South Africa: Supreme Court of AppealYou are here: SAFLII >> Databases >> South Africa: Supreme Court of Appeal >> 2011 >>  ZASCA 196 | Noteup | LawCite
S v Bogaards (864/2010)  ZASCA 196;  1 All SA 376 (SCA) (21 November 2011)
Download original files
Links to summary
Bookmark/share this page
THE SUPREME COURT OF APPEAL OF SOUTH AFRICA
Case No: 864/2010
In the matter between:
JACOBUS BOGAARDS …..................................................................Appellant
THE STATE …...................................................................................Respondent
Neutral citation: Bogaards v The State (864/10)  ZASCA 196 (21 November 2011)
Coram: MTHIYANE, MAYA, MHLANTLA, LEACH, and SERITI JJA
Heard: 24 August 2011
Delivered: 21 November 2011
On appeal from: North Gauteng High Court, Pretoria (Murphy J and Dolamo AJ, sitting as court of appeal):
The appeal succeeds to the extent that the appellant’s convictions on the two main counts and the sentences are set aside.
The order of the high court is set aside and replaced with the following:
MAYA JA (MHLANTLA JA concurring):
 I have had the privilege of reading the judgments prepared by my colleagues Leach and Seriti JJA in this matter. Regrettably, whilst I do agree with some of the reasoning and findings they each advance, I am unable to agree with their respective conclusions. I share Seriti JA’s view that the appeal should ultimately fail, but on a different basis. For the reasons he gives, I agree that the prosecution of Mr Bogaards and his wife was lawful, that the trial was conducted fairly and that the evidence adduced by the State indeed established that Mr Bogaards harboured and concealed the two ‘Boeremag’ trialists, Messrs Gouws and Van Rooyen, after their escape from custody and further failed to report their presence on his farm to the police. But I respectfully part ways with his reasoning here and agree with Leach JA that the State nonetheless failed to prove that such conduct constituted the offences envisaged in ss 11 and 12 of the Protection of Constitutional Democracy Against Terrorist and Related Activities Act 33 of 2004 (the Terrorism Act). In my view, the State established Mr Bogaards’ guilt on the alternative charge of harbouring and concealing escaped prisoners under s 115(e) of the Correctional Services Act 111 of 1998 (the Correctional Services Act).
 The material background facts are set out fully in Seriti JA’s judgment and I need not repeat them. The main question for determination is whether Mr Bogaards may be convicted in the circumstances of this case – where a party is charged under the Terrorism Act for harbouring and concealing persons he knows or ought reasonably to have known or suspected to have committed offences in contravention of ss 11 and 12 of the Terrorism Act alternatively s 115(e) of the Correctional Services Act. Sections 11 and 12 provide:
‘11 Offences relating to harbouring or concealment of persons committing specified offences
Any person who harbours or conceals any person, whom he or she knows, or ought reasonably to have known or suspected, to be a person who has committed a specified offence, as referred to in paragraph (a) of the definition of ‘specified offence’, or who is likely to commit such an offence, is guilty of an offence.
Duty to report presence of person suspected of intending to commit or having committed an offence and failure to so report
Any person who –
has reason to suspect that any other person intends to commit or has committed an offence referred to in this Chapter; or
is aware of the presence at any place of any other person who is so suspected of intending to commit or having committed such an offence,
must report as soon as reasonably possible such suspicion or presence, as the case may be, or cause such suspicion or presence to be reported to any police official.
Any person who fails to comply with the provisions of subsection (1) (a) or (b), is guilty of an offence.’
In paragraph (a) of the definition of ‘specified offence’ the words are said to mean ‘the offence of terrorism referred to in section 2, an offence associated or connected with terrorist activities referred to in section 3, a Convention offence, or an offence referred to in section 13 or 14’. Section 2 of the Terrorism Act in turn renders any person who engages in a ‘terrorist activity’ guilty of the offence of terrorism. (The finely detailed and wide ranging definition of ‘terrorist activity’ comprising numerous actions that fall under the definition is set out in paragraphs  and  below.)
 The charge sheet did not specify the acts of terrorism or terrorist activities which the State alleged that the Bogaards knew or ought to have known or suspected to have been committed by the fugitives. Only during argument in the trial did the State case crystallise: that Mr Bogaards ought to have known that the fugitives faced terrorism charges under the Internal Security Act 74 of 1982 (which was repealed in its entirety by the Terrorism Act) in the ‘Boeremag’ case and, from that fact and their subsequent escape from custody, must at the very least have suspected that they were guilty of those offences.
 The contentions made on Mr Bogaards’ behalf before us were, inter alia, that the elements of the offences created by s 54(4)1 of the Internal Security Act were replaced by the practically identical provisions of ss 11 and 12 of the Terrorism Act and that the definition of ‘terrorist activities’ contained in s 1 of the latter Act includes and broadens that of ‘terrorism’ in the former Act. Bearing in mind the Legislature’s express intention in both Acts to criminalise deeds of terrorism and terrorist activities and the harbouring, concealing and assisting of such offenders, continued the argument, it would be absurd to interprete the Terrorism Act to exclude from prosecution offenders who contravened its ss 11 and 12 in respect of crimes committed under the Internal Security Act merely because of the absence of an ‘expressed transitional provision that regulates the position’ in the Terrorism Act. (Mr Bogaards’ counsel correctly conceded that ‘[n]otwithstanding the extensive transitional provisions [set out in s 27 of the Terrorism Act] it is clear that the legislature did not explicitly provide for a set of facts where, like in casu, the deed of terrorism or terrorist activity was committed under Act 74 of 1982, and the harbouring or concealment and/or failure to report took place after Act 33 of 2004 commenced’.) We were thus urged to categorise the fugitives’ offences committed under the Internal Security Act, before the promulgation of the Terrorism Act and for which they were on trial when they escaped, under the banner of ‘specified offences’ referred to in s 11 of the Terrorism Act as the court below seems to have done.
 Section 27 of the Terrorism Act makes provision, in subsec (1), for the repeal of the Internal Security Act and the following transitional arrangements:
‘(2) All criminal proceedings which immediately prior to the commencement of this Act were instituted in terms of the provisions of the Internal Security Act ... and ... have not been concluded before the commencement of this Act, shall be continued and concluded, in all respects as if this Act had not been passed.
An investigation, or prosecution or other legal proceedings, in respect of conduct which would have constituted an offence under the Internal Security Act ... and which occurred after the commencement of that Act, but before the commencement of this Act, may be conducted and continued as if this Act had not been passed.
Notwithstanding the repeal or amendment of any provision of any law by this Act, such provision shall, for the purpose of the disposal of any criminal proceedings, investigation, prosecution or legal proceedings contemplated in subsection (2) or (3), remain in force as if such provision had not been repealed or amended.’
 The sum of these provisions is that any pending criminal proceedings, investigation, prosecution or other legal proceedings instituted in terms of the Internal Security Act would continue to finality under its provisions, even after its repeal, as if the Terrorism Act had not been promulgated. Evidently, none of these transitional provisions cater for Mr Bogaards’ situation as the conduct for which he was charged and the subsequent criminal proceedings against him occurred after the commencement of the Terrorism Act. The latter Act makes no express provision for a situation such as the present; where the act of terrorism or terrorist activity was committed during the operation of the Internal Security Act and the harbouring and concealment of the offenders occurred after its commencement.
 As Leach JA correctly points out, there is a well established rule of construction that the operation of a statute is prospective, to apply only after its enactment, unless the legislature clearly expressed a contrary intention that the operation should be retrospective, to apply prior to its enactment (see, for example, Peterson v Cuthbert Co Ltd 1945 AD 420; R v Sillas 1959 (4) SA 305 (A); Bellairs v Hodnett & another 1978 (1) SA 1109 (A); National Director of Public Prosecutions v Carolus & others 2000 (1) SA 1127 (SCA); Veldman v Director of Public Prosecutions Witwatersrand Local Division 2007 (3) SA 210 (CC) paras 26 - 27). The presumption therefore is that none of the provisions of the Terrorism Act may be applied retrospectively unless such intention appears expressly or by necessary implication from its provisions.
 I am mindful that the Terrorism Act did not create new offences as such in ss 11 and 12 in view of the provisions of s 54(4) of the Internal Security Act (quoted above) which also criminalised harbouring and concealing a person known or suspected to have committed acts of terrorism and failing to report such person’s presence – the basis of the finding made by the court below (and relied upon by Mr Bogaards) that ‘Van Rooyen and Gouws were charged with ... offences under the Internal Security Act, which are in terms of that Act defined as terrorism. It is therefore fair to say that when [they] escaped from custody they were then charged with conduct equivalent to that contemplated in the definition of terrorism or terrorist activities under the [Terrorism Act]’. However, the provisions of s 27 of the Terrorism Act present an insurmountable difficulty for this view. This is so because they unequivocally provide for offences committed whilst the Internal Security Act was in force to be dealt with after its repeal as if the Terrorism Act had not been passed, thus expressly excluding such offences from the umbrella of ‘specified offences’ as defined under the Terrorism Act. There is simply no room for the modification of the Terrorism Act to ‘fill the gap’ allegedly left by the Legislature in this statute that we were urged by State counsel to execute, by ruling that ‘acts of terrorism as referred to in section 54 of Act 74 of 1982 is conduct equivalent to conduct contemplated in the definition of terrorist activities under present legislation’.
 There is no ambiguity in the wording of ss 11 and 12 of the Terrorism Act. The sections are concerned only with conduct constituting ‘a specified offence’, according to its definition, committed or likely to be committed after the commencement of that Act on 20 May 2005. Therefore, a contravention of these sections arises only upon the commission of an offence which fits such definition (of a ‘specified offence). It was not in dispute that Mr Bogaards’ prosecution targeted the offences committed by the fugitives before their arrest towards the end of 2002. By no stretch of the imagination can such offences be said to constitute specified offences under the Terrorism Act. The presumption against retrospectivity was not rebutted and it was not proved that Mr Bogaards harboured or concealed a person he knew or ought reasonably to have known or suspected to have committed a ‘specified offence’ and thus contravened the provisions of s 11. The same reasoning necessarily applies equally to the charge under s 12 of the Terrorism Act which was based on Mr Bogaards’ failure to report the presence on his premises of persons suspected of intending to commit or having committed an offence of terrorism or offence associated or connected with terrorist activities under the Terrorism Act. He should therefore not have been convicted for a contravention of these provisions.
 But this is not the end of the matter. Mr Bogaards was further charged, in the alternative to contravening the provisions of s 11 of the Terrorism Act, with contravening s 115(e) read with s 1 of the Correctional Services Act. Section 115(e) read (this Act has since been amended as indicated in paragraph ):
‘Any person who ... harbours or conceals or assists in harbouring or concealing an escaped prisoner, is guilty of an offence and liable on conviction to a fine or to imprisonment for a period not exceeding ten years or to such imprisonment without the option of a fine or both.’
 Section 1 defined ‘prisoner’ to mean ‘any person, whether convicted or not, who is detained in custody in any prison or who is being transferred in custody or is en route from one prison to another’. ‘Prison’ means ‘any place established under this Act as a place for the reception, detention, confinement, training or treatment of persons liable to detention in custody or detention in placement under protective custody, and all land, outbuildings and premises adjacent to any such place and used in connection therewith and all land, branches, outstations, camps, buildings, premises or places to which any such persons have been sent for the purpose of imprisonment, detention, protection, labour, treatment or otherwise, and all quarters of correctional officials used in connection with any such prison, and for the purposes of section 115 and 117 of this Act includes every place used as a police cell or lock-up.’ (In terms of s 77 of the amending Act, for the term ‘prisoner’ has been substituted ‘inmate’ and for ‘prison’, ‘correctional centre’.)
 As indicated in the judgment of Seriti JA, the fugitives, who were held in custody at the Central Prison, Pretoria under an order of court which refused their application for bail in previous proceedings, disappeared during the court’s lunch adjournment taken between 12h45 and 14h00 during their criminal trial. It was discovered only at resumption of the proceedings after the break that they had stolen away. All efforts to find them were fruitless until their capture at a Pretoria residential complex several months later. Precisely when and from which point of the court building they launched their escape during the fateful lunch break is not clear from the record. The only evidence led by the State from its witnesses was merely that the fugitives escaped during the lunch adjournment and did not take their lunch which was served in the detention cells.
 Mr Bogaards challenged the validity of the fugitives’ detention warrant required by s 6(1) of the Correctional Services Act for their committal to a correctional centre on the basis that it was signed by a court orderly and that for that reason they were not ‘prisoners’ as envisaged in the Correctional Services Act when they escaped. Reliance for this contention was placed on the judgment in S v Motsasi 1998 (2) SACR 35 (W). There, the court held that to be valid, a detention warrant should be issued by a properly authorised official in the form of the presiding judge or the court registrar or a senior official from the registrar’s office appointed in terms of s 34(1) of the Supreme Court Act 59 of 1959.
 Mr Bogaards’ contention is premised on the notion that the fugitives were being held under a warrant of detention during the court session and I will assume without deciding that this is correct. The basis for the court’s view in Motsasi is not entirely clear. Whilst admittedly sound and standard practice now, there is no such statutory requirement as the court itself acknowledged and there does not appear to have been any policy making provision for such requirement when the case was decided. Mr Bogaards’ counsel threw no further light on this issue as he relied solely on the court’s reasoning. According to the undisputed evidence of the State witness, Inspector Carl Etsebet, the established procedure relating to the issue of a detention warrant at the material time allowed court orderlies (who, incidentally, are the very officials responsible for guarding prisoners in the court premises during court proceedings and manning the detention cells in which such prisoners are held in between the court sessions) to sign such documents until a departmental Circular dated 5 October 2006, after the escape, outlawed that practice. In that event, the relevant detention warrant would have been lawful and that puts paid to the challenge to the applicability of the provisions of s 115(e) of the Correctional Services Act on the basis of its defect.
 But, in my view, the result is the same even on the assumption that the detention warrant was defective and unlawful for the following reasons. First, the detention of a convicted or awaiting-trial person is lawful by virtue of a court order (see Isaacs v Minister van Wet en Orde 1996 (1) SACR 314 (A) at 323g-h; Minister of Justice and Constitutional Development & another v Zealand 2007 (2) SACR 401 (SCA) paras 17 and 18). A defect in a detention warrant, even one which renders it invalid, cannot supersede the authority of the relevant court order. Therefore, the fugitives in this matter were held in lawful custody not by virtue of a detention warrant (which in the present context amounts to no more than an administrative means of proving to the correctional services authorities that the person they are requested to receive is lawfully in custody and may therefore be detained in their facility) but by authority of the court order refusing them bail and remanding them in custody up to the time of their escape. It is primarily that court order authorising that they be kept in custody pending their criminal trial, and not the detention warrant issued under it for their committal in a prison or correctional facility for that purpose, that brought them within the definition of ‘prisoner’.
 They were remanded in custody in the detention cell during the lunch break and were under police control. It matters not whether that control momentarily lapsed, as seems to have happened, when they slipped away. In S v Mtwazi ; S v Ndlovu 1992 (1) SACR 367 (Tk) the Full Court reviewed a matter in which the individual accused were convicted of escaping from custody in breach of s 50(1)(a) of the Transkeian Prisons Act 6 of 1974 which bore definitions of the terms ‘prisoner’ and ‘prison’ that are similar to those contained in the Correctional Services Act. One accused absconded after being given permission by his police guard to go to a shop alone to buy tobacco. The other was removed from his cell to wash a police vehicle and escaped during the absence of the constable who guarded him. In confirming the convictions, the court made comments which I find apposite for present purposes. It said at 370h-371b:
‘The principle that a prisoner escapes from a state of confinement, irrespective of the de facto physical control which is being exercised over him at the time, finds further support in the following extract from the judgment ... in S v Ncube and another 1982 (4) SA 419 (ZS) at 412e, albeit that he was then dealing with an escape from custody prior to incarceration:
“Where every vestige of control is not voluntarily abandoned, and when it is, or ought to be, apparent to a person in lawful custody that a temporary relaxation of the extent of the control exercised over him is not intended by his custodians to be an abandonment of the lawful custody established over him, he is not, in my view, free to take advantage of the situation by decamping.”
A further reason for applying the abovementioned construction to the section is that to apply the “control criterion” would lead to absurd and farcical situations. If the latter criterion were applied, it would then mean that if a warder is incapacitated by natural causes whilst escorting a prisoner outside a “prison”, eg a heart attack, lightning, etc or by other causes over which he has no control, eg a motor vehicle collision, his ward, the prisoner, could then lawfully leave the scene for any place of his choice. Another absurd situation would result if a third person in such circumstances, without any prompting the prisoner, were to grab hold of the arms of the warder. The prisoner would once again not be under de facto control and would be able to leave without fear of contravening s 50(1)(a). These examples clearly reflect that the Legislature could not have intended “control” to be the deciding factor.’
(A full bench judgment of the same division in S v Daphe 1982 (4) SA 60 (Tk) which came to a different conclusion is, in my view, distinguishable since it was partly based on a finding that by leaving the police vehicle escorting him between different towns after it capsized and its driver and only official occupant lost consciousness, the escapee neither acted unlawfully in the circumstances of the case nor had the mens rea to escape as he had telephoned the police shortly thereafter and informed them of his whereabouts.) See also S v Anderson 1999 (1) SACR 153 (N), per Hurt and Nicholson JJ, at 161h-162f.
 In the absence of clear evidence regarding how the escape was executed, one is faced with two probable scenarios: the fugitives escaped either between the courtroom and the court detention cells or from the detention cells themselves. As I see it, the paucity of relevant evidence is not fatal to the State case as either scenario falls neatly within the ambit of s 115(e) having regard to the meaning which the Legislature ascribed to the terms ‘prison’ and ‘prisoner’. Given its general, ordinary meaning, the wording of the expansive definition of ‘prison’ leaves no doubt that a detention cell within a court building falls within its purview. Thus if the escape was launched from the detention cells, the fugitives would obviously be covered by the part of the definition of ‘prisoner’ that refers to ‘any person... who is detained in custody in any prison’. It further seems to me clear from the latter portion of the definition of ‘prisoner’ that a person in lawful custody, including an awaiting-trial prisoner, retains that status even when in transit between different locations (see S v South African Associated Newspapers Ltd v and another 1962 (3) SA 396 (T) at 398D; S v Sibiya 1987 (4) SA 180 (N) at 185H-I; S v Mtwazi supra at 369a-e). Of particular relevance for present purposes are the words ‘“prisoner” means any person ... who is being transferred in custody’. Some of the common definitions of the word ‘transfer’ are ‘to convey or take from one place, person, etc to another … to give or hand over from one to another’ (The Oxford English Dictionary 2ed vol XVIII). To my mind, this portion of the definition of ‘prisoner’ would necessarily apply if the fugitives escaped whilst en route from the courtroom to the detention cells to which they were dispatched until the court session resumed at 14h00. For these reasons, it is competent to convict Mr Bogaards for contravening the provisions of s 115(e) of the Correctional Service Act.
 Regarding the question of appropriate punishment, the provisions of s 115(e) of the Correctional Services Act contemplate a maximum sentence of 10 years imprisonment with or without a fine. Seriti JA’s judgment fully discusses Mr Bogaards’ personal circumstances (he was a 48 year-old farmer with a clean record and had a wife, his co-accused, and a son of an undisclosed age) and other factors relevant to the sentencing process, including the gravity of Mr Bogaards’ transgressions (the numerous charges against the fugitives he harboured and concealed include treason, sabotage, murder and involve allegations of the attempted assassination of former President Nelson Mandela, the manufacture and use of explosive devices, the bombing of a Soweto mosque and a Buddhist temple in Bronkhorspruit and many others), his defiance of police warnings not to provide the fugitives with shelter and the time, money and other valuable resources expended by the state on an extended, nation-wide search for the fugitives. Mr Bogaards’ lack of remorse must also be considered. He denied any wrongdoing until the bitter end. A grudging, limited concession that the fugitives had been on his farm after their escape from prison, which had to be made anyway in view of the overwhelming evidence, came from his counsel only at appeal stage. But he still refused to fully take the court into his confidence and admit liability for his actions. The horrendous consequences for humanity as a whole from acts of terror need not be spelt out. Those who aid and abet such conduct are as dangerous to mankind as the actual perpetrators of such acts. Clearly, a substantial custodial sentence is the only appropriate punishment in the circumstances of this case. In my view, a prison term of five years imprisonment would adequately serve the interests of justice in all the circumstances.
 I would accordingly make the following order:
1 The appeal succeeds to the extent that the appellant’s convictions on the two main counts and the sentences are set aside.
2 The order of the high court is set aside and replaced with the following:
Judge of Appeal
 I have had the benefit of reading the judgments of my colleagues, Seriti JA, Leach JA and Maya JA. Regrettably I find myself unable to agree with the reasoning and conclusion adopted by Seriti and Leach JJA. I concur in the conclusion and order proposed by Maya JA. I agree with her conclusion that the appellant should have been convicted of the alternative count of contravening s 115 (e) of the Correctional Services Act 111 of 1998, in that he harboured or concealed the escaped prisoners, Gouws and Van Rooyen.
 Unlike my colleagues, Leach and Maya JJA, I do not however think it is necessary to go into the question of whether the two escapees were prisoners at the time of their escape. The sole issue for determination on appeal (as identified in the heads and in argument) was whether ‘the warrants [under which Gouws and Van Rooyen were held at the time of their escape] were invalid’. Both Maya and Leach JJA are agreed that the appellant must fail on this issue and that this question must be answered in favour of the State. Regrettably my colleagues find themselves at loggerheads on the question of whether Gouws and Van Rooyen were ‘prisoners’ or whether they were in ‘prison’ at the time of their escape, and invariably come to different conclusions on that issue which as I have said, was never raised by the appellant or the respondent. That was not the basis on which the case was conducted at the trial. The appellant pleaded not guilty and did not tender a plea explanation indicating the basis of his defence. His defence on the alternative charge of escaping only emerged during the cross-examination of the state witness, Inspector Carlo Etsebet, who was in charge of the two escapees in the High Court, Pretoria from where Gouws and Van Rooyen escaped. The only defence disclosed by Advocate Muller SC, for the appellant, during cross-examination was that the detention warrants under which Gouws and Van Rooyen were held were invalid. Not a word was uttered about whether the escapees were ‘prisoners’ at the time of their escape or that they escaped from a ‘prison’. It bears mention that this is not just a legal issue but is also a question of fact. Had the State been alerted to the fact that this was raised as an issue the State might well have conducted its case differently.
 To now hold at his late stage, as Leach JA seeks to do, that the appeal should succeed on this charge on the basis that the Boeremag trialists, Gouws and Van Rooyen were not ‘prisoners’, is with respect a course which is not open to an appellate court and might well result in unfairness to the one side.
 Because the issue was not raised the State probably approached the matter on the basis that the appellant was not disputing that the two escapees were prisoners and all that it had to deal with was the validity or otherwise of the detention warrants. The appellant was represented on appeal by Muller SC, a silk of many years standing. There is no reason to think that he would not have applied his mind to the question whether Gouws and Van Rooyen were prisoners or not at the time of their escape, if he considered that there was the course the appellant wished to pursue. I do not think that it is open to this court to credit the appellant with a defence he did not raise, more so if this would result in prejudice and unfairness to the other party, in this case the State.
 I say all of the above mindful of the fact that the onus of proof in a criminal case is on the State. But as we know in criminal cases depending on how the case is run, the State is not always called upon to prove every element of the offence (which in my view is what happened in the present matter).
 Even if we were called upon on appeal (which I do not believe to be the case) to consider whether Gouws and Van Rooyen were ‘prisoners’ or in ‘prison’ at the time of their escape I would still favour an interpretation (of these words) that accords with logic and common sense and avoid a strictly textual and legalistic approach to the meaning of the words. (See African Christian Democratic Party v Electoral Commission  ZACC 1; 2006 (3) SA 305 (CC) at para 25; Weenen Transitional Local Council v Van Dyk 2002 (4) SA 653 (SCA) para 13; Chauke v Santam Ltd  ZASCA 120; 1997 (1) SA 178 (A) at 183B-C). Before being taken to court Gouws and Van Rooyen were held as ‘prisoners’ in prison. I do not believe that they lost that status when they stepped out of the prison doors en route to court. In my view, they maintained their status as ‘prisoners’ and when they escaped they did so as prisoners. That for me is what makes logical sense.
 For the above reasons I concur in the order proposed by Maya JA.
K K MTHIYANE
JUDGE OF APPEAL
 The appellant, Mr Jacobus Bogaards and his wife, Mrs Elizabeth Bogaards, were arrested on 13 January 2007 and appeared on 15 January in the Modimolle Magistrates’ Court. They again appeared on 22 January when bail was fixed, and the matter was postponed to 16 February where, a charge sheet was handed to them and the matter was post-poned to 12 July 2007 for trial in the Modimolle Regional Court. The trial proceeded and they were charged with contravening ss 11 and 12(1)(b) of the Protection of Constitutional Democracy Against Terrorist and Related Activities Act 33 of 2004 (the Terrorist Act), and one alternative charge of contravening s 115(e) of the Correctional Services Act 111 of 1998 (Correctional Services Act).
 The appellant and his wife, who were out on bail, were subsequently convicted on the two main counts. He was sentenced on count 1 to five years’ imprisonment, two years of which were suspended on certain conditions, and one year imprisonment on count 2, and it was ordered that the sentence on the latter count run concurrently with the sentence imposed on count 1. In the case of his wife the two counts were taken together for purposes of sentence and she was sentenced to two years’ imprisonment, wholly suspended on certain conditions.
 The appellant appealed against his convictions and sentences and his wife appealed against her convictions only. On 11 February 2010, the North Gauteng High Court (Murphy J and Dolamo AJ) confirmed his conviction and sentence and set aside the conviction of his wife. On the same date the high court granted him leave to appeal to this Court, and his bail was extended pending the finalisation of the appeal.
 During 2002 several people were arrested. Amongst these were Mr Herman Van Rooyen (Van Rooyen) and Mr Jan Rudolph Gouws (Gouws). Van Rooyen was arrested on 11 December 2002, in Pretoria. He was a farmer in the BelaBela area, a captain in the South African National Defence commando stationed at BelaBela, and a commander of the reaction force. Gouws was arrested on 10 December 2002, in Pretoria. He worked for Van Rooyen on the farm and he was also a member of the BelaBela commando having a rank of corporal. The appellant was also a member of the commando and he also attended meetings of the commando.
 A total of 22 people including Van Rooyen and Gouws were indicted and appeared before the North Gauteng High Court on several charges. These included murder, sabotage and four counts of terrorism, under s54(1) of the Internal Security Act 74 of 1982. The matter is referred to as S v Du Toit. The arrest and subsequent trial engendered widespread public interest. The trial came to be known as the ‘Boeremag trial’, and it is convenient to refer to it as such in this judgment. The trial commenced on 13 May 2003 and Van Rooyen appeared as accused number 18 and Gouws as accused number 22. The matter is still proceeding before the high court and Van Rooyen and Gouws were never granted or released on bail since their arrest and detention.
 Prior to the beginning of the ‘Boeremag trial’ the police identified the appellant as a possible state witness and on 12 February 2003 a statement was obtained from him. However he was never called as a state witness. In the said statement the appellant, amongst others, stated that ‘[e]k weet van Herman van Rooyen aangesien ons saam vergaderings by die DLU bygewoon het’.
 As stated earlier Van Rooyen and Gouws were kept in detention while the trial was continuing. They were kept at the Central Prison, Pretoria. On 3 May 2006 the trial continued until 12h45 when the court adjourned for lunch. All the accused were present when the court adjourned. At 14h00, when the trial resumed, it was discovered that Van Rooyen and Gouws were missing. Police officers searched the court building and organised road blocks but Van Rooyen and Gouws could not be found. When Van Rooyen and Gouws escaped they did not have any parcels or money on them, except the clothes they were wearing. Police established an operational room to co-ordinate the search of the two escapees. A country wide search was conducted, media statements were issued and photographs of the two escapees were released to the media. A helpline was also established and a contact number was also provided to the media. Interpol was also contacted for assistance. The bank accounts of the escapees were monitored and there were no movements on the said accounts.
 On 22 October 2006 police officers went to the appellant’s farm (the farm) to search for the escapees. At the house the police found the appellant, his wife and a certain Mr Willem Ratter. The house and the surrounding buildings were searched but the escapees could not be found. Superintendent Vreugdenburg, Chief Investigating Officer in the ‘Boeremag’ matter, was in charge of the operation. He informed the appellant that they were looking for Van Rooyen and Gouws and the appellant denied that the escapees were at his house. Police searched only the buildings on the farm and could not find the escapees. The farm was not searched. Superintendent Vreugdenburg further testified and said: ‘[j]ust before we left I told accused 1 that if we don’t find the two escapees on his farm on this day, I gave him a warning that if they are there and I did not find them, he must get rid of them, chase them away, he mustn’t allow them on his farm. He again denied any knowledge of them’.
 On 13 January 2007 police went to the farm and searched the buildings there. In one of the corrugated iron rooms, a motorbike was found. The motorbike was covered with a raincoat and a dirty piece of material. The motorbike had registration numbers and within a few minutes police officers established that the motorbike was registered in the name of Van Rooyen. Superintendent Vreugdenburg asked the appellant whose motorbike it was and the appellant informed him that he bought it about two months prior from Mr Gerald van Rooyen, the father of Van Rooyen, one of the escapees, and that the transaction was done by the exchange of cattle. The appellant also mentioned that the motorbike was brought to the farm by Mr Allen Van Rooyen, brother of the escapee. When asked about the keys of the motorbike the appellant initially said that they were in his son’s room, however he later stated that they were lost.
 Some of the police officers searched the farm itself. They found a tent hidden in the veld on the farm, about 500m away from the house. The tent was in a dry riverbed covered with a piece of cloth. A box, full of food, and army backpacks were found next to the tent. Inside the tent they found sleeping bags and mattresses. One of the bags was opened and inside they found shoes, pants and a shirt which they recognised as belonging to Gouws, one of the escapees. The pants were marked ‘RG’. A fire-arm and a helmet were found. Fingerprints of Van Rooyen were found on the helmet and fingerprints of Gouws were found on the bag of sugar found in the tent.
 On the same day, whilst walking in the vicinity where the tent was discovered, one of the police officers saw Van Rooyen about 10m from where the tent had been situated, prior to its removal. Van Rooyen was carrying an R1 assault rifle. Van Rooyen after a short discussion with the police officer disappeared into the river and left. The articles found at the tent were taken away for forensic tests and the appellant and his wife were arrested. There were indications that the tent was there for a long time.
 On 20 January 2007 the two fugitives, Van Rooyen and Gouws were arrested at the Villa Mignon Complex, Lyttleton, Centurion. They were found in two separate bedrooms in one of the units. In one of the bedrooms they found two R1 assault rifles, with a magazine which was filled with live rounds and one live round in the chamber of the rifle. Cell phones, GPS apparatus, certain documents and a few maps were also found. Keys to the motor bike, that had been found on the farm was also found at the place where Van Rooyen and Gouws were arrested.
 As mentioned earlier, the appellant and his wife were arrested on 13 January 2007. They appeared in the magistrates’ court on 15 January and their case was postponed to 22 January for bail application. On 22 January they again appeared in court. They were granted bail of R10 000 each and the case was postponed to 16 February for further investigation.
 On 8 February 2007, the Deputy Director of Public Prosecutions addressed a letter to the Limpopo Province Regional Court President. The letter reads as follows:
‘THE STATE versus BOGAARDS AND ANOTHER
TRIAL DATE – MODIMOLLE
It has been decided to prosecute Mr and Mrs Bogaards in the Regional Court on charges of contravening sections 11 and 12 of Act no 33 of 2004: Protection of Constitutional Democracy against Terrorism and Related Activities, as well as a charge of contravening section 115(e) of the Correctional Services Act, 111 of 1998.
The prosecution originates from their alleged harboring of and assistance to two accused in the Boeremag trial who escaped from the High Court, Pretoria in May 2006.
It has been decided to approach you with a request to make a regional court magistrate available to hear the matter during a special sitting of a regional court to be arranged at Modimolle…
It appears that a date that will suit the defence, the State and the logistic arrangements of the magistrates court will be 11 - 27 July 2007.’
 On 16 February 2007 the appellant and his wife appeared in court and their case was remanded to 11 July for trial at Modimolle Regional Court. A charge sheet was given to the defence counsel, and on request of the defence counsel, the court ordered the State to furnish the defence with further particulars to the charge sheet on or before 31 March 2007.
 On 19 June 2007 Ramaite SC, acting National Director of Public Prosecutions (NDPP) signed authorisation in terms of s 16(1) of the Terrorist Act, in terms of which he authorised the prosecution of the appellant and his wife on two charges of contravening ss11 and 12(1)(b) of the Terrorist Act.
 On 12 July 2007 the appellant and his wife appeared before the regional court. The authorisation mentioned above was handed to the court. The appellant and his wife who were represented by counsel, pleaded not guilty and the trial proceeded without the defence objecting to the charges put to them. After leading evidence, the State closed its case. The defence closed its case without leading any evidence.
 In his heads of argument, the defence counsel conceded that the evidence adduced by the State proved that the two escapees visited or stayed on the farm for a certain period. In this appeal the defence raised five grounds of appeal and I will deal with them hereunder.
Validity of the Prosecution
 Counsel for the appellant submitted that the prosecution is a nullity because the State instituted the prosecution without the written authority of the NDPP as required by section 16(1) of the Terrorist Act. He further submitted that the prosecution was, as a fact instituted on 16 February 2007, when the charge sheet was given to the appellant and he was informed that he would be prosecuted in the regional court on the said charges.
 Section 16(1) of the Terrorist Act reads as follows: ‘[n]o prosecution under Chapter 2 may be instituted without the written authority of the National Director’. Sections 11 and 12 of the Terrorist Act fall under Chapter 2 of the said Act. In order to determine whether the submission by counsel for the appellant is correct or not, we should determine when the prosecution was instituted in this matter.
 In Minister of Law and Order v Kader  ZASCA 111; 1991 (1) SA 41 (A) the court had to deal with s 64 of the Internal Security Act 74 of 1982 (the Internal Security Act). The said section reads as follows:‘…[n]o prosecution… shall be instituted without the written authority of the attorney-general’. At 51e-f Grosskopf JA said: ‘[w]hat is meant by the institution of a prosecution depends on the context in which the expression is used (cf R v Priest 1931 AD 492 and R v Friedman 1948 (2) SA 1034 (C)). The purpose of s 64 is to ensure that the decision to prosecute a person for a contravention of s 54 is a responsible one, taken by the person who, in terms of s 3 of the Criminal Procedure Act, has the authority to prosecute in the name of the Republic in criminal proceedings. This purpose cannot be achieved if the Attorney-General is required to arrive at a decision on incomplete or preliminary information’.
 In National Director of Public Prosecutions v Moodley 2009 (2) SA 588 (SCA) the court had to deal with the interpretation of s 2(4) of the Prevention of Organised Crime Act 121 of 1998. The section reads: ‘[a] person shall only be charged with committing an offence contemplated in subsection (1) if a prosecution is authorised in writing by the National Director’. The accused were charged with certain offences some of which are affected by s 2(4) quoted above. A draft charge sheet was given to the accused on 10 December 2003 and the written authorisation by the NDPP was dated 24 March 2004. The respondents contended that they were charged on counts which required the NDPP’s authorisation prior to the State obtaining the required authorisation. Scott JA said, para 12: ‘[i]n my view counsel for the appellant correctly submitted that once the prosecution is authorised in writing by the National Director there can be no reason, provided the accused has not pleaded, why the further prosecution of the accused on racketeering charges would not be lawful, even if the earlier proceedings were to be regarded as invalid for want of written authorisation’.
 In this matter, the NDPP signed the authorisation in terms of s 16(1) of the Terrorist Act on 19 June 2007. On the trial date, 12 July 2007 the written authorisation was handed to the court prior to the accused entering their plea. There were no objections from the defence.
 The meaning of the words ‘institution of prosecution’ must be interpreted in the context of the legislation in order to assign to it its proper meaning. As stated in Kader supra at 51e-d, what is meant by the words ‘institution of a prosecution,’ depends on the context in which the expression is used. See also Zuma v National Director of Public Prosecutions 2008 (1) SACR 298 (SCA) para 10 and, S v F 1965 (3) SA 757 (T).
 Counsel for the appellant sought to rely on R v Priest 1931 AD 492, R v Friedman 1948 (2) SA 1034 (C) and Kader supra. These decisions are of no assistance to the appellant. In R v Priest, the court gave meaning to the word ‘prosecution’ within the framework of the Cape Libel Act 46 of 1882. There was no attempt on the part of the court, to give meaning to the word ‘prosecution’ which would be of general application. At 495 Wessels ACJ said ‘…but we have to determine… whether the word was there used in its more general meaning of a commencement of criminal proceedings’. In R v Friedman the accused was charged with the contravention of s 17(3) of the Food, Drugs and Disinfectants Act 13 of 1929. The court had to determine when the prosecution was instituted and it held that the prosecution was instituted against the accused on the date summons was issued. The court, in order to come to that conclusion, was guided by the intention of the legislature and took into account the legal framework of the legislation in question. As stated earlier, in Kader the court said that what is meant by institution of a prosecution depends on the context in which the expression is used.
 In my view, in the context of the Terrorist Act, the words ‘no prosecution may be instituted’, mean that the certificate signed by the NDPP must be available prior to the accused pleading to the charges. The main purpose of the section is, amongst others, to ensure that a decision to prosecute a person on a charge of this nature is taken by the highest official after properly considering all the relevant facts and implications of such a prosecution. If the NDPP can be expected to make a decision to issue the certificate, prior to the proper investigation of the case, that might undermine the objective of the section.
 Preliminary arrangements like the letter addressed by the Deputy Director of Public Prosecutions to the Regional Court President, providing the accused with a charge sheet and further particulars thereto cannot be regarded as the institution of a prosecution. These merely constitute the preparations for the prosecution of the accused.
 My conclusion on this point is fortified by the fact that if the accused had objected before the plea, the State could have simply withdrawn the charges and immediately thereafter re-charged him for the same offences. This would have led to the appellant and his wife being prejudiced by the further delays in the prosecution of their trial which would not have served the interest of justice. I am satisfied that the provision of s 16(1) has been complied with. Such an interpretation satisfies the aims and objectives of the statute without straining the meaning of words employed by the legislature.
Interpretation of ss 11 and 12(1)(b) of the Terrorist Act
 Section 11 reads as follows:
‘Any person who harbours or conceals any person, whom he or she knows, or ought reasonably to have known or suspected, to be a person who has committed a specified offence, as referred to in paragraph (a) of the definition of “specified offence”, or who is likely to commit such an offence, is guilty of an offence.’
Section 12(1)(b) reads as follows:
‘Any person who…(b) is aware of the presence at any place of any other person who is suspected of intending to commit or having committed such an offence, must report as soon as reasonably possible such suspicion or presence as the case made be, or cause such suspicion or presence, to be reported to any police official.’
 The definition clause defines ‘specified offence’ as inter alia, the offence of terrorism referred to in s 2. Section 2 of the Terrorist Act provides that any person who engages in terrorist activity is guilty of the offence of terrorism.
Terrorist activity, with reference to amongst others s 2 mentioned above is defined as:
‘(a) any act committed in or outside the Republic, which –
(i) involves the systematic, repeated or arbitrary use of violence by any means or method…
(iii) endangers the life, or violates the physical integrity or physical freedom of, or causes serious bodily injury to or the death of, any person, or any number of persons;
(iv) causes serious risk to the health or safety of the public or any segment of the public;
(v) causes the destruction of or substantial damage to any property, natural resource, or the environmental or cultural heritage, whether public or private…
(vii) causes any major economic loss or extensive destabilisation of an economic system or substantial devastation of the national economy of a country; or
creates a serious public emergency situation or general
insurrection in the Republic…and
(b) which is intended, or by its nature and context, can reasonably be regarded as being intended, in whole or in part, directly or indirectly, to -…
(ii) intimidate, or to induce or cause feelings of insecurity within, the public, or a segment of the public, with regard to its security, including its economic security, or to induce, cause or spread feelings of terror, fear, or panic in a civilian population; or
(iii) unduly compel, intimidate, or force, coerce, induce or cause a person, a government, the general public, or a segment of a public, or a domestic or an international organisation or body or intergovernmental organisation or body, to do or to abstain or refrain from doing any act or to adopt or abandon a particular standpoint, or to act in accordance to certain principles…’.
‘Terrorist and related activities’ is defined by the definition clause as any act or activity related to or associated or connected with the commission of the offence of terrorism, or an offence associated or connected with a terrorist activity, or a Convention offence or an offence referred to in ss 11 and 14.
Section 1(7) of the definition clause reads as follows:
‘For the purpose of this Act a person ought reasonably to have known or suspected a fact if the conclusions that he or she ought to have reached, are those which would have been reached by a reasonably diligent and vigilant person having both-
(a) The general knowledge, skill, training and experience that may reasonably be expected of a person in his or her position; and
(b) The general knowledge, skill, training and experience that he or she in fact has.’
 As stated earlier, Van Rooyen and Gouws appeared as accused 18 and 22 in the ‘Boeremag trial’. They faced several counts including murder, sabotage and terrorism. As far as the terrorism counts are concerned they are alleged to have contravened s54(1)(i),(ii),(iii) and (iv) of Internal Security Act 74 of 1982. Section 54(1) reads as follows:
‘Any person who with intent to-
overthrow or endanger the State authority in the Republic;
achieve, bring about or promote any constitutional, political, industrial, social or economic aim or change in the Republic;
induce the Government of the Republic to do or to abstain from doing any act or to adopt or to abandon a particular standpoint; or
put in fear or demoralize the general public, a particular population or the inhabitants of a particular area, in the Republic, or to induce the said public or such population group or inhabitants to do or to abstain from doing any act,
in the Republic or elsewhere -
commits an act of violence or threatens or attempts to do so;
performs any act which is aimed at causing , bringing about, promoting or contributing towards such act or threat of violence, or attempts, consents or takes any steps to perform such act;
conspires with any other person to commit, bring about or perform any act or threat referred to in paragraph (i) or act referred to in paragraph (ii) or to aid in the commission, bringing about or performance thereof; or
incites, instigates, commands, aids, advises, encourages or procures any other person to commit, bring about or performs such act or threat,
shall be guilty of the offence of terrorism and liable on conviction to the penalties provided for by law for the offence of treason.’
 Section 11 prohibits a person from harbouring or concealing persons whom he or she knows, or suspects or ought reasonably to have known or suspected to be a person who has committed terrorist activities or who is likely to commit such activities. Section 12(1)(b) requires any person who is aware of the presence at any place of a person who is suspected of having committed terrorist activities or likely to commit terrorist activities to report the presence of such person to any police official. The effect of the presumption contained in s 1(7) is that if the State relies on the suspicion of the accused, it would be sufficient to show that a reasonable, diligent and vigilant person with the general knowledge, skill, training and experience of the accused would have so known or suspected.
 In Powell NO v Van Der Merwe NO 2005 (5) SA 62 (SCA), paras 36 and 37, Cameron JA said:
‘This Court has endorsed and adopted Lord Devlin’s formulation of the meaning of ‘suspicion’:
“Suspicion in its ordinary meaning is a state of conjecture or surmise where proof is lacking; ‘I suspect but cannot prove’. Suspicion arises at or near the starting point of an investigation of which the obtaining of prima facie proof is the end…”
To the passage already adopted I would add the sentence that immediately follows, since it has a bearing on the present:
“When such proof has been obtained, the police case is complete; it is ready for trial and passes on to its next stage”.
Ferreira and Swanepoel were not ready to charge Powell or Nell. Prima facie proof was as yet lacking. Lord Devlin went on to point out:
“Another distinction between reasonable suspicion and prima facie proof. Prima facie proof consists of admissible evidence... Suspicion can take into account matters that could not be put in evidence at all… Suspicion can take into account also matters which, although admissible, could not form part of prima facie case.”’
See also Duncan v Minister of Law and Order 1986 (2) SA 805 (A) at 819I, and Kader at 50H-I.
 As indicated above, it is undisputed that the escapees stayed on the farm during the period alleged by the State. Their tent was a mere 500m away from the farm house. Van Rooyen’s motor-vehicle was kept in a shed near the house. It is inconceivable that the appellant would have been unaware of the presence of Van Rooyen and Gouws on his farm. There was only one entrance to the farm, and the appellant would necessarily have known about their movements in and out of the farm. I find on this evidence that the appellant harboured Van Rooyen and Gouws.
 As pointed out earlier, Van Rooyen and Gouws were arrested during December 2002 and they were on trial, facing several counts of terrorism, sabotage and murder. Their arrest and trial received wide media coverage. The appellant, Van Rooyen and Gouws were known to each other and were all members of the local Commando. Van Rooyen’s brother and father were known to the appellant. Police officers took a statement from the appellant as a potential state witness in the ‘Boeremag trial’. After the escape of Van Rooyen and Gouws, police mounted a massive manhunt looking for them. Their escape received wide media coverage. On 20 October 2006 police officers went to the farm looking for Van Rooyen and Gouws. Police officers after failing to find Van Rooyen and Gouws on the farm, ordered the appellant not to allow them on his farm.
 From the facts mentioned in the preceding paragraph, it is clear to me that the appellant knew that Van Rooyen and Gouws were some of the accused in the ‘Boeremag trial’. With his general knowledge and skills, the appellant knew, or suspected or ought reasonably to have suspected that Van Rooyen and Gouws had committed terrorist activities or are likely to commit terrorist activities.
 Counsel for the appellant submitted that a conviction under s 11 is not competent unless the harboured persons have been charged with terrorism under the Terrorist Act. He further submitted that the court a quo erred in finding that a conviction under s 11 is competent despite Van Rooyen and Gouws not having been charged under the Terrorist Act.
 The definition of ‘terrorist activities’ as defined in s 1 of the Terrorist Act include the activities which are regarded as terrorist activities in terms of the Internal Security Act. Some of the activities which form the basis of the charges that Van Rooyen and Gouws are facing under the Internal Security Act are activities which are equivalent to activities contemplated in the definition of terrorist activities under the Terrorist Act. Van Rooyen and Gouws were charged with activities that could constitute terrorism under the Terrorist Act.
 Counsel for the State contended, correctly so in my view, that it is clearly the express intention of the legislature in both the Internal Security Act and the Terrorist Act to make provision for the prosecution of people committing deeds of terrorism or terrorist activities as well as people harbouring, concealing or assisting such persons. Under both Acts failure to report such persons is an offence.
 In Janse Van Rensburg v Minister of Defence 2000 (3) SA 54 (SCA) para 18, Melunsky AJA said: ‘[a] court fulfils its function by attempting to give effect to the intention of the lawgiver…All other methods of interpretation should be considered with a view to arriving at the intention of the legislator’.
The interpretation contended for by the appellant, will obviously undermine the intention of the legislator which is to criminalise people who harbour or fail to report people who are suspected of having committed terrorist activities or who are likely to commit such activities.
 Some of the activities Van Rooyen and Gouws are alleged to have engaged in are classified as terrorist activities under the Terrorist Act. My view is that it is sufficient for the State to prove that Van Rooyen and Gouws are known or suspected or ought reasonably to have been suspected to have committed activities that could constitute terrorism under the Terrorist Act, or are likely to commit such activities.
 Counsel for the appellant further submitted that in the charge sheet, the State limited itself to the allegation that the escapees committed terrorist activities, and the State failed to prove that the escapees committed acts of terrorism.
 Section 88 of the Criminal Procedure Act 51 of 1977 (the Criminal Procedure Act) provides that where a charge sheet is defective for want of an averment which is an essential ingredient of the relevant offence, the defect shall, unless brought to the notice of the court before judgment, be cured by evidence at the trial proving the matter which should have been averred.
 The failure by the State to make reference to the suspicion of the persons being likely to commit terrorism or that the escapees were persons the appellant reasonably ought to have suspected of terrorism in the charge sheet was cured, as contemplated in s 88 of the Criminal Procedure Act, by the evidence led by the State.
 I am satisfied that the appellant harboured or concealed persons whom he suspected or ought reasonably to have suspected of committing terrorist activities or who are likely to commit terrorist activities. By his conduct the appellant contravened the provisions of s 11 of the Terrorist Act.
 The second count, as stated earlier relates to the alleged failure by the appellant to report the presence of the escapees on his farm to the police as required by s 12(1)(b) of the Terrorist Act. It is common cause that the appellant, despite his knowledge of the presence of the escapees on his farm, failed to report their presence to the police. His failure to report the presence of the escapees on his farm is consistent with his intention to harbour them. I am satisfied that the State has proved beyond any reasonable doubt that the appellant contravened both ss 11 and 12(1)(b) of the Terrorist Act.
Misdirection by trial court
 In his heads of argument counsel for the appellant submitted that the magistrate committed a serious misdirection by disallowing cross-examination pertaining to the reason and period that the observation unit of the police was on the farm, before the unit was discovered on 13 January 2007. He further submitted that the refusal to allow the question rendered the trial unfair as envisaged by s 34 read with s 35(3) of the Constitution.
 During cross-examination, Superintendent Vreugdenburg was asked how long the police task force members had been on the farm, when they were discovered. Superintendent Vreugdenburg refused to answer the question saying that he could not disclose the police investigating technique. The trial court ruled that the Inspector need not answer the question.
 In Shabalala v Attorney-General, Transvaal  ZACC 12; 1996 (1) SA 725 (CC), the court had to deal with, inter alia the right of an accused to a fair trial and the requirements of a fair trial in a given case. Mahomed DP para 52, said: ‘[t]he court in each case would have to exercise a proper discretion balancing the accused’s need for a fair trial against the legitimate interest of the State in enhancing and protecting the ends of justice’.
 Allowing the defence’s question in the context of this case, would have exposed the police investigation techniques whilst at the same time it would have been of no assistance to the accused. The accused’s need for a fair trial in this case, does not require the exposure of the police investigating techniques. Besides that, the evidence that the defence intended eliciting by the question mentioned earlier is irrelevant in the context of this case. The appellant did not suffer any prejudice. Consequently it cannot justifiably be said that the trial court misdirected itself in any manner.
 My view is that there is no basis for interference with the conviction of the appellant and therefore the appeal on conviction should fail.
 It is trite that punishment is pre-eminently within the discretion of the trial court. The court of appeal can interfere with the sentence imposed by the trial court only when it is vitiated by irregularity, misdirection or is disturbingly inappropriate. See S v Rabie 1975 (4) SA 855 (A) at 857D-F. In S v Le Roux 2010 (2) SACR 11 (SCA), para 35, Mlambo JA said:
‘As stated in many cases, which it is not necessary to cite, sentence is a matter for the discretion of the trial court, and a court of appeal must focus on whether that discretion was exercised judicially. As an appeal court we should be slow to interfere in sentences imposed by trial courts where the exercise of their discretion is beyond reproach’.
 At the time of imposition of the sentence, the appellant was 48 years old and a first offender. He was a farmer and he participated in the activities of the Commando in his area. He was married and he has at least one child.
 Counsel for the appellant contended that the sentence imposed induces a sense of shock and is therefore glaringly inappropriate. He also contended that a fine or suspended sentence would be appropriate. He relied on the case of Thatcher v Minister of Justice and Constitutional Development 2005 (4) SA 543 (C). Reliance on the Thatcher case is misplaced as it is distinguishable from this case. In the Thatcher case, unlike the present case, there was a plea bargain where the sentence was negotiated.
 In this matter there are various aggravating circumstances. The appellant, who is convicted of serious offences, showed no remorse. The escape of Van Rooyen and Gouws caused the State enormous financial expenses. The ‘Boeremag trial’ could not continue for several months. Attempts to trace and re-arrest both Van Rooyen and Gouws caused the State huge financial and human resources expenses.
 After considering all the circumstances of this case, including the personal circumstances of the appellant, the seriousness of the offence and the aggravating circumstances, I am not persuaded that there are any factors which justify any interference with the sentence imposed by the trial court. The offences committed by the appellant are of such a serious nature and in the circumstances of this case they call for direct imprisonment. A fine or suspended sentence would be wholly inappropriate.
 In the circumstances, I would dismiss
the appeal against both the conviction and sentence.
W L Seriti
Judge of Appeal
 Having read the judgment of Seriti JA, and while I agree with his finding that the appellant had indeed harboured and concealed the escapees Gouws and Van Rooyen at some stage after they had escaped from custody during the ‘Boeremag trial’ and that the appellant failed to report their presence on his farm to the police, I respectfully do not agree with his conclusion that the appellant thereby contravened either s 11 or s 12(1)(b) of the Protection of Constitutional Democracy Against Terrorist and Related Activities Act 33 of 2004 (the Terrorism Act). I am also of the view that the State failed to prove its alternative charge that the appellant had committed an offence under s 115(e) of the Correctional Services Act 111 of 1998. My reasons are set out below.
 At the outset it is necessary to consider the relevant provisions of the Terrorism Act.
(a) In order to secure a conviction under s 11, the State had to prove that an accused harboured or concealed a person who he or she knew, or ought reasonably to have known or suspected, ‘to be a person who has committed a specified offence as referred to in para (a) of the definition of “specified offence”, or who is likely to commit such an offence . . .’.
(b) The offences set out in sub-paragraph (a) of the definition of ‘specified offence’2 are:
‘the offence of terrorism referred to in s 2, an offence associated or connected with terrorist activities referred to in s 3, a Convention offence, or an offence referred to in ss 13 or 14 (insofar as it relates to the aforementioned sections)’ – I interpose to mention that only offences involving ‘terrorism’ and ‘terrorist activities’ are of relevance in this case.
(c) Section 2 provides that any person ‘who engages in a terrorist activity is guilty of the offence of terrorism (emphasis provided).
(d) ‘Terrorist activity’ as referred to both in part (a) of the definition ‘specified offence’ and in s 2, is defined as meaning (the reader is advised to take a deep breath): 3
(vi) is designed or calculated to cause serious interference with or serious disruption of an essential service, facility or system, or the delivery of any such service, facility or system, whether public or private, including, but not limited to-
(viii) creates a serious public emergency situation or a general insurrection in the Republic, whether the harm contemplated in paragraphs (a)(i) to (vii) is or may be suffered in or outside the Republic, and whether the activity referred to in subparagraphs (ii) to (viii) was committed by way of any means or method; and
(ii) intimidate, or to induce or cause feelings of insecurity within, the public, or a segment of the public, with regard to its security, including its economic security, or to induce, cause or spread feelings of terror, fear or panic in a civilian population; or
(iii) unduly compel, intimidate, force, coerce, induce or cause a person, a government, the general public or a segment of the public, or a domestic or an international organisation or body or intergovernmental organisation or body, to do or to abstain or refrain from doing any act, or to adopt or abandon a particular standpoint, or to act in accordance with certain principles, whether the public or the person, government, body, or organisation or institution referred to in subparagraphs (ii) or (iii), as the case may be, is inside or outside the Republic; and
(c) which is committed, directly or indirectly, in whole or in part, for the purpose of the advancement of an individual or collective political, religious, ideological or philosophical motive, objective, cause or undertaking.'
(d) Section 1 goes on further to define ‘terrorist and related activities’ as
‘. . . any act or activity related to or associated or connected with the commission of the offence of terrorism, or an offence associated or connected with a terrorist activity, or a Convention offence, or an offence referred to in ss 11 to 14.'
 The indictment served on the appellant and his wife was terse in the extreme. It largely amounted to a regurgitation of provisions of the Terrorism Act. In respect of the first count, it merely alleged that the appellants had contravened s 11:
‘In that the accused during the period 6 May 2006 to 13 January 2007 and on the farm Waterbessiefontein, Buffelspoort in the district of Waterberg, wrongfully and intentionally harboured or concealed persons, to wit Herman van Rooyen and Jan Rudolf Gouws, being persons in respect of whom the accused knew, or ought to have known or suspected, had committed a specified offence as envisaged in para (a) of the definition of “specified offence” in (the Terrorism Act) to wit terrorism and/or acts relating to terrorist activities as described in the said Act.'4
 In the light of the definitions already set out, the State was obliged to show that the two escapees were persons who had committed terrorist activities or acts relating to terrorist activities in order to obtain a conviction. But noticeable by its absence in the indictment is any reference to any specific act of terrorism or terrorist activity which the State alleged the appellants knew or ought to have known or suspected that the escapees had committed. Interestingly, as Seriti JA mentions in his judgment, when the matter was postponed in the magistrates' court on 16 February 2007, the State was ordered to furnish further particulars to the charge on or before 31 March 2007. Section 87(2) of the Criminal Procedure Act 51 of 1977 requires further particulars to a charge to be entered in the record. This was not done, presumably as the State did not so provide the particulars as ordered.
 It may well be debatable in the circumstances whether the indictment as framed is a proper charge which informed the appellant and his wife of the case they were to meet. Indeed, in the light of the definition of terrorist, the charge can be properly described as ‘. . . a vague, confusing, and conclusory articulation of the factual and legal basis for the claims and a general “kitchen sink” approach to pleading the case’5 ─ and I am surprised that the defence made no attempt to squash it. As no such attempt was made, I intend to proceed on the basis that the State was entitled to secure a conviction on proof that the appellant knew, or ought reasonably to have known or suspected, that the escapees had committed any of the myriad of actions referred to in the definition of terrorist activity or any act relating to any such activity.
 The State initially attempted to do this by proving certain handwritten notes made by the escapees after their escape which contained detailed plans to overthrow the government and to free their co-accused in the Boeremag trial from detention in a high security jail in Pretoria. However, although the trial court took this document into account as proof that the appellant knew or ought to have known that the escapees were intent on committing acts of terrorism, ultimately the State correctly accepted in this court that it had not shown that the appellant had any knowledge of either the escapees’ notes or their violent plans. The State was thus constrained to fall back on an argument that the appellant must have known that the escapees had been charged in the Boeremag trial with various offences under the Internal Security Act 74 of 1982 and that this, coupled with their escape, must have caused him at least to suspect that they were guilty of those charges. This in turn, so the argument went, established the appellant’s guilt under s 11 of the Terrorism Act.
 Essentially this argument is based on the contention that the actions of the escapees committed before the Terrorism Act came into operation, and which were the subject of charges under the Internal Security Act on which they were being tried when they escaped, are to be construed as ‘specified offences’ under the Terrorism Act. This was the approach of the high court which, in dismissing the appellant’s appeal, said:
‘. . . . it appears that Van Rooyen and Gouws were charged with, as I have indicated, offences of treason, attempted murder, murder and also offences under the Internal Security Act, which are in terms of that Act defined as terrorism. It is therefore fair to say that when Van Rooyen and Gouws escaped from custody they were then charged with conduct equivalent to that contemplated in the definition of terrorism or terrorist activities under the present legislation.’
 In my view, however, this is where the State’s case is fatally flawed. With effect from 20 May 2005, the Internal Security Act was repealed in its entirety by s 27 of the Terrorism Act. However, the latter section contained various ‘sunset’ provisions which read as follows:
‘(2) All criminal proceedings which immediately prior to the commencement of this Act were instituted in terms of the provisions of the Internal Security Act, 1982 (Act 74 of 1982), and which proceedings have not been concluded before the commencement of this Act, shall be continued and concluded, in all respects as if this Act had not been passed.
(3) An investigation, or prosecution or other legal proceedings, in respect of conduct which would have constituted an offence under the Internal Security Act, 1982, and which occurred after the commencement of that Act but before the commencement of this Act, may be conducted, instituted and continued as if this Act had not been passed.
(4) Notwithstanding the repeal or amendment of any provision of any law by this Act, such provision shall, for the purpose of the disposal of any criminal proceedings, investigation, prosecution or legal proceedings contemplated in subsection (2) or (3), remain in force as if such provision had not been repealed or amended.'
 Not only is it trite that legislation, particularly that giving rise to criminal liability, is presumed not to be retrospective in effect, but it is clear from these provisions that the legislature drew a line in the sand to distinguish between events giving rise to criminal liability which had occurred before 20 May 2005 when the Internal Security Act was repealed, and those which occurred thereafter. Under s 27(2) the prosecution of the charges against the escapees in the Boeremag trial is to continue as if the Terrorism Act had not been passed, and their actions which are the subject of those charges could therefore never be the subject of charges under the Terrorism Act. Moreover, under s 27(3), even if the escapees had not already been charged under the Internal Security Act before 20 May 2005, any criminal activities on their part before that date which constituted an offence under the Internal Security Act, fall to be charged under that Act and not the Terrorism Act. Any possible doubt in that regard is removed by the specific provisions of s 27(4).
 It is therefore clear that any criminal conduct which constituted an offence under the Internal Security Act that occurred before that Act was repealed is to be dealt with and prosecuted under that Act, and not under the Terrorism Act. That being the case, actions which constituted an offence under the Internal Security Act cannot be the subject of a charge brought under the Terrorism Act. To hold otherwise would be to apply the Terrorism Act retrospectively and to fly directly in the face of the legislature’s specific provision to the contrary.
 To regard offences under the Internal Security Act as specified offences under the Terrorism Act would give rise to absurdity. It would mean that if the appellant concealed and harboured the escapees well-knowing that they had committed offences under the Internal Security Act before its repeal, he could be convicted in terms of s 11 of the Terrorism Act of the offence of harbouring persons on the basis that he knew they had committed specified offences envisaged under the Terrorism Act while, in fact and in law, the escapees themselves could not be charged or convicted of those offences by reason of s 27 of that Act.
 Having regard to the provisions of s 54 of the Internal Security Act quoted by Seriti JA,6 I accept that actions of a person of a nature that would constitute the offence of terrorism under that Act might well also amount to terrorist activity (and therefore ‘terrorism’) under the Terrorism Act. However, as is clearly apparent from what I have said, those actions committed before the Terrorism Act came into operation on 20 May 2005 cannot be regarded as terrorist activities under the Terrorism Act and, thus, cannot be specified offences as envisaged under that Act. Had that not been the legislature’s intention, it could easily have specifically provided for specified offences to include offences under the Internal Security Act. Not only did the legislature fail to do so but it clearly and specifically provided for offences committed before the Internal Security Act was repealed to be dealt with as if it had not been repealed.
 As this issue was not ventilated until it was raised with counsel during the appeal, both sides were allowed to submit written argument on the point. The reasoning I have set out was gleefully adopted by counsel for the appellant. On the other hand counsel for the State, realizing the problem he faced, submitted in his written argument that this court should modify the legislature’s language to rule that terrorism under s 54 of the Internal Security Act ‘is conduct equivalent to conduct contemplated in the definition of terrorist activities in the present legislation’.7 However, although it may at times be permissible to modify statutory provisions by reading into a statute something which has been clearly omitted, a court must guard against making rather than interpreting legislation, and it certainly cannot supplement legislation if the effect of doing so flies in the face of the legislature’s clear and specifically expressed intention. Here the clear intent of the legislature is for the Terrorism Act not to operate retrospectively, and offences under that Act must thus be limited to conduct after it came into operation. There is just no room to supplement the Terrorism Act in the manner suggested.
 The State therefore cannot look to the alleged criminal activities of the escapees under the Internal Security Act to prove that they were persons whom the appellant knew, or ought to have known or suspected, had committed an offence referred to in para (a) of the definition of ‘specified offence’ in the Terrorism Act. As this was the only conduct of the escapees relied upon, the State failed to prove an essential element of its main charge under s 11 (count 1 if the indictment) and the appellant was wrongly convicted on that charge.
 On a parity of reasoning, the appellant was also wrongly convicted of an offence under s 12. In that charge (count 2) it was alleged he had failed to report the presence of the escapees on his farm to the police, well knowing that they were persons who had committed terrorism or terrorist activity related offences as envisaged by the Terrorism Act. Again the State relied on the escapees’ actions which were the subject of the charges brought against them under the Internal Security Act as proof of this fact. And again, it must fail as the conduct underlying such charges cannot amount to terrorism or acts related to terrorist activities under the Terrorism Act.
 As the convictions under neither s 11 nor 12 of the Terrorism Act can stand, it is strictly speaking unnecessary to consider the issue of whether the State had complied with the provisions of s 16(1) of the Terrorism Act. That having been said, if the section is to be purposively interpreted the factors mentioned by Seriti JA in his judgment may well lead to the conclusion that a prosecution under chapter 2 of the Terrorism Act is only ‘instituted’ when the accused is asked to plead. However, even if I had concluded otherwise in regard to the charges under the Terrorism Act, it would have been unnecessary to decide the issue. A similar issue arose in this court in both Moodley8 and De Vries9 in regard to s 2(4) of the Prevention of Organised Crime Act 121 of 1998 (POCA) which is similar in nature and wording to s 16(1) of the Terrorism Act in that it provides that an accused ‘shall only be charged with committing an offence’ envisaged by s 2(1) of POCA if a prosecution is authorised by the National Director of Public Prosecutions. As in the present matter, in both those cases the necessary authorisation had been made available before the appellant was asked to plead. In each case this court decided that once the required written authorisation had been handed in, the prosecution was thereafter lawful and that it was thus unnecessary to decide at what stage a person was ‘charged’ as envisaged by s 2(4) of POCA. On a parity of reasoning it would be unnecessary in the present case to reach a decision on precisely when a prosecution is ‘instituted’ as envisaged under s 16(1) of the Terrorism Act, as that section had clearly been complied with before the charges were put to the appellant and he was asked to plead thereto. Accordingly, had I not reached the conclusion which I have in regard to the charges under the Terrorism Act, I would probably have found it unnecessary to decide the issue.
 I turn to the alternative charge on count 1. On this count it was alleged that by harbouring the escapees on his farm the appellant had contravened s 115(e) of the Correctional Services Act 111 of 1998 which, as it read at the time, provided that any person who ‘harbours or conceals or assists in harbouring or concealing an escaped prisoner’ commits an offence (the Act has since been amended to delete reference to a ‘prisoner’ and to replace it with an ‘inmate’).10
 Again it is necessary to give close consideration to the statutory provisions relevant to this issue.
‘. . . any person, whether convicted or not, who is detained in custody in any prison or who has been transferred in custody or is on route from one prison to another prison’.
(b) In turn, a ‘prison’ was defined as meaning (again it would be best for the reader to take a fairly deep breath):
‘. . . any place established under this Act as a place for the reception, detention, confinement, training or treatment of persons liable to detention in custody or to detention in placement under protective custody, and all land, outbuildings and premises adjacent to any such place and used in connection therewith and all land, branches, outstations, camps, buildings, premises or places to which any such persons have been sent for the purpose of imprisonment, detention, protection, labour, treatment or otherwise, and all quarters of correctional officials used in connection with any such prison, and for the purposes of sections 115 and 117 of this Act includes every place used as a police cell or lock-up.’ (My emphasis.)
 It hardly needs to be said that ‘custody’ in the definition of ‘prisoner’ must mean lawful custody. In order to establish that Van Rooyen and Gouws were in lawful custody when they escaped, the State proved that after the Boeremag trial had begun they had unsuccessfully applied for bail, their application having been refused by the high court on 26 July 2004. For present purposes I am prepared to accept that the effect of this was that Van Rooyen and Gouws were ordered to remain in custody and that this order was operative at the time of their escape.
 In an attempt to get around this difficulty, counsel for the appellant relied on s 6(1)(a) of the Correctional Services Act 111 of 1998 which provides that ‘[a] person may not be committed to a prison without a valid warrant for his or her detention’. The evidence established that at the relevant time the escapees, and those of their co-accused who were in detention, were being held at the C-Max Prison in Pretoria, from where they were taken daily to attend their trial in the high court. Each day a warrant was issued in respect of each of them, directing the head of the prison to take the person concerned into custody and to redeliver him to court the next day. The day before the escape in question, such a warrant was issued for each escapee.
 A warrant of detention amounts to a directive issued by the court obliging the prison authorities to hold a person in custody. (There are also other important consequences which flow from the issue of such a warrant, unnecessary to detail for purposes of this judgment).11 It hardly needs to be said that the issue of a warrant of detention (generally done using a form called a ’J7’) is therefore a serious matter. Unfortunately, the warrants relied on by the State were sloppily prepared using a standard form to be issued by a magistrate, despite the trial in question being conducted in the high court. But more importantly, although they bear the stamp of the registrar of the high court, each warrant was signed (purportedly on behalf of the magistrate) by a court orderly, a policeman by the name of Nkandiwiri who held the rank of inspector.
 Understandably, the validity of this warrant was attacked by counsel for the appellant who, in doing so, relied heavily on the judgment in S v Motsasi 1998 (2) SACR 35 (W) in which it was held that although there is no statutory requirement to such effect, a warrant of detention should be signed by the presiding judge or by the registrar or a senior official in the registrar's staff.12 If this is indeed correct, the issue of a warrant by court orderly would not amount to proper compliance. I should mention as an aside that Superintendent Vreugdenburg, when testifying stated that problems had been experienced in getting court orderlies to issue warrants and that he and his investigating team had taken to issuing them. If the judgment in Motsasi is correct, such warrants may also be invalid.
 Be that as it may, in the light of the view I have of the matter, it is unnecessary to reach a final conclusion on the validity of the warrants as, for a number of reasons, it is an issue of no relevance.
(a) First, what counsel for the appellant did not mention when relying upon Motsasi, is that the court in that matter went on to hold that where a person is being detained under a judge's order, such detention does not become unlawful merely because there is not a valid warrant of detention.13 That conclusion must surely be correct. The lawfulness of the detention flows from the court’s order not the manner of detention.14 An error in the completion of a warrant, which might visit it with invalidity, cannot override the order of a judge. Thus even if the warrants were invalid (and my prima facie view is that they were) this did not result in Van Rooyen and Gouws being held unlawfully in custody.
(b) Moreover, even if the warrants were valid, they operated only until Van Rooyen and Gouws were returned to the high court on 3 May 2006, and were then discharged. They could not, and did not, direct the head of the prison to detain them thereafter. It is not suggested that either Van Rooyen and Gouws escaped from the C-Max prison during the currency of the warrants - in fact as set out below, they escaped during the course of the court's lunch adjournment.
(c) Inspector Etsebet, who was at court on the day of the escape, testified that there are police cells at the high court to hold persons in custody. As appears from the definition of prison in s 1 of the Correctional Services Act quoted above, such a cell or lock-up is regarded as being a prison ‘for the purposes of s115 and117’. Those sections relate to escaping and aiding escapes, and visit certain conduct with criminal liability. Accordingly it is only if a person commits an offence under those sections (eg by escaping or aiding and escape) that for that limited purpose the cell or lock-up is to be regarded as a prison. As it would not otherwise be regarded as a prison (ie if no such unlawful conduct occurs) it is not necessary for a person to be detained in a cell lock-up under a warrant as envisaged by s 6(1). In addition, the various provisions of the latter section – which, inter alia, require that as soon as possible on admission the prisoner be given a bath or shower, undergo a health examination and be provided with the written information relating to various rules and disciplinary requirements – are all requirements clearly intended to apply only where a prisoner is taken up in a prison as defined and not merely held for a while in a police cell or lock-up. This too indicates that the requirement of a warrant does not apply when persons are held in police cells at court.
 Consequently, when Van Rooyen and Gouws were returned to the high court on 3 May 2006, the police were entitled to hold them without a specific warrant. And as they escaped after their return, the debate on the validity of the warrants issued the previous day, but discharged, is thus irrelevant. For present purposes I am prepared to accept that they were lawfully in police custody, but this does not mean that the State has proved that the appellant committed an offence by harbouring them after their escape which, for the reasons set out by Seriti JA in his judgment, I accept the appellant clearly did.
 As I have said, s 115 makes it an offence to harbour an escaped ‘prisoner’. This is in contradistinction to s 117 which makes it an offence for ‘any person’ to escape from custody (as Van Rooyen and Gouws appear to have done). While a person in custody is commonly referred to as a ‘prisoner’, that cannot be the case here where the word can only have the meaning ascribed to it by the legislation. By definition, a ‘prisoner’ must be a person in custody as that is an element of the definition. But the converse is not necessarily the case. The definition of prisoner also requires him or her to be ‘detained . . . in any prison or who has been transferred in custody or is on route from one prison to another prison’. There may be many people who are in custody who, by reason of this further requirement not having been satisfied, are not prisoners as defined. But s 115 does not make it an offence to harbour ‘any person’ who escapes from custody – it relates solely to the harbouring of escaped prisoners, and as it is a provision which creates a criminal offence it is to be strictly construed. An ‘escaped prisoner’ under s 115 therefore cannot be interpreted as meaning ‘any person who escapes’ as envisaged in s 117 – the latter having a far wider connotation.
 The issue is thus not whether the appellant harboured any people who had escaped from custody. To secure a conviction the State had to prove that Van Rooyen and Gouws were prisoners who had escaped from a prison. But as Van Rooyen and Gouws had absconded from court, the State had to fall back on the provision in the definition of prison that for the purposes of s 115 every place used as a police cell or lock-up is to be regarded as a prison. In order to succeed on this issue, it was incumbent on the State to show that the escape had been made from a police cell lock-up at the court. Without doing so the State would have failed to prove an essential element of an offence under s 115(e).
 It is on this issue that the State’s case flounders. The only evidence in regard to the escape was that of Superintendent Vreugdenburg and Inspector Etsebet who both tersely stated that Van Rooyen and Gouws were present when court adjourned at 12h45 on 3 May 2006 but were missing at 14h00 when court reconvened. Apart from implying that Van Rooyen’s sister, who worked as a stenographer in the court building may have assisted as she knew the layout of the building, Vreugdenburg gave no further evidence in regard to how or from where the escape was made. It seems improbable that if Van Rooyen and Gouws did escape from a cell or lock-up, Vreudenburg would not have mentioned it. It may well be that Van Rooyen and Gouws were left unattended at some stage and took advantage of that state of affairs to make good their escape. Indeed, according to the hearsay evidence of Etsebet they escaped between the court and the cells where they did not arrive to take their meal But quite what happened, one does not know. However, the one thing the State did not prove is that their escape was effected from a cell or a lock-up. In addition, if they escaped while in the courtroom (which in my view is clearly not a prison) or even on their way from there to the cells, I do not see how that can be construed as having occurred while they were being ‘transferred in custody or (were) en route from one prison to another prison’ as set out in the definition of prisoner. The clause ‘from one prison to another prison’ clearly qualifies both the transfer in custody and the en route requirement, and bearing in mind the necessity to construe that provision strictly, I cannot accept that the escapees were on their way from one prison to another either while in court or when on their way to the cells, even if those cells are to be regarded as a prison.
 Accordingly, in my judgment, the State did not establish that Van Rooyen and Gouws were prisoners as defined at the time that they escaped and, that being so, it failed to prove that the appellant harboured escaped prisoners. At the most the State proved that he had harboured persons who had escaped from custody, but that was not an offence under s115 with which he was charged. It therefore failed to prove any of the offences levied against the appellant in its indictment, and his convictions and sentences must be set aside.
 I am aware that I have reached this conclusion by way of a point not argued by counsel for the appellant. But whether Van Rooyen and Gouws were prisoners as defined has always been a live issue. That they were not prisoners at the material time was not a defence which the appellant was obliged to raise. It was at all times incumbent upon the State to prove every fact necessary to prove the guilt of the appellants and, to do so, it had to prove that the escape had been made at a time when Van Rooyen and Gouws were prisoners. The defence attempted, unsuccessfully, to show that they had not been prisoners as they were not in lawful custody due to the alleged defects in the warrants, but that did not absolve the State from having to prove the other facts necessary to show that they were prisoners within the ambit of the definition. Nor did the failure of the appellant’s counsel to raise the issue at the trial alter that situation. The defence was not called on to challenge the State’s evidence which merely established the approximate time of the escape, and no more. There was thus no obligation to challenge any aspect of the State’s evidence in regard to the escape, and there was no obligation on the defence to point out to the prosecution the hiatus in its case. As was observed by the Constitutional Court in S v Molimi  ZACC 2; 2008 (2) SACR 76 (CC) para 40:
‘There is no obligation on the defence to assist the prosecution in the execution of its duties and the advancement of its case. If that was so, an unwarranted burden would be imposed on the accused who has to contend with the allegations levelled against him or her.’
 In these circumstances, not only is there no room for the State to have assumed that the defence accepted that Gouws and Van Rooyen were prisoners at the time they escaped, but there can be no question of the State having been prejudiced by the failure of the defence to raise the issue. The duty of an appeal court is to consider whether the trial court reached the correct conclusion on the evidence led before it and, in doing so, to consider any legal issue which is relevant. Whether the proven facts establish the commission of the crime with which an accused is charged is clearly a question of law,15 and it would lead to an ‘intolerable situation’ if this court was bound to confirm a decision that is ‘clearly wrong’.16 As raising this legal issue on the evidence led can cause no prejudice to the State, and as the liberty of the subject is involved, the issue is one which this court can raise mero motu.
 I must express my disquiet about this outcome. I gravely suspect that the appellant was guilty of a criminal offence by harbouring the two escapees. However, the prosecution appears to have been distracted by the futile exercise of trying to link him to the escapees’ notes of their plans for violent insurrection and, possibly as a result, lost focus of the essential elements it had to prove to establish criminal liability on the appellant's part. But whatever my suspicions may be, the State was required to prove an offence beyond a reasonable doubt and, in my view, it failed to do so. If the appellant was in fact guilty of any offence, the State has only itself to blame.
 My conclusion in regard to the merits of the convictions renders it unnecessary to deal with any further issues raised. For the sake of completeness, I should mention that my prima facie view in regard to whether the trial court misdirected itself is that the issue of how long the police had been keeping the appellant's farm under observation before 13 January 2007 was not privileged and Vreugdenburg ought to have been directed to answer the questions put to him in this regard. On the other hand, I doubt that this gave rise to an unfair trial. No further comment on this is required.
 For the reasons set out above, I would uphold the appeal and set aside the appellant's convictions and sentences.
L E Leach
Judge of Appeal
For Appellant: GC Muller SC
Boonzaaier & Du Plessis Attorneys, Rietfontein
For Respondents: AG Janse van Rensburg
The National Prosecuting Authority, Pretoria
1Section 54(4) of the Internal Security Act provided:
‘Any person who has reason to suspect that any other person intends to commit or has committed any offence referred to in subsection (1), (2) or (3) and any person who is aware of the presence at any place of any other person who is so suspected of intending to commit or having committed such an offence, and who–
harbours or conceals that other person;
fails to report or cause to be reported to any member of the police such presence of that other person at any place, as the case may be, shall be guilty of an offence and liable on conviction to the penalty to which the person whom he so harboured or concealed or to whom he so rendered assistance or whose presence he so failed to report or cause to be reported would have been liable on conviction of the offence which the last mentioned person intended to commit or committed, as the case may be’.
2Set out in s 1 of the Terrorism Act.
3In s 1 of the Terrorism Act.
4This is my translation of the indictment which was written in Afrikaans.
5Reportedly the comment of a judge in a federal appeal court in the United States – see http://www.abajournal.com/news/article/7th_circuit_slaps_lawyer_for_unintelligible_writing_full_of_gibberish.
6Para 57 above.
7I quote the written argument.