Court of Appeal · Decided 27 May 2011

Shadrake Alan v Attorney-General

Feedback & support

[2011] SGCA 26Case no. Civil Appeal No 212 of 2010

What happened, in plain English

There is no plain-English summary of this judgment yet.

Ask about this judgment

Laws this judgment relies on

Constitution of the Republic of Singapore

Grounds of decision

Official text from eLitigation

Andrew Phang Boon Leong JA (delivering the judgment of the court):
Introduction
[1] This is an appeal against the decisions of the trial judge (“the Judge”) in Attorney-General v Shadrake Alan [2011] 2 SLR 445 (“Shadrake 1”) and Attorney-General v Shadrake Alan [2011] 2 SLR 506 (“Shadrake 2”). In Shadrake 1, the Judge found Mr Alan Shadrake (“the Appellant”) in contempt of court for eleven of the fourteen impugned statements. In Shadrake 2, the Judge sentenced the Appellant to six weeks’ imprisonment and a fine of $20,000 (in default of which, two weeks’ imprisonment, to run consecutively to the first term of imprisonment).
[2] The present appeal raises – in some instances for the very first time before this court – important issues relating to the law of contempt which will need to be clarified before the relevant principles are applied to the facts. Foremost amongst these issues are the fundamental ones relating to the test for liability for contempt of court on the ground of scandalising the judiciary (“scandalising contempt”) as well as what the Judge characterised as the defences to scandalising contempt.
The factual background
[3] This case arose from an application by the Attorney-General (“the Respondent”) to commit the Appellant, the author of Once a Jolly Hangman: Singapore Justice in the Dock (Strategic Information and Research Development Centre, 2010) (“the book”), for contempt of court in relation to certain passages contained in the book.
The decision below
The decision on liability
[4] In the court below, the Judge first undertook an extensive and comprehensive survey of the Singapore and Commonwealth case law on the applicable test for liability for scandalising contempt.
[5] In the result, the Judge departed from previous decisions of the Singapore High Court with regard to the test for the actus reus in cases of scandalising contempt. The Judge did so in holding that the “real risk” test (viz, that the impugned statement must pose a real risk of undermining public confidence in the administration of justice before it is held to be contemptuous), in contradistinction to the “inherent tendency” test, was to be applied in Singapore (see Shadrake 1 at [50]).
[6] In so far as the test for mens rea was concerned, the Judge held that the only mens rea needed for finding liability was that the publication of the allegedly contemptuous statement was intentional; it was not necessary to prove an intention to undermine public confidence in the administration of justice (see Shadrake 1 at [55]).
[7] Turning to what the Judge characterised as the defences to scandalising contempt, the Judge considered that the defences of justification and fair comment in the law of defamation were not applicable (see Shadrake 1 at [59]–[69]), although his position with regard to the defence of justification was, with respect, unclear. The Judge was of the view that the only defence to scandalising contempt was that of fair criticism (see Shadrake 1 at [70]–[76]). According to the Judge, in order to raise a defence of fair criticism, the alleged contemnor must be able to show some objective basis for his contemptuous statements, the cogency of the rational basis required to support his allegations increasing correspondingly with the seriousness of the allegation made (see Shadrake 1 at [72]). The alleged contemnor must also be able to show that the allegation was made in good faith, viz, that the alleged contemnor must genuinely believe in the truth of the criticism he made (see Shadrake 1 at [73]). The Judge further held that the mere fact that the criticism is in outspoken language does not mean that it should necessarily be penalised (see Shadrake 1 at [75]). He also surmised that, contrary to English case law and earlier Singapore cases, there should not be a limit on the kind of criticisms which can be made against the court subject to the above criteria being met (see Shadrake 1 at [76]). We pause to observe, parenthetically, that the preferable approach might be to view the concept of fair criticism as going towards liability rather than as an independent defence – a point which we will deal with in more detail below (see generally at [59]–[86]).
[8] Applying the law to the facts, the Judge held that three of the statements (viz, the third, sixth and twelfth statements, reproduced respectively at [92], [104] and [123] of Shadrake 1) did not amount to contempt, whilst the remaining eleven statements were contemptuous and did not qualify for the defence of fair criticism (see Shadrake 1 at [133]–[136]).
The decision on sentence
[9] The Judge held that imprisonment would be the norm for the author of a publication which scandalises the court (see Shadrake 2 at [26]). This is because to constitute scandalising contempt, a statement must not only pose a real risk of undermining public confidence in the administration of justice but must also fall outside the ambit of fair criticism – the satisfaction of these criteria ipso facto making for a high degree of culpability (see Shadrake 2 at [26]).
[10] Taking into account the high level of the Appellant’s culpability, including his stated intent to repeat his contempt by publishing an expanded second edition of the book; previous sentencing precedents (which were on the low end); the fact that the precise extent of the book’s circulation in Singapore was unclear; the fact that the Appellant was not a person with a credible and established reputation; and the desire to signal that the courts have no interest in stifling legitimate debate on the death penalty and other areas of the law, the Judge sentenced the Appellant to six weeks’ imprisonment (see Shadrake 2 at [42]).
[11] Further, to send a signal that those who hope to profit from controversy by scandalising the court may expect to have their profits disgorged by a stiff fine in addition to other punishment, the Judge imposed a $20,000 fine on the Appellant, with two weeks’ imprisonment in default (see Shadrake 2 at [37] and [43]).
The issues
[12] The main issues are relatively straightforward and are encompassed within the following three grounds of appeal relied upon by the Appellant:
(a) First, that the Judge had erred in his statement of the test for liability for scandalising contempt. In this regard, the Appellant contends that whilst the Judge correctly adopted the label of a “real risk” test, he erred in defining the content of the said test;
(b) Second, that the Judge had erred in his interpretation of the passages held to have given rise to the contempt; and
(c) Third, that the sentence meted out by the Judge was manifestly excessive.
[13] We will proceed to deal with each of these issues seriatim. In dealing with the first issue, we will address not only the test for liability for scandalising contempt but also what the Judge characterised as the possible defences to contempt. This particular issue relates, in substance, to the applicable legal principles. The second issue involves the application of the relevant legal principles to the facts of this case – in particular, the impugned statements. At this point, viz, the confluence of both law and fact, we arrive at the most significant aspect of the present appeal, not least because of the very real effect it will have on the result of the case in general and the Appellant in particular. To this end, we will undertake an analysis of each statement in relation to the relevant legal principles. The third issue involves both the discussion of applicable legal principles in relation to sentencing in the context of scandalising contempt, as well as the application of those principles, if appropriate, to the facts of this case. We use the phrase “if appropriate” because there is, of course, the threshold issue of liability (which falls within the purview of the first and second issues). Put simply, before we can even begin to consider the issue of sentence, liability for contempt on the part of the Appellant must first be established (and it is important to note – and, indeed, it is axiomatic – that such liability must be established by the Respondent beyond a reasonable doubt). If no liability for contempt is established in the first place, then the issue of sentence does not need to be considered. That is why we have left the discussion of the legal principles relating to sentence to the final part of this judgment.
[14] Let us now turn to consider the applicable legal principles relating to the test for liability for scandalising contempt as well as what the Judge characterised as the possible defences available.
The legal principles
Introduction
[15] Although counsel for the Appellant, Mr M Ravi (“Mr Ravi”), focused on the test for liability for scandalising contempt, we will also deal with the possible defences thereto because they are (as we shall see) directly relevant to the facts of the present appeal. As already mentioned, however (at [7]), these possible defences, particularly fair criticism, may be better viewed as going towards the issue of liability rather than constituting independent defences as such (see also the detailed analysis below at [59]-[86]).
[16] Before proceeding to deal with the applicable legal principles with respect to liability for scandalising contempt, it would be appropriate to provide some background on the law relating to contempt.
[17] At the most general level, it should be noted that the law relating to contempt of court operates against the broader legal canvass of the right to freedom of speech that is embodied both within Article 14 of the Constitution of the Republic of Singapore (1985 Rev Ed, 1999 Reprint) (“the Constitution”) as well as the common law. The issue, in the final analysis, is one of balance: just as the law relating to contempt of court ought not to unduly infringe the right to freedom of speech, by the same token, that right is not an absolute one, for its untrammelled abuse would be a negation of the right itself. Indeed, this last mentioned point is embodied in Art 14(2) of the Constitution which provides that “Parliament may by law impose … restrictions designed to … provide against contempt of court”. In this regard, the Singapore Parliament has in fact provided the courts with the jurisdiction to punish for contempt in s 7(1) of the Supreme Court of Judicature Act (Cap 322, 2007 Rev Ed) (see also the Singapore High Court decision of Attorney-General v Lingle [1995] 1 SLR(R) 199 (“Lingle”) at [7]).
[18] Not surprisingly, such an approach (from balance) is adopted in other jurisdictions as well. In the Appeal Committee of the Court of Final Appeal decision of Wong Yeung Ng v Secretary for Justice [1999] 3 HKC 143 (“Wong Yeung Ng (CFA)”), for example, Litton PJ, delivering the judgment of the majority, observed thus (at 147):
[19] Contemptuous acts are generally classified into two broad categories, viz, contempt by interference and contempt by disobedience (see, eg, You Xin v Public Prosecutor [2007] 4 SLR(R) 17 at [16] (“You Xin”)). The former category (viz, contempt by interference) comprises a wide range of matters such as: (1) disrupting the court process itself (eg, contempt in the face of the court, alternatively termed “ex facie contempt”); (2) acts which risk prejudicing or interfering with particular legal proceedings (“sub judice contempt”); and (3) acts which interfere with the course of justice as a continuing process (eg, publications which “scandalise” the court and retaliation against witnesses for having given evidence in proceedings which are concluded) (see generally, eg, You Xin at [16] and the Privy Council decision of McLeod v St Aubyn [1899] AC 549 at 561). The latter category (viz, contempt by disobedience) consists in the disobedience of court orders as well as the breach of undertakings given to the court (ibid).
[20] The present appeal relates to a charge of scandalising contempt which, as noted in the preceding paragraph, falls within the first category, viz, contempt by interference.
[21] Finally, an important point should be emphasised. It is not new but is nevertheless of vital importance because it undergirds the law relating to contempt in general and scandalising contempt in particular. This relates to the very raison d’être of the law of contempt. It is trite that the law relating to contempt of court is not intended to protect the dignity of judges. As Lord Atkin put it in the Privy Council decision of Ambard v Attorney-General for Trinidad and Tobago [1936] AC 322 at 335 (“Ambard”), “justice is not a cloistered virtue” but a public one. An act of scandalising contempt is a public injury rather than a private tort, in so far as it involves undermining public confidence in the administration of justice. In the words of this court in Pertamina Energy Trading Ltd v Karaha Bodas Co LLC [2007] 2 SLR(R) 518 at [22] (“Pertamina Energy Trading”):
[22] Put simply, the fundamental purpose underlying the law relating to contempt of court in general and scandalising contempt in particular is to ensure that public confidence in the administration of justice is not undermined. Indeed, this fundamental purpose – which finds acceptance across all Commonwealth jurisdictions - has been reiterated time and again in the Singapore context (see, eg, Pertamina Energy Trading (cited in the preceding paragraph); as well as the Singapore High Court decisions of Attorney-General v Wong Hong Toy [1983–1984] SLR(R) 34 at [26]; Attorney-General v Zimmerman Fred and others [1985–1986] SLR(R) 476 (“Zimmerman”) at [9]; Attorney-General v Hertzberg Daniel [2009] 1 SLR(R) 1103 (“Hertzberg”) at [20]; and Attorney-General v Tan Liang Joo John [2009] 2 SLR(R) 1132 (“Tan Liang Joo John”) at [11]).
[23] We turn now to the test for liability for scandalising contempt, viz, the actus reus and mens rea for the offence. As the mens rea requirement is well established, this judgment focuses instead on the actus reus. That said, for the avoidance of doubt, the necessary mens rea was succinctly and rightly enunciated by the Judge as follows (see Shadrake 1 at [55]; see also above at [6]):
The test for liability for scandalising contempt
Introduction
[24] Although there are two main tests for liability for scandalising contempt, viz the “real risk” test and the “inherent tendency” test, we note that Mr Ravi has attempted to introduce (in substance at least) a third test (the “clear and present danger” test).
The “real risk” test
(1) The test
[25] The “real risk” test means precisely what it says – that, before a statement can be held to be contemptuous, it must pose a real risk of undermining public confidence in the administration of justice. As the Judge correctly pointed out, a statement which only poses a remote possibility that public confidence in the administration of justice would be undermined would not be held to be contemptuous (see Shadrake 1 at [51]).
[26] The “real risk” test can in fact be traced back to the English decision of R v Duffy and others, ex parte Nash [1960] 2 QB 188 (“Nash”), which related to a situation of sub judice contempt (see also above at [19]). In that case, Lord Parker CJ held that the test was whether there “was … a real risk, as opposed to a remote possibility, that the article [in question] was calculated to prejudice a fair hearing” (see Nash at 200; emphasis added). The “real risk” test was also applied in the House of Lords decision of Attorney-General v Times Newspapers Ltd [1974] AC 273 at 298–299 (“Times Newspapers”), where Lord Reid held that:
[27] With regard to the “real risk” test, the Judge observed as follows in the court below (see Shadrake 1 at [51]–[54]):
[28] In our respectful view, the Judge’s elaboration of the “real risk” test in [51], viz, that “a real risk, as defined in the cases, is not to be equated with a serious or grave risk, but merely something more than a de minimis, remote or fanciful risk” [emphasis added in bold italics], gives rise to possible ambiguity. If what the Judge meant was that a “real risk” was not to be equated with “a serious or grave risk” inasmuch as the latter connotes a “clear and present danger”, this would be consistent with our rejection of the “clear and present danger” test for the reasons set out below (at [38]–[50]). However, it is possible to interpret the Judge’s elaboration in a different way, for he observed in the same paragraph ([51]), citing Lord Reid in Times Newspapers, that “any degree of risk above the de minimis level, including “a small likelihood”, amounts to contempt, with the seriousness of the risk going only to mitigation” [emphasis added in bold italics], and repeated this observation in [77]. If this was the meaning intended by the Judge, it should be borne in mind that there is an argument to the contrary, viz, that “a small likelihood” of risk would not, ex hypothesi, constitute a “real risk” within the meaning of the “real risk” test (cf the approach of Brooke and Dubin JJA in the Ontario Court of Appeal decision of R v Kopyto (1988) 47 DLR (4th) 213 (“Kopyto”), where a “serious risk” test was endorsed. Kopyto is discussed in greater detail below at [43]-[44]). While it is uncertain as to which meaning the Judge intended to attribute to the requirement of “real risk”, it is clear that his elaboration in [51] was based on a recognition of the signal importance of ensuring that public confidence in the administration of justice is not undermined.
[29] In the final analysis, although the two contrasting approaches toward the “real risk” test described briefly in the preceding paragraph are interesting, we are of the view that it would be not be helpful to attempt an elaboration (especially a theoretical one) of the “real risk” test. We state this with great respect to the Judge but the brief analysis above demonstrates the very real danger of semantic analysis trumping practical factual considerations. The difficulty in seeking to elaborate on this particular test in the abstract is obvious once we recognise that the test cannot be divorced from the factual context of each case (which inevitably admits of innumerable permutations). Each case ultimately turns on its own particular facts and whether there is a “real risk” that public confidence in the administration of justice is – or might be – undermined depends very much on the court’s objective assessment of the relevant facts of the case itself (see also Shadrake 1 at [52]-[53], quoted above at [27]). It is of course true that some legal test is required. Our simple point, however, is that, in this instance at least, seeking to elaborate upon a legal test whose efficacy is to be demonstrated more in its application rather than its theoretical elaboration is, with respect, perhaps an approach that should be avoided. The fact of the matter is that the great strength of the “real risk” test lies, inter alia, in its practical robustness. Put simply, we are of the view that the “real risk” test is adequate in and of itself and, hence, does not require further elaboration. What is clear is that the “real risk” test will not be satisfied in a situation where the risk of undermining public confidence in the administration of justice is remote or fanciful. And, as explained below (at [39]), where there is, at the other end of the legal spectrum, a situation that would have satisfied the more stringent “clear and present danger” test, that particular situation would clearly fall within the purview of the less stringent “real risk” test. However, there will be many situations that lie in between and, as already emphasised, much will depend on the particular facts and context of the case in question.
[30] We would add that, even if the Judge had indeed held the view that “a small likelihood” of risk constituted a real risk (a view which we disagree with for the reason stated in [28] above), such a “small likelihood” of risk would merely be a “technical” contempt that may not even attract any sanction at all (see Lord Reid’s view in Times Newspapers at 299, and also Shadrake 1 at [77]). Such a situation would be a highly borderline or marginal case of contempt and, viewed from the perspective of substance and even practical principle, might not even merit the initiation of contempt proceedings by the Respondent in the first place. We would think that, where such a situation arises, the Respondent would consider it appropriate to initiate contempt proceedings only if other very exceptional circumstances exist. That having been said, we do not think – as already explained above – that an elaboration of the “real risk” test is helpful to the courts which must decide on the facts before them. The basic question is a simple one: is there, having regard to the facts as well as surrounding context, a “real risk” that public confidence in the administration of justice is - or would be - undermined as a result of the impugned statement?
[31] This brings us to the next issue as to who constitutes “the public” for the purposes of the law relating to scandalising contempt. However, before proceeding to consider this issue, we note that the Judge’s emphasis – in reliance on prior local cases – on the small physical size of Singapore itself is, at best, a neutral factor (see Shadrake 1 at [52], and reproduced above at [27]). With the advent of the technological age in general and the Internet in particular, the geographical size of a particular jurisdiction is no longer an important factor vis-à-vis the application of the “real risk” test, for the very simple reason that even in a geographically large jurisdiction, information can still be disseminated both quickly and widely. A further factor mentioned by the Judge (at [52] of Shadrake 1, and reproduced above at [27]) which, in our view, is also neutral at best, is his reference to “the fact that judges in Singapore are the sole arbiters of fact and law in cases coming before the courts”.
[32] Returning to the issue as to who constitutes “the public”, it will be recalled that the Judge observed (see Shadrake 1 at [52], and reproduced above at [27]) that “those who come into contact with the impugned publication may not always be average reasonable persons” [emphasis added]. He added that:
(See also the Hong Kong Court of First Instance decision of Secretary for Justice v Oriental Press Group Ltd [1998] 2 HKC 627 (“Oriental Press Group”) which the Judge cited and relied upon). If all the Judge is stating is that regard should always be had to the factual context of the case itself, this would be correct. However, there is, with respect, a certain degree of ambiguity in these observations inasmuch as they may suggest that the concept of the average reasonable person may, on occasion at least, be dispensed with. If this is what was meant, we must respectfully disagree. In our view, “the public” must, by definition, comprise the average reasonable person. It is true that different persons might respond differently to the same impugned statement. However, the court concerned must make an objective decision as to whether or not that particular statement would undermine public confidence in the administration of justice, as assessed by the effect of the impugned statement on the average reasonable person. As Lord Radcliffe observed, albeit in another context (that relating to the doctrine of frustration in the law of contract), in the House of Lords decision of Davis Contractors Ltd v Fareham Urban District Council [1956] AC 696 (“Davis Contractors”) (at 728):
[33] In this regard, we respectfully disagree with the view expressed in the joint judgment of Chan CJHC and Keith J in Oriental Press Group (at [55]) (cited in Shadrake 1 at [28]):
With respect, the proposition just set out is not, as explained above, consistent with what we understand to be the general position on this particular aspect of the law. Nor ought it to be, given the serious consequences to the persons found in contempt of court.
[34] However, it should be emphasised (especially in light of the observations by Lord Radcliffe in Davis Contractors (cited above at [32])) that the court proceeds on an objective inquiry and does not substitute its own subjective view as to who comprises “the public”, and, in so far as the observations by Lord Radcliffe just referred to suggest otherwise, they ought to be qualified accordingly. What is clear, in our view, is that the concept of “the public” cannot differ according to different factual matrices although these matrices are the relevant backdrop against which to ascertain whether or not public confidence in the administration of justice has been – or might be – undermined.
[35] Finally, we would caution that the illustrations referred to by the Judge, viz, rants made at a dinner party and a letter circulated by a disappointed litigant to holders of high public and legal office (see [54] of Shadrake 1, also reproduced above at [27]) cannot admit of only one correct answer in every case. Much would, in our view, depend on the precise facts and context in which the impugned statement is made. For instance, we note that the Judge was careful to qualify his view (at [54] of Shadrake 1) that there would be no contempt in the dinner party rant scenario if the rants “are shown to have been ignored” [emphasis added], because in such a case there would be no real risk of undermining public confidence in the administration of justice. The point we are making here is simply that having regard for the precise facts and context in which the impugned statement is made is crucial.
(2) A summary
[36] Let us summarise. Put simply, the “real risk” test is an adequate formulation in and of itself and requires no further theoretical elaboration. It is, at bottom, a test that means precisely what it says: is there a real risk that the impugned statement has undermined – or might undermine – public confidence in the administration of justice (here, in Singapore)? In applying this test, the court must avoid either extreme on the legal spectrum, viz, of either finding that contempt has been established where there is only a remote or fanciful possibility that public confidence in the administration of justice is (or might be) undermined or finding that contempt has been established only in the most serious situations (which is, as we shall see in the next section of this judgment, embodied within the “clear and present danger” test). In undertaking such an analysis, the court must not substitute its own subjective view for the view of the average reasonable person as it is clear that the inquiry must necessarily be an objective one. Much would depend, in the final analysis, on the precise facts and context in which the impugned statement is made.
[37] As already mentioned, extreme positions in relation to the “real risk” test must be assiduously avoided – whether in form and/or in substance. This brings us to one of the main arguments canvassed by counsel for the Appellant, Mr Ravi, in the present appeal which was, as we shall demonstrate in a moment, merely an attempt to introduce a wholly different (and stricter) test, viz, the “clear and present danger” test, under the guise of the “real risk” test. Let us turn now to consider this particular argument.
(3) The Appellant’s attempt to introduce the “clear and present danger” test under the guise of the “real risk” test
[38] As already alluded to in the preceding paragraph, Mr Ravi sought to go further by arguing that there would be a “real risk” of undermining public confidence in the administration of justice only if the statement concerned posed a clear and present danger that public confidence in the administration of justice would be undermined. A moment’s reflection would reveal that Mr Ravi was attempting to equate the concept of a “real risk” with the concept of a “clear and present danger” in so far as the potential undermining of public confidence in the administration of justice is concerned. It would take little further reflection to conclude that such an attempted “equation” is, with respect, deeply flawed. Both concepts clearly do not have the same meaning. Let us elaborate.
[39] Whilst it would follow that a statement which posed a clear and present danger would simultaneously pose a real risk, the converse does not follow. In other words, a statement which would pose a real risk would not necessarily pose a clear and present danger. The criterion centring on the concept of a “clear and present danger” is an extremely stringent standard compared to the concept of a “real risk” and, to that extent, the latter would encompass the former – but not vice versa. Mr Ravi was, in substance, attempting to introduce a completely different test (viz, the “clear and present danger” test) into Singapore law under the guise of the “real risk” test. As we have seen, both tests might overlap inasmuch as situations which fall within the scope of the “clear and present danger” test would also fall within the scope of the “real risk” test. However, as we have emphasised, the converse does not follow.
[40] It is important, in our view, to emphasise that if a particular statement poses a “real risk”, this would be sufficient to render that statement contemptuous. Indeed, the very concept of a “real risk” is a weighty one and it therefore comes as no surprise that the “real risk” test is the predominant (indeed, almost exclusive) test that applies throughout the Commonwealth (and see generally the case law cited below at [45]-[49]).
[41] The “clear and present danger” test applies, in the main, in the United States’ (“US”) context (see, eg, the US Supreme Court decision of Bridges v State of California (1941) 314 US 252 (“Bridges”)), where the concept of freedom of speech is inextricably linked to the unique culture as well as constitutional position (ie, the First Amendment) in the US (see Bridges and the Constitutional Court of South Africa’s decision of The State v Russell Mamabolo (2001) 3 SA 409 (CC) (“Mamabolo”) at [21]). With the exception of a seemingly solitary and divided Canadian decision which we will deal with in more detail below (at [43]-[44]), the “clear and present danger” test appears to apply in no other Commonwealth jurisdiction. To return to this particular test as set in its US context, it should, first, be noted that in the US there does not even appear to be a concept of scandalising contempt to begin with (see Bridges) – the US Supreme Court in Bridges considered criticism of the courts, no matter how unrestrained, made after a decision has been rendered, to be an exercise of the right of free discussion and free speech. As alluded to above, the concept of freedom of speech has – owing to the unique cultural as well as constitutional heritage of the US – been accorded a paramountcy in a manner quite different from other Commonwealth jurisdictions. This is not surprising when we consider the language of the First Amendment itself, as follows:
The US First Amendment is clearly quite different from the corresponding articles in the respective constitutions of Commonwealth jurisdictions (of which Art 14 of the Singapore Constitution is a representative illustration). This is not to state that freedom of speech is absent – or even lacking, for that matter – in Commonwealth countries. There is, instead, far more attention accorded to the issue of balance between the right to freedom of speech on the one hand and its abuse on the other (inter alia, by conduct amounting to contempt of court). One might add that the paramountcy accorded to the right to freedom of speech in the US is not, with respect, necessarily an approach that ought to be emulated as it could (as we have noted above at [17]) actually result in possible abuse and consequent negation of the right itself. This is no mere parochial rhetoric but is, rather, premised on logic and commonsense. Hence, it is no surprise, therefore, that jurisdictions across the Commonwealth (which are numerous as they are diverse and which, of course, include Singapore) adopt, instead, the approach from balance (see also above at [17]-[18]).
[42] Returning to the “clear and present danger” test, it was adopted in the US context with respect to sub judice contempt (see, eg, Bridges itself). On this related note, however, there is, in our view, no reason in principle why the “real risk” test (as opposed to the “clear and present danger” test) ought not to apply in situations of sub judice contempt as well. Indeed, in the English context, the “real risk” test first enunciated in Nash related in fact to a situation of sub judice contempt, the test there being whether there was a real risk that the impugned statement would prejudice a fair hearing (see also above at [26]).
[43] That having been said, there is one apparently solitary Commonwealth decision which furnishes some (but, as we shall see, by no means clear) support for the “clear and present danger” test in the context of scandalising contempt – the Ontario Court of Appeal decision of Kopyto. Kopyto was perceptively summarised and analysed by Mortimer VP in the Hong Kong Court of Appeal decision of Wong Yeung Ng v Secretary for Justice [1999] 2 HKC 24 (“Wong Yeung Ng (CA)”), which affirmed the decision of the Court of First Instance in Oriental Press Group, and whose decision the Appeal Committee of the Court of Final Appeal in Wong Yeung Ng (CFA) refused to give leave to appeal against. In Wong Yeung Ng (CA), Mortimer VP summarised the decision in Kopyto as follows (at 41-42):
[44] As Mortimer VP aptly pointed out in the passage quoted from his judgment in preceding paragraph, the decision in Kopyto in so far as the legal test for scandalising contempt is concerned was a split one. Brooke and Dubin JJA clearly opted for the “real risk” test, whereas Cory and Goodman JJA opted for the “clear and present danger” test. The remaining judge, Houlden JA, did not need to decide on the precise test to be adopted. Hence, Kopyto is by no means a strong Commonwealth decision in favour of the “clear and present danger” test. Indeed, to the best of our knowledge, there does not appear to be any decision subsequent to Kopyto which definitively favoured the approach of Cory and Goodman JJA.
[45] On the contrary, the vast majority of decisions in other jurisdictions point in the opposite direction (viz, in favour of the “real risk” test).
[46] For example, in Hong Kong, although the Court of Appeal in Wong Yeung Ng (CA) did not really give a definitive pronouncement as to which view in Kopyto ought to be preferred, the Court of Final Appeal in Wong Yeung Ng (CFA) did not grant leave to appeal to the appellant, one of the grounds in his application being that the Hong Kong courts should adopt the Kopyto “clear and present danger” test instead of the “real risk” test.
[47] And, in Mamabolo, the Constitutional Court of South Africa agreed with the minority’s view in Kopyto, expressing misgivings (at [35]) about the:
[48] We would add that the views of Brooke and Dubin JJA in Kopyto are more persuasive than those of Cory and Goodman JJA because they acknowledged the clear difference between the Canadian and the US position, citing at [72] the following observations of the New Zealand Court of Appeal in Radio Avon (at 234):
[49] In summary, the “clear and present danger” test does not represent the law in Singapore and Mr Ravi’s attempt to introduce it under the guise of the “real risk” test fails.
[50] Let us turn now to consider the other main test, viz, the “inherent tendency” test as well as its relationship to the “real risk” test.
The “inherent tendency” test
(1) The test
[51] The “inherent tendency” test was first articulated by the Singapore High Court in Attorney-General v Wain Barry J and Others [1991] 1 SLR(R) 85 (“Wain”) (the related judgment with regard to sentence is reported at Attorney-General v Wain Barry J [1991] 1 SLR(R) 108), where Sinnathuray J observed as follows (at [54]):
[52] With respect, there does not appear to be any clear authority for the “inherent tendency” test and, indeed, the learned judge in Wain did not cite any such authority. However, as the Judge correctly observed, this particular test was “subsequently referred to in decisions of the [Singapore] High Court and seems to have developed a life of its own” (see Shadrake 1 at [34]). The various Singapore High Court decisions which appeared to adopt the “inherent tendency” test include Attorney-General v Chee Soon Juan [2006] 2 SLR(R) 650 at [31]; Hertzberg at [25]–[34]; and Tan Liang Joo John at [12].
[53] We do note, however, that, in the Supreme Court of Queensland decision of Attorney-General for State of Queensland v Colin Lovitt QC [2003] QSC 279 (“Lovitt”), Chesterman J did observe (in a similar vein to Sinnathuray J in Wain (above at [52])), as follows (at [58]):
(2) The relationship between the “inherent tendency” test and the “real risk” test
[54] The test laid down in Wain (reproduced at [51] above) is somewhat ambiguous in so far as it does not clearly set out the precise relationship between “real risk” and “inherent tendency”.
[55] It may be possible, on one reading at least, to interpret Wain as conceptualising “inherent tendency” in contradistinction to “real risk”; indeed, this is what several courts have done in subsequent cases.
[56] However, a holistic reading of Wain suggests that the learned judge did not intend to divorce the test from its actual or potential impact on public confidence in the administration of justice. It is, in fact, axiomatic that the law in general and the law relating to scandalising contempt in particular do not – and cannot – operate in a hermetically sealed environment. That this is the case is clearly illustrated by the actual application of the law to the facts by the learned judge himself (and a similar interpretation could arguably be taken of Lovitt). This must surely be the case as, in our view, it would be contrary to both logic as well as commonsense for the “inherent tendency” test – or any test for that matter – to be stated only at a purely abstract or theoretical level. Indeed, even a theoretical formulation must have in view the vital sphere of application, having regard of course to the particular facts and context of the case in which that formulation is applied. In our view, in none of the decisions cited above (at [52]) is there any indication whatsoever that the court concerned had ignored the particular facts and context of the case at hand in arriving at its decision. Looked at in this light, the apparent distinction between the “inherent tendency” test on the one hand and the “real risk” test on the other is, in our view, a “legal red herring”.
[57] As stated earlier, however, we note that subsequent courts have appeared to define the “inherent tendency” test in contradistinction to the “real risk” test. Although those cases would have been decided the same way even if the “real risk” test was applied, this provides us reason enough to eschew the use of the term “inherent tendency”. As the Judge noted (at [50] of Shadrake 1), emphasising the test to be that of “real risk” would avoid controversy and misunderstanding by conveying precisely the legal test to layperson and lawyer alike. We therefore unequivocally state that the “real risk” test is the applicable test vis-à-vis liability for scandalising contempt in Singapore.
[58] We turn now to consider the concept of fair criticism. As we shall see, although it was apparently assumed in the court below that this concept is a defence in the law relating to contempt, it might well be the case that – whilst still relevant (and even vital to the ultimate outcome of the case concerned) – this particular concept is an integral part of the entire process of inquiry as to whether or not the impugned statement is in contempt of court in the first place. Put simply, the concept of fair criticism might go to the issue of liability rather than constituting a defence after a finding of liability for contempt has been made. Let us now turn to the concept itself.
The concept of fair criticism
Introduction – liability or defence?
[59] It had, apparently at least, been assumed in the court below that the concept of fair criticism is a defence to a charge of contempt of court (see Shadrake 1 at [70]–[76]). However, and as was suggested to counsel during oral submissions before this court, the precedents cited in support of this approach towards the concept of fair criticism are – at best – ambiguous. Put simply, it could be said that these precedents, whilst discussing the concept of fair criticism, nevertheless do not treat it as a separate defence as such but, rather, as an integral part of the process of analysis as to whether or not the impugned statement is in contempt in the first place. Hence, the question posed in the heading to this section of the judgment.
[60] To elaborate, none of the key decisions expressly refers to the concept of fair criticism in the context of a separate defence. For example, in Ambard (which is often cited as one of the seminal cases with regard to fair criticism), the key passage in Lord Atkin’s judgment (delivered on behalf of the Privy Council) is as follows (at 335):
[61] One would immediately note that there is no mention in the passage just quoted of fair criticism being a formal legal defence as such. Indeed, it appears that Lord Atkin was focusing on the concept of fair criticism in the context of liability. This approach also appeared to be adopted much earlier in the English decision of The Queen v Gray [1900] QB 36 (“Gray”), where Lord Russell CJ observed thus (at 40):
[62] More recently, in the English Court of Appeal decision of Regina v Commissioner of Police of the Metropolis, ex parte Blackburn (No 2) [1968] 2 QB 150 (“Blackburn”), Lord Denning MR observed (in a similar vein) thus (at 155):
[63] In the same decision, Salmon LJ also observed, as follows (at 155-156):
[64] Finally, Edmund Davies LJ observed (at 156; and, significantly, citing Gray (see above at [61])):
[65] In Radio Avon, the court observed that there was a “right” to criticise the work of the courts. The relevant extract reads as follows (at 230):
It will be noted that the court in Radio Avon cited the passage from Ambard in coming to its conclusion that it was in the interest of freedom of speech that criticism put forward fairly and honestly for a legitimate purpose should not be found to be in contempt. As we have earlier demonstrated, Ambard appears to focus on fair criticism in the context of liability, rather than as a formal defence.
[66] In the Australian High Court decision of Nationwide News Pty Ltd v Wills (1992) 177 CLR 1, Mason CJ observed thus (at 32–33):
[67] In the same decision, Brennan J stated as follows (at 38-39):
[68] It is, of course, possible to interpret the statements of principle set out in the preceding paragraphs of this part of the judgment as being consistent with the concept of fair criticism as a defence inasmuch as that particular interpretation would also result in the absence of contemptuous conduct as well. However, the nature, tenor and thrust of these statements of principle are, in our view, more consistent with the concept of fair criticism as going towards liability instead.
[69] There is, however, at least one decision that refers to the “defence” of fair criticism. In Gilbert Ahnee and others v Director of Public Prosecutions [1999] 2 AC 294 (“Ahnee”), the Privy Council noted (at 306):
However, as just mentioned, the prior decisions cited by the Privy Council in Ahnee do not constitute (at least unambiguously) authority for conceptualising fair criticism as a defence.
[70] We note that the major treatises on the law of contempt shed little light on the issue of whether fair criticism relates to liability or operates as a formal legal defence. For example, CJ Miller, Contempt of Court (Oxford University Press, 3rd Ed, 2000) at para 12.31 et seq purports to deal with “[p]ossible [d]efences” [emphasis added], noting that defences to scandalising contempt “are not questions which have been subjected to detailed analysis in an English court”, and that “any conclusions must necessarily be tentative”. The text proceeds to analyse case law on the “possible defences” of justification and fair comment, without drawing any firm conclusions as to whether or not these concepts constitute defences to scandalising contempt. Similarly, in David Eady & ATH Smith, Arlidge, Eady & Smith on Contempt (Sweet & Maxwell, 3rd Ed, 2005) at para 5-252 et seq, the learned authors note that the defences available to scandalising contempt are uncertain.
[71] The reports of law commissions in England and Australia are similarly inconclusive. The Australian Law Reform Commission in its report Contempt (Report No 35, Australian Government Publishing Service, Canberra, 1987) (“the ALRC Report”) cited the passage from Ambard for the proposition that there was a right to criticise the judiciary (see the ALRC Report at para 413). However, no elaboration on the characterisation of such a right to criticise was proffered. The ALRC Report went on to debate the existence of the defences of justification and fair comment, noting that there were conflicting case law authorities on the issue of the defences to contempt. The ALRC subsequently tentatively rejected both defences (see the ALRC Report at paras 415 and 416). Notably, there was never any mention of a defence of fair criticism as such (see the ALRC Report at para 437 et seq). In our view, the closest that the ALRC Report came to addressing the concept of fair criticism (as understood by the Judge) was in an ambiguous statement at para 438, which reads in relevant part, as follows:
The italicised portion suggests that there is some other concept which is akin to, but not the same as, a “formal defence of justification” [emphasis added]. It is possible that, having earlier cited the Ambard passage at para 413 of the ALRC Report, the ALRC was in fact alluding to some concept of fair criticism not amounting to a formal defence – albeit, and with respect somewhat confusingly, in the context of the concept of justification.
[72] In the report of the Law Commission of England and Wales entitled Offences Relating to Interference with the Course of Justice (Law Commission No 96, 7 November 1979) (“the EWLC Report”), it was recommended at para 3.70 that the offence of scandalising contempt should be replaced by a new offence which punishes the publication or distribution of false matter:
In other words, a person will not be convicted of contempt if either (a) the allegation made by him or her is true; or (b) the allegation is false, but the alleged contemnor honestly believed it to be true and was not reckless as to whether it was true or false. This approach bears some similarity to the role given to the concept of fair criticism by the Judge, and may be thought to conceive of fair criticism as a defence. However, there is no reference to the concept of fair criticism whatsoever in the EWLC Report, much less any analysis of fair criticism existing as a formal legal defence.
[73] In the 1974 report of the Committee on Contempt of Court chaired by Phillimore LJ (“the Phillimore Report”), which was referred to in the EWLC Report, the Phillimore Committee recommended that justification should be a defence to contempt if the impugned statements were demonstrated to be for the public benefit (see the Phillimore Report, at paras 165–167). We note that the Phillimore Report did not cite any case law for its proposed norms, and that, again, no reference was made to the concept of fair criticism as such.
[74] We turn now to examine s 5 of the Indian Contempt of Courts Act 1971 (“ICCA”), which appears to address the concept of fair criticism (and which, to the best of our knowledge, is probably the only attempt in the Commonwealth to deal legislatively with this concept). Section 5 reads as follows:
We also note that s 8 of the ICCA provides that:
[75] With respect, there is a degree of ambiguity in s 5 concerning whether fair criticism, under Indian law, relates to liability or operates as a formal legal defence. On one view, it may be thought that s 5 treats the concept of fair criticism as being synonymous with the defence of “fair comment” in defamation. This view may be derived from the fact that the savings provision in s 8 refers to “[o]ther defence[s]”, thus possibly implying that s 5 was intended to refer to a “defence”. It may also stem from the Report of the Sanyal Committee which, in recommending the codification of contempt laws (eventually resulting in the ICCA), discussed “fair criticism” under the chapter entitled “Defences”. However, it should be noted that no elaboration is provided in the ICCA as to how precisely the concept of “fair criticism” should be characterised, and there is no express reference to it being a “defence” as such. Furthermore, s 5 appears to be a codification of the position in Ambard and Gray (see the Report of the Sanyal Committee, Chapter IX section 3; and see Aiyer’s Law of Contempt of Court, Legislature & Public Servants (Law Book Company, 5th Ed, 1976) at 91), which decisions we have earlier noted to be ambiguous vis-à-vis the characterisation of the concept of “fair criticism”.
[76] The other possible interpretation of s 5 is that the ICCA merely treats fair criticism as being related to the issue of liability rather than constituting a defence. In this regard, we note that a 2006 legislative amendment expressly provided for a defence of justification in s 13 of the ICCA, as follows:
Looked at in this light, the only defence expressly referred to in the ICCA is one of justification. This suggests that the s 5 reference to “fair criticism” was not intended to establish a defence of “fair criticism” as such. Indeed, it may indicate that “fair criticism” in s 5 should be characterised in a manner akin to how we have characterised Ambard and Gray, viz, as being related to liability instead of being a formal legal defence (see [60]–[61] above).
[77] We are not well placed, and certainly do not propose, to definitively determine how the concept of “fair criticism” should be characterised under the ICCA. Our only observation is that there is potential ambiguity with regard to the precise role and operation of the concept of fair criticism in Commonwealth case law and (possibly) even legislation.
[78] However, whether one approach is adopted instead of the other is not merely a theoretical issue. If the concept of fair criticism relates to liability, then the evidential burden would be on the party relying on it. The legal burden, on the other hand, would be on the Respondent to prove beyond a reasonable doubt that the impugned statement does not constitute fair criticism, and that it presents a real risk of undermining public confidence in the administration of justice. If, however, the concept of fair criticism constitutes a defence, then the legal burden would shift to the alleged contemnor to show on the balance of probabilities that the impugned statement amounts to fair criticism.
[79] In Singapore, although the court’s power to punish for contempt of court is constitutionally and statutorily expressed (see above at [17]), there is no statutory provision (contra the ICCA) which provides guidance on whether fair criticism relates to liability or operates as a formal legal defence. In our view, given the ambiguity in case law and authorities from other Commonwealth jurisdictions, the provision of any defence to scandalising contempt – which is itself manifestly a question of policy – is more properly addressed by Parliament (as was done in India via the ICCA in general, and the 2006 legislative amendment of the ICCA in particular). However, until defences to contempt are rigorously considered and legislatively provided for, it falls to the courts to determine how the concept of fair criticism functions in practice under the common law.
[80] We prefer viewing the concept of fair criticism as going towards liability for contempt of court. Indeed, given that scandalising contempt is quasi-criminal in nature, this approach has the additional benefit of ensuring that the alleged contemnor is not disadvantaged vis-à-vis the implications with regard to the burden of proof (discussed above at [78]). At this juncture, we note that a caveat is necessary: the precise characterisation of the concept of fair criticism was not fully canvassed before us (see also above at [59]). Accordingly, our views on this point must necessarily be taken, to that extent, to be provisional in nature. We do, however, apply – where the concept of fair criticism has been raised – the approach we have stated at the outset of this paragraph to the impugned statements in this particular appeal.
The applicable principles
[81] Even if, as we have assumed, the concept of fair criticism goes towards liability for contempt of court, this by no means renders nugatory the various principles and factors that have already been laid down in previous decisions – notably, in Tan Liang Joo John, where Judith Prakash J cited (at [14]) the passage from Ambard (which is, in turn, cited above at [60]) in support of her proposition that “[f]air criticism does not amount to contempt of court”. The learned judge then proceeded to pertinently observe as follows (at [15]-[23]):
[82] The above extract was also cited in part by the Judge in Shadrake 1 (at [70]). In our view, the factors enunciated in the preceding paragraph (albeit by no means exhaustive) are very helpful in assisting the court concerned to arrive at a conclusion as to whether or not the impugned material is in contempt of court. Indeed, it is at least arguable that Prakash J appeared, in Tan Liang Joo John, to treat the concept of fair criticism as going towards liability rather than as an independent defence.
[83] We pause here to address two arguments raised by the Respondent. The Respondent argued, first, that statements calling into question the independence, impartiality and integrity of courts can never amount to fair criticism. Second, the Respondent contended that allegations made outside the formal avenues (eg, the relevant court proceedings or the removal mechanism provided for in Art 98(3) of the Constitution) can never amount to fair criticism.
[84] We do not agree with these arguments. With regard to the first argument (viz, limiting the scope of fair criticism to criticism that does not call into question the independence, impartiality and integrity of courts), we are of the view that this overly limits the ambit of fair criticism. This view was also taken by the Judge (see Shadrake 1 at [76]) and in Tan Liang Joo John (at [21]–[23]). Indeed, the Respondent’s argument appears to render the concept of fair criticism nugatory, since most allegedly contemptuous statements, by their very nature, call into question the independence, impartiality and integrity of courts. We are fortified in taking this view by the wealth of Commonwealth authority holding that an imputation of judicial impartiality or impropriety is not ipso facto scandalising contempt (see the High Court of Australia decision of R v Nicholls (1911) 12 CLR 280 at 286; the New South Wales Supreme Court decision of Attorney-General for New South Wales v Mundey [1972] 2 NSWLR 887 at 910–911; the New Zealand Court of Appeal decision of Radio Avon at 230–231; the English Court of Appeal decision of Blackburn at 155–156; and the Privy Council decision of Ahnee at 1314).
[85] With regard to the second argument (viz, limiting the scope of fair criticism to criticism raised through the formal avenues provided for in the law), we are of the view that this is too onerous a limitation on the right to free speech. An alleged contemnor should not be precluded from proffering fair criticism merely because he or she did not have the means, or did not choose, to air his or her rationally supported criticisms via any of the formal legal avenues. Indeed, we agree with the Judge’s view that resort to the Constitutional inquiry process cannot be the only way in which courts and judges can be criticised for wrongdoing. As the Judge noted in Shadrake 1 at [68]:
A summary
[86] Whilst the concept of fair criticism could either go towards liability for contempt or constitute a defence to what would otherwise be considered contemptuous conduct, we prefer to utilise the concept in the former sense. In approaching this concept, the court should bear in mind the various factors set out (albeit non-exhaustively) in Tan Liang Joo John (above at [81]). The court ought always to apply this concept not only in relation to the precise facts and context but also bearing in mind the following key question throughout: does the impugned statement constitute fair criticism, or does it go on to cross the legal line by posing a real risk of undermining public confidence in the administration of justice – in which case it would constitute contempt instead?
Our decision
The issue of liability
Introduction and context
[87] Having stated the applicable principles of law, we turn now to apply them to each of the eleven statements which the Judge found to be in contempt of court. We also note that the Respondent has not filed a cross-appeal against the decision of the Judge in the court below against either liability or sentence.
[88] Before considering each of the statements, the relevant background and context should also be noted.
[89] First, in so far as the scope of distribution/publication of the book is concerned, counsel for the Appellant, Mr Ravi, had informed the Judge that almost 6,000 copies of the book had been sold internationally, especially at airports, and that some copies were being circulated in Singapore (see Shadrake 2 at [27]). Mr Ravi further stated that (ibid):
[90] Ms Kumarassamy Gunavathy, a senior analyst with the Media Authority of Singapore, also deposed by affidavit that she bought the book from Kinokuniya Book Store of Singapore Pte Ltd in Ngee Ann City. She stated that the book had been sold in Singapore by several other bookstores, including Select Books Pte Ltd and Mary Martin Book Sellers Pte Ltd. The Appellant chose not to cross-examine Ms Gunavathy on her affidavit, and therefore the contents of her affidavit are taken to be accepted by the parties (see Shadrake 1 at [84]).
[91] Second, the Judge noted that the Appellant held himself out as an investigative journalist (see Shadrake 1 at [85]–[86]). A perusal of the following parts of the book illustrates this point:
(a) The announcement on the back cover (“investigative journalism has come to mean the kind of brave reporting that exposes injustice … Alan Shadrake’s hard-hitting new book cuts through the façade of official silence to reveal disturbing truths about Singapore’s use of the death penalty”);
(b) The bibliographical note (“Alan Shadrake is a renowned veteran investigative journalist and author whose 50-year career has taken him around the world”);
(c) The preface (“meticulous search of legal files and archived cases going back to 1963”, references to unnamed lawyers, Central Narcotics Bureau officers and an anonymous librarian at the National Library);
(d) Extracts at page 21 (“yet another major scoop in my long career as an investigative journalist”); and
(e) A full page photograph of the Appellant posing outside the Supreme Court of Singapore, captioned: “Author Alan Shadrake spent months scrutinising court and other records all over Singapore to unearth judicial scandals”.
[92] Accordingly, the statements should be read as being statements emanating from a person who has held himself out as an investigative journalist and their actual or potential effect on the public should be assessed accordingly. It should be noted that the Appellant does not disagree with the Judge’s holding on this particular point.
[93] We turn now to consider the eleven statements seriatim in light of the legal principles set out above.
The first statement
[94] The first statement is found at page viii of the book, and reads as follows:
[95] The Judge was of the view that “legal system” in this statement could not be read to include everything but the courts; indeed, the Judge considered that the courts are the first thing that comes to mind when a reference is made to a “legal system” (see Shadrake 1 at [88]). Accordingly, he held that the first statement constituted a clear accusation that the courts (and other parts of the legal system) took politics, international trade and business into account in determining who lives or dies on the gallows (ibid).
[96] On the other hand, Mr Ravi argues, on behalf of the Appellant, that the Judge misinterpreted the first statement. He points out that nowhere in the context of this passage is there any reference to improprieties on the part of a judge or a court. He also points out that in the first substantive chapter of the book, ten pages down from the first statement, the target of the Appellant’s criticism is the executive (narcotics police, CNB officer, and prosecution). Accordingly, he contends that adopting a contextual interpretation, the Appellant’s key argument in the first statement is that prosecutorial decisions weigh heavily in the outcome of certain death penalty cases. The Appellant further argues that the Judge should have taken the same approach to “legal system” as that in the third statement (which was found not to be contemptuous). The third statement, found at page 27 of the book, reads:
[97] Although Mr Ravi purported to rely on a contextual approach, he has misapplied it. The first statement, whilst occurring at page viii of the book, is actually located within the second page of the entire book itself. It is also part of the Preface which attempts to give an overview of the book. More importantly, the context of the first statement, viz the sentences which immediately precede it on the same page, is as follows:
[98] There are, in fact, two main themes within the very title of the book itself, viz, “Once a Jolly Hangman – Singapore Justice in the Dock” [emphasis added]. Both themes are reflected in the passage from the Preface quoted in the preceding paragraph. The first theme is the examination of the particular career of the “most prolific executioner in the world”. This first theme is the leitmotif to the second broader – and main – theme of the book (disingenuously slipped in, as it were, in the form of the subtitle “Singapore Justice in the Dock”), the distinction between the two themes being emphasised by the short – but pithy and telling – sentence just before the first statement itself, viz, “[a]nd something else” (quoted in bold italics above). And that “something else”, as the first statement puts it, concerns “[s]omething sinister” [emphasis added]. Further, part of that machinery which allegedly perpetrates such undesirable results is “the Singapore legal system” [emphasis added]. Even when read literally (as opposed to contextually), the meaning is clear. Whilst Mr Ravi refers to executive action as well as prosecutorial decisions, it is axiomatic (especially to the average reasonable person who is a member of the public and, we might add, a fortiori, to persons who are legally trained) that a key institution in the administration of justice is the courts (which, inter alia, actually decide cases by applying the relevant law to the evidence before them). And, as already alluded to earlier in this paragraph, this meaning is buttressed by the subtitle of the book itself which encapsulates this theme even more pithily and succinctly, viz, “Singapore Justice in the Dock”. Indeed, the Appellant has placed the courts in his own “dock” and has found the result, amongst other things, “sinister”. If further support is required for the interpretation just adopted, one need refer no further than one page back, viz, to the very first page of the Preface – and, indeed, to the very first sentence thereof, which reads as follows: “It was never my intention to go to Singapore and write a book about its revered but much feared chief executioneror its justice system” [emphasis added in italics and bold italics]. One might add that – whether by design or inadvertence – the Appellant was clearly utilising the word “or” in the sentence just quoted in a conjunctive sense, having regard to the sense as well as context concerned. Finally, the back cover of the book (which also attempts to encapsulate the themes of the book) drives home the interpretation we have discerned in no uncertain terms – inter alia, in the sentence which states that the book “reveals thecruelty and imprudenceof anentire judicialsystem” [emphasis added in italics and bold italics].
[99] We would add that the third statement (reproduced above at [96]) – albeit referring to the term “legal system” – was held not to be contemptuous in the court below. We agree with the Judge. Indeed, there is a fundamental difference between the third statement and the first statement. In the latter, there is a clear reference to the judicial function of determining liability and imposition of the death sentence (“determine who lives and who dies on the gallows”), whereas there is no such unambiguous reference in the former. Giving the Appellant the benefit of the doubt, as the Judge rightly did, there is reasonable doubt that the third statement actually referred to the courts. For the reasons given above, however, the same cannot be said of the first statement.
[100] Put simply, the first statement, whilst short in length is long on meaning – a meaning that attributes by way of summary and theme to, inter alia, the judiciary, the role of achieving sinister ends and by “[working] in secret” at that. The first statement further suggests that the judiciary is influenced by considerations relating to “politics, international trade and business” when determining liability for capital offences. This is the very antithesis of the impartial administration of justice which the judiciary is pre-eminently responsible for and which is expected as a given by the public itself. In our view, it is clear, beyond a reasonable doubt, that there is a real risk that such a statement with such a meaning would undermine public confidence in the administration of justice in Singapore. We further note that fair criticism was not raised by the Appellant with respect to the first statement.
The second statement
[101] The second statement is found at page 3 of the book. It refers to the case of Julia Bohl, a German national who was arrested and initially charged with the capital offence of trafficking more than 500 grams of cannabis, which was subsequently reduced to a non-capital offence of trafficking in a smaller amount:
[102] The Judge noted that the reference to a sentence of “only five years” and that it was a “slap on the wrist arranged by the Singapore government” constituted a clear reference to the courts, since sentencing is generally (and was indeed in that case) the province of the courts (see Shadrake 1 at [89]). Further, the Judge rejected the Appellant’s explanation of the second statement in which the latter purported to rely on page 142 of Mr Subhas Anandan’s (“Anandan”) autobiography (Subhas Anandan, The Best I Could (Marshall Cavendish Editions, 2009)). In the Judge’s view, Anandan had merely stated that the prosecution could not have preferred a capital charge because the amount of cannabis seized was below the threshold of 500 grams (see Shadrake 1 at [90]). Accordingly the Judge found that the second statement referred to the courts and was therefore contemptuous.
[103] The Appellant contends that a fuller reading of Anandan’s book belies the interpretation arrived at by the Judge, as follows:
(a) First, the Appellant argues that Anandan did in fact convey the message that the diplomatic negotiations played a part in the change of charge. This may be gleaned from two key passages at pages 142 and 143 of Anandan’s book. To an ordinary reader, the suggestion from Anandan’s book is that the Australian authorities acted too slowly in lobbying on behalf of Nguyen, and therefore stood no chance of saving his life, whereas the German embassy acted early and effectively in negotiating a reduction in the charges put to Bohl. Accordingly, the Judge’s criticism of the Appellant’s “attitude towards his sources” is unwarranted, since Anandan’s book, which includes a first-hand account of the Bohl case, highlights the links between diplomatic negotiations and prosecutorial decision-making.
(b) Second, the word “court” appears nowhere in the impugned passage. By contrast, the “Singapore government” is clearly referenced in the passage.
(c) Third, while it is the case that sentencing is generally “the province of the courts”, courts can only sentence on charges which are brought and proven by the prosecution. The sting in the Appellant’s attack as borne out by Anandan’s first-hand testimony is that there was diplomatic influence in the bringing of charges against Bohl, not in the adjudication of her case. A contextual view supports this – the Appellant refers at page 8 to the Bohl case as one in which “[t]he charges against her were suddenly – and ‘miraculously’ – modified”; again a reference to the charging decision, not the adjudication of the case.

The judgment continues past this page. Read the rest at eLitigation.

Shadrake Alan v Attorney-General

Laws this judgment relies on

Grounds of decision

Introduction

The factual background

The decision below

The decision on liability

The decision on sentence

The issues

The legal principles

Introduction

The test for liability for scandalising contempt

Introduction

The “real risk” test

(1) The test

(2) A summary

(3) The Appellant’s attempt to introduce the “clear and present danger” test under the guise of the “real risk” test

The “inherent tendency” test

(1) The test

(2) The relationship between the “inherent tendency” test and the “real risk” test

The concept of fair criticism

Introduction – liability or defence?

The applicable principles

A summary

Our decision

The issue of liability

Introduction and context

The first statement

The second statement