - is integral to our 
tradition.’65 

 
The flaws I have described above are part of a deeper weakness in the law. The words of the 
Act are largely the language of ordinary speech, but ordinary speech does not define their 
legal purpose. Much of the disagreement over whether Part 3A conflicts with freedom of 
expression comes from a reluctance to address the distinction between legal meaning and 
colloquial meaning. ‘Hate’ used in ordinary discourse has many meanings, including justified 
dislike. As Tony Wright MP observed in the quote above, it includes loathing of religious 
bigotry. Even in this most passionate use of the term, inciting others to feel the same should 
not amount to religious hatred in the statutory sense.    
 
Part 3A will have to be interpreted in the light of much other law. It will be interpreted through 
prior case law on racial hatred – after all, this statute has come about primarily because of 
that well-established principle of statutory interpretation. If incitement to racial hatred was not 
understood in the light of restrictive precedent on the meaning of ‘racial group’, this new 
statute would probably never have been proposed.   
 
It may be however that the meaning of ‘hatred’ will be left to juries as judges fall back on the 
argument that the word should be given its ‘ordinary meaning’.66 Back in 1987, ATH Smith 
gave a vivid example of the conflicting outcomes which can result from this approach.67 While 
there is a powerful argument that juries normally do not need to be directed on the meaning of 
ordinary words,68 more guidance would be helpful here.  
 
There are however limits. The statute must be understood within all the usual guides to 
legislative intention. It is the constitutional duty of the judge to discern the legal meaning69 of 

                                                      
64 L Philip Barnes ‘Was the Northern Ireland Conflict religious?’ (2005) 20(1) Journal of 
Contemporary Religion 55, 56 and ff. 
65 Dr Tony Wright, HC Debs 21 June 2005 vol 435 col 730.   
66 See Brutus v Cozens [1973] AC 854, though note that this case is often mistakenly taken 
as authority for the proposition, without more, that the meaning of an ordinary word is not a 
question of law.  
67 ATH Smith The Offences Against Public Order (London: Sweet & Maxwell, 1987) 42-43, 
citing two cases in which leaflets were handed out to soldiers, urging them to desert.  Despite 
the similar facts, one was held to amount to ‘insulting’ behaviour while the other was not. See 
also Geoffrey Bindman ‘Incitement to racial hatred’ (1982) NLJ 229, 300, on ‘ordinary 
meaning’ in Brutus v Cozens, above. Bindman argued that the uncertainty this causes had 
discouraged prosecutions.  See also Timothy AO Endicott’s critical commentary in ‘Questions 
of law’ (1998) 114 LQR 292. 
68 FAR Bennion Statutory Interpretation: A Code (4th ed.) (London: LexisNexis Butterworths, 
2002) 1016.  
69 In short, ‘the meaning that correctly conveys the legislative intention’: FAR Bennion 
Statutory Interpretation: A Code (4th ed.) (London: LexisNexis Butterworths, 2002) 381.  There 



 12 

the text - not to construe it with unfettered discretion either for the benefit of the accused or to 
help the state achieve a conviction.   
 
Part 3A will also be governed by human rights law.  Article 10 ECHR states that freedom of 
expression can be limited only when that is objectively necessary to protect, for instance, 
public order in a democratic society. The test of necessity has been glossed in case law and 
is not a matter of discretion for governments alone. States interfering with this right must 
demonstrate a pressing social need and give relevant and sufficient reasons for what they 
do.70 The court has emphasised the importance of ‘pluralism, tolerance and 
broadmindedness’.71  Freedom of religious expression too is protected by Article 9 ECHR and 
that applies to atheists and critics of religions as much as believers. Any limits must be 
proportionate to the aim of achieving one of the public goods listed in the Convention.  All this 
will limit the scope of Part 3A. 
 
If I make what seems like an obvious point, it is because it has not always been fully grasped                              
by legal commentators, who can appear to argue that the courts will do as they may with the 
law.  Andrew Myers in the New Law Journal in June 2005 made the extraordinary claim that  
 

‘a (true) report about Hindu extremists burning a bus full of children on their way 
to Sunday school in India, or a BBC documentary [which] draws attention to the 
discrimination of women in countries that have adopted Sharia law … [is] likely to 
stir up hatred of Hindus and Muslims respectively, and thus be crimes’.  He 
added that ‘If the Bill is passed, WH Smith will commit a crime every time a Bible 
or Koran is sold in its shops.’72   

 
This is simply wrong. This reading of the provision (even in its original draft at first reading) 
could not possibly be endorsed by any responsible British court. Denmark has been held in 
breach of the Convention for convicting a documentary-maker of aiding and abetting the 
dissemination of a racist statement. In a radio interview, members of a far-right racist 
organisation used foul racist abuse. Despite the exceptionally offensive language, the 
European Court of Human Rights stressed that ‘freedom of expression constitutes one of the 
essential foundations of a democratic society.’73   
 
The Racial and Religious Hatred Bill’s most formidable parliamentary critic, Lord Lester, 
stated in debate that there was not ‘the slightest doubt’ that reading from a sacred text of one 
of the main world religions could not in itself amount to an offence under the Bill, even in its 
original draft.74 The analogy with selling a sacred text is clear.  But even without the guidance 
of the ECHR decision or the observations of a senior legal figure, it is wrong to suggest that a 
court could have correctly interpreted the law in this way. The British judiciary have a 
chequered record in protecting civil liberties but this does not mean that legislation can 
impliedly empower them to convict BBC documentarians of religious hatred for fair reporting 
of world events. And it surely goes without saying that no British court would have convicted 
booksellers for selling bibles.  
 

                                                                                                                                                        
may be real doubt over what the legal meaning is – even when the literal meaning seems 
clear - but the range of possible meanings is limited. My thanks to Francis Bennion for 
explaining his concept to me (private correspondence).      
70 Handyside v UK (A/24) (1979-80) 1 EHRR 737, 753-754. 
71 Handyside, ibid 754.  
72 Andrew Myers ‘A crime to tell the truth’ (2005) 155 NLJ 957. 
73 Jersild v Denmark (A/298) (1995) 19 EHRR 1 at [31]. 
74 HL Deb 24 January 2006 vol 441 col 1068.  This refers solely to reading aloud from one of 
these texts, in the absence of any other activities inciting hatred: such a reading could of 
course contribute to an offence if it was used as part of those activities.  On the other hand, it 
is possible that the offence as the government originally framed it could in fact have caught 
someone who read solely from a sacred text, without intent to stir up hatred, in an inflamed 
situation.  Nevertheless, as the offence now requires proof of purposive intention, it is difficult 
to imagine a situation where someone could be unjustly convicted in such circumstances.  



 13 

Even so, the complaints in their political shape have force.  Fears of a ‘chilling effect’ are not 
just rhetoric. And it was disingenuous for government spokespersons to claim that the first 
draft of the Bill would be construed strictly by judges.75 As Bennion has emphasised, it is often 
supposed that criminal statutes are to be construed strictly.76 In fact it is not the job of the 
court to approach a clear statute with the presumption that the defendant is to be set free if at 
all possible.  If a statute is clear, it is to be interpreted to achieve its purpose, and this is as 
true of criminal statutes as it is of civil ones.  A statutory provision with penal effect is to be 
interpreted narrowly only if there is real doubt over what it means. So, if  governments decide 
during the passage of a penal Bill that judges are to construe it strictly, they have two choices: 
they can word the Bill so that it directs the judges to do so, or they can leave the Bill so 
ambiguous that the judges are forced to. Unfortunately the government chose ambiguity. 
  
While it is a central principle that a person should not be penalised except under clear law, it 
is not the only criterion, and it is certainly not a licence to patch up deficiencies which were 
pointed out to Parliament during the passage of the Bill.  Armed by the interpretative rule in 
section 3 of the Human Rights Act 1998, a judge can indeed adopt a secondary interpretation 
which protects human rights, but the Joint Human Rights Committee had already stated that 
in its view the proposed legislation was human rights-compliant.77 It was misleading to argue 
that judges would repair any flaw. And the paramount concern in providing for the accurate 
interpretation of any piece of legislation must be to draft an adequate bill in the first place - not 
to add savings, make parliamentary statements and leave judges with an unnecessary duty to 
clarify it.  
 
Furthermore, the presumption that new legislation is to be interpreted in the light of related 
law may not prove particularly helpful in interpreting Part 3A.  This would limit interpretation, 
but not enough to relieve all reasonable fears.  First, Part 3A is separate from the provisions 
on incitement to racial hatred and the wording used is not identical. Although it is part of ‘the 
family of all public order offences dealing with hatred against groups of people’,78 there is 
greater scope for interpreting the statutory language differently.  
 
Second, precedent on racial hatred states that the word ‘hatred’ is to be given its ordinary and 
natural meaning (the usual presumption unless there is doubt).  There is little evidence that 
key people have thought about the need to distinguish the word ‘hatred’ from its colloquial 
connotations. During the passage of the Bill, Government spokespersons stated that there 
was no need to include a definition of the word because ‘it is a word that a jury could perfectly 
well understand.’79  In most areas of criminal legislation this is indeed acceptable practice. But 
while it is common for penal legislation to use terms from everyday speech, this particular 
area of crime policy is riddled with a loose use of language.   
 
The term ‘hate crime’ is unsuitable to describe the statutorily aggravated offences in the UK, 
yet has become the preferred label.  Perhaps because of this commentators regularly use 
‘hatred’ as a synonym when discussing the provisions of statutes which enhance sentence 

                                                      
75 It was suggested that this would be achieved by guidelines issued at a later date. Charles 
Clarke MP HC Deb 21 June 2005 vol 435 col 679; Paul Goggins MP, HC Stand Committee E 
28 June 2005 vol 435 col 38; 30 June 2005, col. 98; HC Deb 11 July 2005 vol 436 col 599 
and 31 January 2006 vol 442 col 192; Lord Bassam of Brighton HL Deb 11 October 2005 vol 
437 col 280.  Reference was also made by Lord Mackay of Clashfern to a Home Office 
memorandum on guidance in HL Deb 25 October 2005 vol 438 cols 1095 and 1125, but this 
is not now available. 
76 FAR Bennion Statutory Interpretation: A Code (4th ed.) (London: LexisNexis Butterworths, 
2002)  705-709. 
77 Fourth Report of Session 2004-05, Scrutiny: First Progress Report, HL 26, HC 224 para 
1.129. Eighth Report of Session 2004-05, Scrutiny: Fourth Progress Report, HL Paper 60, HC 
388 para 2.64. Fourteenth Report of Session 2004-05, The Convention on the Elimination of 
Racial Discrimination para 94.  These comments referred to similar provisions in earlier Bills. 
78 Lord Lester HL Deb 24 January 2006 vol 441 col 1075.   
79 Lord Falconer of Thoroton HL Deb. 11 October 2005 vol 437 col 163-164. 



 14 

where an offence is aggravated.80 This is useful shorthand for a complex topic, used to make 
the law clearer to the passing reader, but its results are unfortunate. Offences aggravated by 
racial hostility can include incidents in which offensive racist language was used, whether or 
not the defendant intended them to have a racist impact. They are wider in scope than the 
concept of a ‘hate’ crime implies. A violent youth might use a foul racial epithet while 
committing an offence for other motives. He need not feel racist hatred. It may be enough that 
the mens rea is proven for the basic offence and that there be evidence that he abused the 
victim using racist language.  In DPP v Pal, Simon Brown LJ stated:  
 

‘... I would wish to make it perfectly plain that I will reject also an argument put 
before us ... to the effect that section 28(1)(a) has no application to ... street 
arguments when insults may be thrown without thought being given to whether 
the same are racially abusive, which conduct is already covered by other 
offences’. 81   

 
Beyond that, the evidential burden is on the defendant to show that he or she did not intend 
those words to be racially hostile.  In DPP v Green, Rafferty J said:  
 

‘Section 28(1)(a), as distinct from (b), creates a racially aggravated offence 
without the requirement to prove a racist motive. Disposition at the time is 
irrelevant.’82 

 
It is ‘everyday racism’ which such law captures, not only the pre-meditated obsessions of a 
few racist activists. Conviction of a crime of ‘hatred’, in contrast, carries connotations of 
premeditated activity, such as that associated with far-right racist organisations. 
 
There are good arguments for punishing careless racial slurs. In a society where racism is rife 
and certain groups are repeatedly targeted to their fear and detriment, use of racial epithets 
during an offence can be inimical to public order and community relations, beyond any 
damage caused by the basic offence. The state can declare that it ‘takes racism seriously’ by 
enhancing sentence, and such laws may deter some citizens from open expressions of 
prejudice.83 
 
There are serious implications, though, if the public and employers do not understand what it 
is that the legislation punishes. If many think that a conviction for an aggravated crime is proof 
of an act of ‘hatred’, they will attach a tag of pre-meditated bigotry to convictions for 
something else. If the offenders are characterised as purveyors of extreme forms of racism, 

                                                      
80 There are many examples in the press (see eg The Guardian 2 October 2003; The 
Scotsman 1 December 2004, The Herald 22 November 2004; The Herald 23 March 2005; 
The Herald 21 January 2006).  There are also examples in parliamentary debates (HC Deb 
31 January 2006 vol 442 col 233) and government reports (see eg the Select Committee on 
Religious Offences Religious Offences in England and Wales vol 1 HL  95-I  (2003) para 2.12; 
Working Group on Hate Crime, Consultation Paper, (Scottish Executive: The Stationery 
Office, 2003) para. 1.5; NIO Consultation Paper Race Crime and Sectarianism Crime 
Legislation in Northern Ireland, 2002, 6; also in the Summary Paper, 3).  The first report does 
state the law precisely in Appendix 5, as does the second report in its footnote.  There are 
also several examples in academic discussions. 
81 DPP v Pal 2000 WL 191187 (DC) at [14]; [2000] Crim LR 756. 
82 DPP v Green [2004] EWHC QB 1225 at [16]. 
83 On the other hand, in an interesting article, Paul H Robinson and John M Darley argue that 
criminal legislation often does not act as deterrent, and further argue that there is a 
responsibility to make law known when it is ‘counter-intuitive’, as I suggest this law is.  ‘Does 
Criminal Law Deter? A Behavioural Science Investigation’ (2004) 24(2) Oxford Journal of 
Legal Studies 173-205.  
For a thoughtful and intelligent discussion of arguments for and against such legislation in the 
UK, see in particular Mohammed M Idriss ‘Religion and the Anti-Terrorism, Crime and 
Security Act 2001 [2002] Crim LR 890.  For an insightful discussion, which has not dated, of 
the broader aims of public order law, see ATH Smith The Offences Against Public Order 
(London: Sweet & Maxwell, 1987) ch 1. 



 15 

the message is lost.  Those convicted should be stigmatised for their express acts of abuse, 
not for imagined acts of hatred. 
 
‘Hatred’ is a resonant word and it threatens to dominate wherever racial and religious crime 
are discussed.  Take the conviction of Abu Hamza al-Mazri for both incitement to religious 
hatred and soliciting to murder (among others) in 2006.84 In press commentary on his case, 
the hatred offence was given more emphasis, particularly in headlines.85  I did a small 
experiment by searching Technorati, a blog search engine, for the words ‘Abu Hamza’.  27 of 
100 posts mentioned Hamza’s convictions. Of these, only 11 mentioned soliciting to murder at 
all, while 13 focused solely on incitement to racial hatred.86  While this survey can hardly be 
described as a representative sample of public perceptions, it suggests that some people will 
equate Hamza’s case with Griffin and Collett’s.  If they then compare Hamza’s conviction with 
the jury’s failure to reach a verdict in Griffin and Collett, this may be politically troublesome.   
 
These sorts of misunderstandings are particularly worrying when we consider a recent 
American study of attitudes towards “hate crime” legislation. The researchers found that 
people who opposed “hate crime” legislation were more forceful in trying to get others to 
share their views than were those who supported such legislation.87 If it happens that people 
also tend to behave this way in the UK, then this could heighten the confusion in which Part 
3A and its ilk are already mired.  Popular discussion (in everyday life, not just in public 
debate) may be dominated by hostile discourses in which the opponents are more outspoken 
than the supporters. 
 
What proportion of the British public have anything like an accurate understanding of hostility 
and hatred as terms of art in criminal law?  Yet, without this, how will recruiters know why they 
are excluding someone from an appointment,88 or foreign immigration officers know why they 
are refusing a British person a right of residence abroad?  How will the public be able to 
debate free speech cases if people misguidedly compare the outcomes of trials for different 
crimes? The government must address this before the calls for prosecutions under Part 3A 
come flooding in. 
 
 
Compliance with Convention rights 
 
The most vociferous attacks on the Bill concerned freedom of expression. This is a right 
imbued with such totemic power in the contemporary British press that a reader might be 
forgiven for thinking that cities might fall and economies slide into the ocean before the British 
public would countenance the silencing of a single voice. The various views of British people 
are not of course represented in media ideologies. But a government should not 
underestimate the power of the speech trump card. 
 
The problem is that the freedom of expression claim is not as straightforward as it seems. As 
a political argument that the legislation could create a ‘chilling effect’ which scares the 
layperson to silence, it has power.  As a legal argument, however, there are two difficulties 
with it.   
 

                                                      
84 BBC News 7 Feb 2006 http://news.bbc.co.uk/1/hi/uk/4690224.stm  
85 A few also emphasised it in the body of their reports: see eg Philip Johnston The Telegraph 
8 February 2006; Paul Owen and agencies The Guardian 9 February 2006; Gary Younge The 
Guardian 4 Feb 2006; Daily Mail 12 January 2006. 
86 www.technorati.com , accessed 7 March 2006.  Four posts were in another language which 
I did not translate; two were repeat posts and two were inaccessible.  I counted only those 
posts in which people discussed Hamza’s convictions in their own words.   
87 Edward Dunbar and Andres Molina ‘Opposition to the Legitimacy of Hate Crime Laws’ 
(2004) 4(1) Analyses of Social Issues and Public Policy 91.  
88 The Solicitor General for Scotland, for instance, remarked in committee that the impact of a 
conviction for a religiously-aggravated offence ‘on future job applications and emigration 
would be significant’:  Scottish Parliament Report, Justice 2 Committee, 10 December 2002, 
col 2430. The implication for a conviction for incitement to religious hatred in England is clear. 



 16 

First, the parliamentary Joint Committee on Human Rights said repeatedly that even the 
earlier, broader version of the Bill was unlikely to contravene Convention rights. A reconvened 
Committee did change its position and say of the equivalent provisions in the original version 
that the new offences ‘could arguably have an adverse effect on free speech’ in terms of 
Art.10 unless amended ‘to make specific reference to advocacy of religious hatred that 
constitutes incitement to hostility, violence and discrimination’.89  The Committee did not 
expand on this but it probably meant (a) a need to prove purposive intention; (b) a wish to 
restrict the offences to words or behaviour which are threatening (leaving out those which are 
abusive or insulting); and (c) a legislative guarantee that this would not interfere with 
legitimate debate. But ‘arguably’ does not suggest that the Committee felt certain, and at best 
there is disagreement on this point. Given the safeguards in UK law, it is unlikely that any 
conviction would have breached Convention duties.  Since the legislation has been narrowed 
down even Lord Lester, the most outspoken member of the Committee, has declared himself 
content.90 The more likely outcome is not an ECHR appeal, but the complete failure of the 
legislation to achieve any convictions for anti-Muslim hatred at all. The UK has repeatedly 
been criticised for not adjusting British criminal law to cover religious hatred as a surrogate for 
race.91 
 
Second, the UK may already be in breach of Convention rights. Both its approach to ‘racial’ 
hatred and the English common law of blasphemy appear incompatible with the Convention.  
Assumptions about the interpretation of ‘racial’ hatred mean prosecutions will usually only be 
brought where there has been incitement to hatred against the ‘mono-ethnic’ faith groups. 
Also, common law on blasphemy and blasphemous libel92 protects only Christianity.  Both 
these distinctions raise a question under Article 14 ECHR.  Article 14 requires that the 
Convention rights be secured without discrimination on a range of named grounds, which 
include race and nationality. It is justifiable to discriminate only where this is done objectively 
for a reasonable purpose and is no more than is necessary to achieve that purpose. Both the 
narrow interpretation of ‘racial’ and the selective protection offered by blasphemy seem 
arbitrary in the light of racial equality.  What are the reasonable purposes which justify these 
distinctions?  Furthermore, as Part 3A will not apply to Scotland, there will be a difference 
between the two jurisdictions. However, as Northern Ireland has long had separate provision 
on incitement to religious hatred which differs from British law, this may not itself found a case 
on discrimination. 
 
On blasphemy, there is Article 9 ECHR. This protects freedom of thought, conscience and 
religion. The Commission has held in Kokkinakis v Greece that this includes the right to 
express religious beliefs and to proselytise,93 as part of the right of freedom of expression 
under Article 10.  However, the decision was not unanimous and Mr Kokkinakis’ manner while 
proselytising was ‘inoffensive’. It is not clear how more robust religious rivalry would be 
treated. Even so, Art.5 ECHR permits punishment only when ‘prescribed by law’, and the law 
of blasphemy was described by the Law Commission over twenty years ago as unacceptably 
vague and broad.94 The European Court of Human Rights did hold in Wingrove v UK that the 
high threshold test for blasphemy set by the English courts was a safeguard against 

                                                      
89 HC/HL First Report (2005-06) para 5.2.  
90 HL Deb 24 January 2006 vol 441 col 1075. 
91 Concluding observations of the Committee on the Elimination of Racial Discrimination: 
United Kingdom of Great Britain and Northern Ireland CERD/C/63/CO/11 para 21. The 
European Commission against Racism and Intolerance also criticised the distinction which is 
made in UK law between ‘racial’ and ‘religious’ groups: Second Report on the United 
Kingdom, CRI (2001) 6 para 43; General Policy Recommendation N°7: National legisla tion to 
combat racism and racial discrimination CRI (2003) 8 para IV 18(d); and Third Report on the 
United Kingdom, CRI (2005) 27 para 14.   
The Select Committee on Religious Offences has pointed out however that the problems are 
now mostly addressed by other law: Report of the Select Committee on Religious Offences 
Religious Offences in England and Wales (2003) HL  95-I, paras 95-98. 
92  See ch 3 of the Report of the Select Committee on Religious Offences, ibid. These 
offences no longer exist in Scotland. 
93 Kokkinakis v Greece (A/260-A) (1994) 17 EHRR 397.   
94 Law Commission Offences Against Religion and Public Worship (1985). 



 17 

arbitrariness.95 The legal advice which informed Wingrove has however been criticised96 and 
although the matter is not settled, the balance of arguments suggests that English blasphemy 
law is incompatible. 
 
The legal position is not simply that the UK has an imperfect body of law which has been 
criticised by international observers. It may have an enforceable obligation to change its law.  
It needed to act on incitement to religious hatred, and it may have to abolish the law of 
blasphemy.  Whether Part 3A is the best way to achieve the former remains to be seen. 
 
 
Does freedom of expression need special protection? 
 
In response to fears, the Lords added to the Bill a saving explicitly protecting freedom of 
expression, now found in section 29J: 
   

‘Nothing in this Part shall be read or given effect in a way which prohibits or 
restricts discussion, criticism or expressions of antipathy, dislike, ridicule, insult 
or abuse of particular religions or the beliefs or practices of their adherents, or of 
any other belief system or the beliefs or practices of its adherents, or 
proselytising or urging adherents of a different religion or belief system to cease 
practising their religion or belief system.’ 

 
This should have been unnecessary, given the emphasis on freedom of expression in prior 
law and given that Part 3A is now worded more restrictively than the law on racial hatred 
because it leaves out ‘abuse’ and ‘insult’ and confines the criminal act to ‘threat’.  Lord Lester 
called the new saving ‘a unique addition to the criminal code,’97 but that is hardly a 
compliment.  It adds a further layer to the legislation. While the presumption is that savings 
are liberally construed, any unforeseen omission will still raise questions which will need to be 
resolved in court. It is also not the purpose of a saving to narrow a provision further, as Lord 
Lester appeared to intend.98  This is not good drafting.  And even if it did achieve what Lord 
Lester desired, it would give Article 10 ECHR too high a priority over Article 9.   
 
On the other hand, the Bill’s defenders were unconvincing too. On safeguards, Home 
Secretary Charles Clarke protested that ‘a complaint must be accepted by a police officer’,  
‘the police must investigate and acquire evidence’ and ‘the case must be referred to the 
Crown Prosecution Service, which will assess it by applying evidential and public-interest 
tests’.99 All are routine requirements. More significant is that the Attorney General must give 
consent for prosecution,100 and usually the Director of Public Prosecutions will too. But 
Bennion rails against laws which remove the right of private prosecution.101 
 
Charles Clarke listed several more ‘safeguards’ which resembled the normal criminal justice 
process, but he also noted that the case must go before a jury. This in fact is one of the 
problems in interpreting the legislation.  Viscount Colville of Culross pointed out that juries do 
not give reasons, so to ensure the law is applied accurately, and to furnish an appeal, much – 
perhaps too much - depends on what the judge said in summing up.102  Furthermore, the 

                                                      
95 (17419/90) (1997) 24 EHRR 1 at [61]. 
96 Report of the Select Committee on Religious Offences Religious Offences in England and 
Wales (2003) HL  95-I at paras 12, 14 and 40.  See also ECRI Second Report on the United 
Kingdom (CRI, 2001) 6 para 43, in which the blasphemy law is described as discriminatory. 
For a recent critique of the blasphemy laws, see FAR Bennion ‘A New Reason for Blasphemy 
Reform’ (2005) JP 169, 609, 629. 
97 HL Deb 24 January 2006 vol 441 col 1075. 
98 HL Deb 25 October 2005 vol 438 cols 1076-1077.  On savings, see FAR Bennion Statutory 
Interpretation: A Code (4th ed.) (London: LexisNexis Butterworths, 2002) 619. 
99 Paul Goggins HC Standing Committee E 30 June 2005 col 105. 
100 At 29L.  
101 FAR Bennion ‘Gilding the Lily on Religious Hatred’ (2005) 14(1) The Commonwealth 
Lawyer 35. 
102 HL Deb 11 October 2005 vol 437 cols 198-199. 



 18 

House of Lords recently emphasized in Wang that a jury is entitled to direct a verdict of guilty 
where there is some evidence to support it, however perverse that verdict may seem to be.103 
 
Clarification in courts of appeal would help, but it has been argued specifically in defence of 
the new law that few people have been prosecuted for incitement to racial hatred and that few 
prosecutions are expected under the new law, because its aim is to deter.104  Perhaps, then, 
the ‘freedom of expression’ saving is necessary in this complex area of law.  Ideally, a well-
drafted provision should be narrow enough that any remaining doubt would be dispelled by 
reference to Convention rights and precedent on racial hatred. But as lay juries must reach a 
verdict ‘in all the circumstances’, it is probably better that a saving was included.   
 
 
Silencing hate speech: the evolution of signifiers 
 
This brings us back to lay conceptions of the new law. Will the disquiet which enveloped the 
Bill now deaden public debate? I mentioned above that the racial hatred law has led to few 
prosecutions. Government spokespersons claim that it has been a deterrent105 but it is 
dangerous to assume that the legislation has silenced the right people. Together with the 
public uncertainty over the scope of Part 3A, this lack of prosecutions and convictions may 
enhance, not dispel, the chilling effect.  In the context of the Anti-terrorism, Crime and 
Security Act 2001, Adam Tomkins says it ‘is the splintering of terrorism from ordinary criminal 
justice that results in human rights abuses’.106 Without cases to adjudicate upon, bad law is 
not even judicially clarified, let alone exposed to judicial review.   
 
Yet talk of the chilling effect within religious debate can be deceptive. This claim is most 
familiar in debates about terrorism, where it makes more sense. Supporters of small terrorist 
organisations rarely write for national newspapers.  Tamil Tigers have not been testing out the 
limits of their rights of free speech under the Terrorism Act 2000107 in the pages of the 
Guardian or the Times.  The limits of incitement to religious hatred, however, will quickly be 
tested by mainstream commentators. The BNP website will soon demonstrate the furthest 
reaches of the nasty but lawful.  Terrorism and religious debate should not be carelessly 
compared. 
 
The key problem may not be the chilling effect, but the difficulty of constructing a strong case 
to take to trial.  We saw above that a major justification for creating Part 3A is that extremist 
organisations specialise in pushing the limits of the law. Yet it is obvious that racist 
organisations will adapt to the new law just as they did to the old.  
 
Hatred is a subset of obnoxious speech.  That criminal law permits most of us to be rude and 
nasty most of the time is simply one aspect of our freedom to tell glaring, provably 
nonsensical, lies. The duty to truth is imposed only on relatively few professionals speaking 
within their sphere of authority. Tackling particular aspects of incitement to hatred is always 
going to be problematic for the criminal law, because freedom of expression permits people 
not just to present a biased version of the truth but also to tell dangerous lies. If the link 
between the lie and the harm is too indirect, no offence is committed. Incitement to racial 
hatred and now incitement to religious hatred are statutory exceptions to the general rule that 
the law permits people blatantly to mislead each other.  
 
The trial of Nick Griffin and Mark Collett for incitement to racial hatred was hailed as proof that 
the existing law catches religious hatred.  It is not.  The case could only be brought because 
Griffin and Collett were secretly recorded in private and so spoke more openly than they 

                                                      
103 R v Wang [2005] 1 WLR 661, [2005] EWCA Crim 3238.   
104 Paul Goggins HC Deb 21 June 2005 vol 435 col 759. 
105 Paul Goggins HC Deb 21 June 2005 vol 435 col 759. 
106 Adam Tomkins ‘Readings of A v Secretary of State for the Home Department’ [2005] 
Public Law 259, 265. 
107 The Liberation Tigers of Tamil Eelam (the ‘Tamil Tigers’) are a proscribed organisation 
under Sch.2 of the Terrorism Act 2000: see The Terrorism Act 2000 (Proscribed 
Organisations) (Amendment) Order 2001 SI 1261.  



 19 

would in public. Their speeches contain several references to Asians and ‘Whites’. Without 
these, there would have been insufficient evidence to prosecute.  If the retrial succeeds, it will 
not demonstrate that anti-Muslim hatred could have been captured within the racial hatred 
law, as some have claimed. It will simply show that careless crypto-racists can get caught.   
 
When only some types of incitement to hatred are made an offence, and when some types of 
abuse are identified as worse than others by the use of specified words and symbols, bigoted 
ingenuity quickly adapts.  In Scotland, where sectarianism at football matches has become a 
regular form of ritual sparring, the law of religious aggravation has had to be made more 
sophisticated, and thus wider.  
 
Section 74 of the Criminal Justice (Scotland) Act 2003 mostly copies the English provisions 
on religious aggravation. The Scottish legislation, however, has some additions. The English 
law specifies in section 28 of the Crime and Disorder Act 1998, as amended, that the 
aggravation is founded on ‘hostility based on the victim's membership (or presumed 
membership) of a racial or religious group’.  Membership can include association with 
members of that group. A ‘religious group’ means ‘a group of persons defined by reference to 
religious belief or lack of religious belief’.  
 
The Scottish equivalent provides that the aggravation is founded on ‘malice and ill-will based 
on the victim's membership (or presumed membership) of a religious group, or of a social or 
cultural group with a perceived religious affiliation’108 which as in England and Wales includes 
association with members of that group.  A religious group means: 
 
‘a group of persons defined by reference to their-  

(a) religious belief or lack of religious belief; 
 
But it also covers: 
 

(b) membership of or adherence to a church or religious organisation; 
(c) support for the culture and traditions of a church or religious organisation; or 
(d) participation in activities associated with such a culture or such traditions.’109 

 
The Scots provision is broader because of the kinds of offence it aims to tackle.  Scottish 
sectarianism is a complex subject, but sectarian hostility (or ‘malice and ill will’) was seen by 
the Scottish Parliament as associated most frequently with football rivalry. This is expressed 
by oblique linguistic associations and referential symbols. For an outsider, this symbolism is 
bewildering. Rangers fans have thrown potatoes onto the pitch to annoy the Catholic/Irish-
identified Celtic fans,110 while their opponents have worn German flags as a provocative 
celebration of the new Pope.111 Both sides have adapted apparently harmless songs such as 
the traditional ‘Fields of Athenry’ and Tina Turner’s ‘Simply The Best’. Now, singing even the 
harmless lyrics is inflammatory.   
 
Football fans, in close contact with both fellow supporters and their rivals, develop exquisitely 
arcane references which can be interpreted, accurately, as sectarian insults. The British 
National Party cannot draw on such close interaction between itself, its members and its 
targets and so has less opportunity for such sophistication. Even so, bigotry is fluid and fast-
changing and new terms of abuse spread fast. Allusions are easily understood. Recently a 
Scottish court official was suspended after allegedly sending an email describing Irish people 
as ‘green scum’.112 The statutory convolutions in which the Scottish Parliament wound itself to 
tackle sectarianism are instructive for those tackling expressions of religious hatred in 
England and Wales. It seems easier to gather evidence against minority religious preachers 

                                                      
108 s 74(2)(a) – my emphasis added. 
109 s 74(7) – my emphasis added. 
110 Devil’s Advocate Sunday Herald August 14 2005. 
111 Alan Pattullo The Scotsman 25 April 2005. 
112 Douglas Fraser The Herald February 1 2006.  



 20 

than against racist far-right activists.  Nick Griffin, after all, denies that he is a racist.113  Can 
Part 3A meet the challenge? 
 
 
The problem of proving intention 
 
The final question then is: can Part 3A work?  Branch after branch has been hacked off. Lord 
Lester insisted that the Government’s original proposals suffered from ‘the twin vices of 
vagueness and uncertainty and over-breadth and a lack of proportion’.114 The Lords then took 
out the ‘likely to’ test and limited the offence to ‘threatening’ words, material or behaviour. 
They aimed to create a statute which would do precisely the job for which it was designed, 
and no more.  
 
But that ‘likely to’ test so criticised by the Lords was inserted in Britain’s incitement to racial 
hatred provisions in 1976,115 precisely because research had found that the need to prove 
purposive intention116 was so restrictive that prosecutions were likely to fail.117  So, Parliament 
broadened the offence. No longer was there a need to prove intent to stir up hatred: it was 
enough that it was ‘likely to’ be stirred up.  
 
Despite this, no floodgates opened. Peter Thornton noted that there were few prosecutions 
under section 5A, and they rarely succeeded.118  In 1986 Parliament revised the law again: 
this time the prosecution could prove either intent or likelihood. It was still not enough.  
Attorneys General feared that high-profile cases would fail and were reluctant to consent to 
prosecutions.   
 
One might think that a religious hatred provision would be different. But between 1970 and 
1987, Northern Ireland had just such a provision on incitement to religious hatred - with a 
purposive intention test just like that in Part 3A.  And, as Brigid Hadfield pointed out in 1984, 
the wording of the Northern Irish provision was ‘such as to render highly unlikely a successful 
prosecution under it’.119 Three years later Parliament re-wrote the Northern Irish provisions to 
include an alternative ‘likely to’ test, based on the British offence.120 Yet even though the 
Northern Irish law encompasses actions which ‘arouse fear’ as well as those which stir up 
religious hatred, records show just one conviction so far.121   
 
Part 3A covers a wider range of actions than the 1965 Act did (particularly actions in private 
dwellings) but the threshold test of intent is the same.  It will be possible to infer intent, even if 
the person denies it.122 But the prosecution will probably have to prove the purposive intention 

                                                      
113 BBC news report February 3 2006. 
114 HL Deb 25 October 2005 vol 438 col 1074.  
115 The new provision, inserted by s 70 Race Relations Act 1976, became s 5A of the Public 
Order Act 1936.  It was repealed by s 40(3), Sched. 3 Public Order Act 1986. 
116 On purposive intention, see Richard Card (2000) Public Order Law (Bristol: Jordans, 2000) 
139 and 187. 
117 Home Office Racial Discrimination Cmnd 6234 (1975) at para 126.  CERD praised the 
change: see Ineke Boerefijn and Joanna Oyediran in Sandra Coliver Striking a Balance  
(Essex: Article 19, 1992) 27.  
118 Peter Thornton Public Order Law (London: FTPL, 1987) 61.  
119 Brigid Hadfield ‘Incitement to religious hatred’ [1984] PL 231, 241. On the other hand, 
Patricia M Leopold (‘Incitement to hatred – the history of a controversial criminal offence’ 
[1977] PL 389, 401-402) rather dismissively claims that this was less a flaw of the legislation 
and more that ‘a Northern Irish jury was unable to be objective about a case of incitement.’ 
120 The Public Order (Northern Ireland) Order 1987, Part 3 (ss.8-17).  The change followed 
criticisms that the earlier provisions were ineffective: see T Murphy in Sandra Coliver Striking 
a Balance  (Essex: Article 19, 1992) 267.  
121 HL Deb 3 March 2005 vol 431 col 34 (WA).  The data date back only to 1993, as there are 
no records available before that.  The previous legislation only led to one (failed) prosecution 
between 1970 and 1992: see T Murphy in S Coliver ibid. 
122 See Anthony Lester and Geoffrey Bindman on intent in the 1965 Act: Race and Law 
(Harmondsworth: Penguin, 1972) 362.  



 21 

to stir up religious hatred.123  What is worrying that both Hadfield and Thornton mention cases 
where no conviction was ever achieved – not because the police did not investigate, or 
because the Attorney General refused to allow a prosecution, but because the test of 
purposive intention was too difficult to surmount.124   
 
It will be even harder to get a conviction under Part 3A, because it permits abuse and insult. A 
person will have a defence under the freedom of expression saving if he intended ‘only’ to 
insult or express antipathy.  In the example above from the BNP website, it is unlikely that the 
comments would amount to words which are ‘threatening’.125  Also, the authors could argue 
that they are expressing antipathy towards the religious practices of some Muslims who find 
authority for their actions in the Koran, but not all Muslims. We may suspect otherwise, but as 
the article links passages of the Koran to the actions of ‘some’ followers, a jury could decide 
that the article was protected by section 29J. The website does elsewhere disavow both racist 
and anti-Muslim intent126 and this might persuade a jury not to convict.127 So if the BNP 
website can continue publishing such comments, what comfort can Muslims draw from the 
new law? 
 

‘The second clip has another woman relating how another Muslim has told her 
daughter “Us Pakis are going to have to watch what we’re doing with you young 
’uns now.”  
Together the clips make it crystal clear that for me to have used the phrase “Paki 
street thug” to describe a specific type of swaggering young scumbag – 
equivalent, as I told the jury on Friday, to the “hoodies” and “white trash” who 
besmirch our own community – was not intended to be, and could not possibly 
be taken to be, any kind of racist slur or insult. In Keighley, as in fact all over 
working class Britain, the word is a simple, neutral, descriptive term, used by all 
communities.’ 128 
 

To back up their claim that they were not using abusive language, Griffin and Collett were 
allowed to show the jury video extracts of a documentary which alleges that South Asian 
people refer to themselves as ‘Pakis’. This was supposed to demonstrate that the word is 
inoffensive. Bindman has pointed out that the 1965 test also allowed wily defendants to 
produce dubious material in the courtroom, supposedly to prove their innocent intentions.129  
For incitement to racial hatred today, proof of intention is not essential, but for religious hatred 
it is.  Prosecutors will fear bringing a case where purposive intention has to be proved, if it can 
become a showroom for arguments like this.   
 
The legislative history in Britain and Northern Ireland suggests that the new Part 3A will be 
almost unenforceable. Without a confession, it will be very difficult to prove the purposive 
intention. Perhaps some extremist clerics who confess their intentions may be caught, which 
would be no bad thing; and much more legitimate than deporting them without trial. But those 
at whom the legislation was supposedly aimed – racist activists – will have little trouble 
adapting their rhetoric. The Lords have pruned this statute so hard they have left it a stump. 
Minority faith leaders will see this as a betrayal.  
 
The only way in which we can have genuinely free expression is to protect those minorities 
who would also like to exercise it.  A society does not have freedom of expression if white 

                                                      
123 See Richard Card Public Order Law  (Bristol: Jordans, 2000) 87. 
124 Brigid Hadfield ‘Incitement to religious hatred’ [1984] PL 231, 242; Peter Thornton Public 
Order Law (London: FTPL, 1987) 61. 
125 The word has been given its ordinary meaning: Brutus v Cozens [1973] AC 854.   
126 http://www.bnp.org.uk/candidates2005/manifesto/manf4.htm (‘we are not ‘racists’’) and 
http://www.bnp.org.uk/articles/islamic_menace.htm  (‘we have nothing against Muslims’). 
127 Although not always - see Patricia M Leopold’s discussion of R v Jordan [1967] 1 All ER 
486, in ‘Incitement to hatred – the history of a controversial criminal offence’ [1977] PL 389, 
395. 
128 From Nick Griffin’s online blog 30 January 2006 http://freespeechontrial.blogspot.com/  
129 Geoffrey Bindman ‘Incitement to racial hatred’ (1982) NLJ 229, 300. 



 22 

activists can freely stir up hatred and so ensure that their minority victims become too 
frightened to reply.  
 
But the result of the final amendments to Part 3A is - in law at least - almost entirely a return 
to the status quo.  Its sole effect is likely to be on public perceptions of what is unlawful. The 
greatest barrier to public enlightenment is likely to be pundits who will not be able to resist 
telling tall tales about the silencing of free speech.  Parliament must accept some of the 
blame. Part 3A may be so misunderstood that perhaps it is just as well that it has been made 
useless. But international observers will not stay silent, and the UK may have to answer for its 
errors.