The Official Secrets Act 1989: the Right
Balance?

1. N Stevens'*'

The celebration of the centenary of a controversial Act of Parliament by means of
the enactment of yet another, sharing the same"title and basic purpose, must be
something of a rarity. Yet this is precisely what has occurred this year, with the
granting of the Royal Assent to the Official Secrets Act 1989 on May 11. The
commencement date has yet to be announced at the time of writing (September,
1989).

The new Act has not closed the circle, however: far from it, in fact. Whilst, prior
to the Official Secrets Act 1889, no criminal sanctions applied to the unauthorised
disclosures of "official" information, not only has the 1989 Act retained the use of
such sanctions, it has also provided for offences of strict liability in connection with
some kinds of disclosures. Whether criminal sanctions are necessary, or even
desirable, in connection with disclosures which fall short of espionage, or are in no
way of assistance to an enemy or potential enemy, is itself a highly vexed question.
This article does not address that issue as such, but seeks to assess the new Act
mainly in terms of the objectives which, according to the Government, it sought to
achieve. In this context, the necessity of resorting to criminal sanctions in some
cases has been more-or-Iess assumed: the question, therefore, is: will these be
directed towards the appropriate cases? In a world in which "freedom of
information", "the public interest" and "open Government" are, increasingly,
ntuch~used (some would say over-used) phrases, it is essential that a piece of
legislation that imposes, or retains, criminal sanctions in cases of disclosures of
even delicate information must strike the right balance. All of the above interests,
as well as the occasional requirements of confidentiality, and even the risk of harm
to the public service or other branches of Governmental activities, must be
weighed in the balance if one is to answer that question satisfactorily. This, the
Government claims to have done. The purpose of this article is to examine that
claim, against the background of a century of Official Secrecy.

The growth of official secrecy and background to the Act
It is strange, perhaps, to contemplate that the first Official Secrets Act was

passed in 1889. Prior to that date, no criminal sanctions existed for preventing

"Of The School of Law, University of Buckingham.

169



THE DENNING LAW JOURNAL

cases of "breach of official trust"l as such. Instead, the authorities had to rely on
mainstream criminal law, such as the Larceny Act 1861. The best known example
of a case which highlighted this gap in the law - indeed, the case which is
commonly thought to have precipitated the 1889 Act - was that of Charles Marvin,
in 1878. Marvin was a temporary clerk at the Foreign Office, who supplemented
what he described as a "miserable" wage by writing articles for The Globe. One
such article contained the gist of a secret treaty, to which Marvin had access in the
course of his employment, and which (being endowed with a prodigious memory)
he was able to recall and duplicate at The Globe's offices. Later, he made a copy of
the entire treaty, taking the precaution of using his own notepaper, and this, too,
was published.

Marvin was identified as the source of what was a politically embarrassing leak.
The case against him was dismissed, however, since he had not actually removed
the document, nor even stolen the paper used to transcribe it.

In 1888, a "Breach of Official Trust" Bill was introduced in the Lords. On
receiving the Royal Assent, it became the Official Secrets Act, 1889. Like its
successor Act of 1911, this Act aroused controversy, not least in the manner in
which it underwent its Commons' stages. Like the 1911 Act, it sought to address
two distinct problems: espionage (in Section 1); and unauthorised disclosures of
official information (in Section 2). In both cases, however, the 1889 Act was
somewhat narrower in scope than that of 1911; section 2, in particular, since it
applied criminal sanctions to unauthorised disclosures by Crown servants or
contractors only. Furthermore, such disclosures were unlawful only when made
"to any person to whom [the information] ought not, in the interest of the State or
otherwise in the public interest, to be communicated ... ,,2 The Act (and
particularly section 2) was not especially effective (few successful prosecutions
being brought) so that, by the turn of the century, attempts to strengthen its
provisions were being made.

For many years, there was a common misconception that the Official Secrets
Act 1911, passed hurriedly as it was,3 was a piece of emergency legislation, in
response to the growing threat of German activities including espionage. The
1909 report of a Defence Committee, however, suggests that not only the contents
of the new Bill (and especially section 2), but even the manner of its introduction
in Parliament, had been carefully considered, and planned, well in advance of
1911.4

1. The original tide of the Bill.
2. This is, of course, the basis of the "public interest" defence which, particularly since the Pontillg

case (infra) has been so controversial, and which is conspicuously absent from the 1989 Act.
3. It underwent all its Commons stages in the afternoon of August 18; s.1 was not even debated!
4. A newspaper had published details of warships. Lamenting their inability to prosecute the

newspaper (as opposed to the source, ifknown) under the 1889 Act, .the Committee's report comments:
"We ought to be in a position to prosecute the proprietor of the newspaper ... "; and later, " ... such a
Bill would excite less opposition if it were introduced by the Secretary of State for War than by the
Home Secretary, and that this might be done on the plea of it being a measure of precaution of great
importance in national defence".

170



THE OFFICIAL SECRETS ACT 1989

The 1911 Act "strengthened" the law in a number of ways. Thus (regarding
Section 1) whilst the 1889 Act had been silent as to the burden of proving the
required intention on the part of th~ accused, section 1(2) of the 1911 Act clearly
relieved the prosecution of any such burden; and placed the burden of dispruving
such an intention on the accused.5 Moreover, the prescribed criminal intention
was itself broadened to include "any purpose prejudicial to the safety or interests
of the State." The range of both "prohibited places" and "prohibited conduct"
were also extended, so that (for example) it became criminal merely to "approach"
or be "in the neighbourhood" of a "prohibited place".6 (The 1889 Act had been
narrower, requiring actual entry into one of the places actually itemised in the Act,
or, if outside it, making a sketch or plan of such a place). Finally, the 1911 Act
"upgraded" section 1 offences to felonies, punishable by penal servitude7 for
between three and seven years.

The extended scope of section 1 can be seen in the case of Chandler v. D.P.P., 8
which established that the provisions of section 1 could be used in cases of
sabotage, even where no damage was done, and this despite the margin heading
"Penalties for Spying". In the Chandler case, some C.N.D. protestors sought
temporarily to incapacitate Wethersfield airbase (where NATO command aircraft
were based) by means of a peaceful "sit-down" type protest. Their long-term
purpose, they maintained throughout, was nuclear disarmament. This, they
argued, was positively beneficial to society, and could hardly, therefore, be
described as "prejudicial to the safety or interests of the State.,,9

They were nevertheless convicted, and appealed (eventually) to the House of
Lords, who upheld the convictions. In his judgment, Lord Reid stated: "
... 'State' is not an easy word. It does not mean the Government or executive ...
perhaps the country, or the realm are as good synonyms as one can find ... ". Later
he added,1O arguably perpetuating - albeit inadvertently - the misconception
concerning the Act's genesis, " ... The 1911 Act was passed at a time of grave
misgivings about the German menace, and it would be surprising and hardly

5. It stated: " ... if any [information] relating to ... any prohibited place within the meaning of this
Act ... is made, obtained or communicated by a person other than a person acting under lawful
authority it shall be deemed to have been made ... for a purpose prejudicial to the safety or interests of
the State unless the contrary is proved" ("a purpose prejudicial to the safety or interests of the state"
being the intention prescribed by the Act).
6. "Prohibited places" now included, as well as arsenals, dockyards, camps and the like, "any work of

defence, arsenal dockyard, camp, ship, telegraph or signal station belonging to His Majesty", and any
other place, whether belonging to His Majesty or not where ships, arms or other "defence" materials
were built, stored, made or repaired. Furthermore, the Act empowered. the Secretary of State to
"declare" such things as railways, roads or other means of communication, or any place "belonging to
His Majesty" to be a "prohibited place" for the purposes of the Act: Official Secrets Act 1911, s.3.
7. Till its abolition in 1948: Criminal]ustice Act 1948, 5.1. The actual term was increased once more

in 1920 to a maximum of fourteen years: Official Secrets Act 1920, s.8(1).( The distinction between
felonies and misdemeanors itself was, of course, abolished in 1967: Criminal Law Act 1967, s.1.)
8. [1964] A.C. 763.
9. Supra.

10. At p.701.

171



THE DENNING LAW JOURNAL

credible that the Parliament of that date intended that a person who deliberately
interfered with vital dispositions of the armed forces should be entitled to submit
to a jury that Government policy was wrong and that what he did was really in the
interests of the State." At all events, the case seems to have decided that "interests
of the State" could not be divorced from the policy of the Government of the day,
and that, for practical purposes, the two are synonymous.

Apart from the precise legal issues involved, the decision was not a popular one,
and prompted the Attorney-General I1 to make a statement to the effect that
prosecutions under section 1 should really be confined, as far as possible, to cases
of espionage or (at least) clear cases of sabotage. The section was not, for instance,
used against the Greenham Common women, though several were prosecuted for
(inter alia) criminal damage and offences of obstruction. The Attorney's statement
was taxed, however, in the notorious A.B. C. trial in 1978,12when section 1 charges
were again brought. Whilst acknowledging that section 1 extended to cases of
sabotage, as well as espionage, the trial judge, Mars-Jones J, indicated that, in
cases of this kind, involving "investigative journalism", such charges were
"oppressive". The section 1 charges were quietly dropped, though the defendants
were convicted of various charges under section 2. Less controversially, section 1
has been used in the various celebrated "spy-trials", such as those of George
Blake in 1960, and of Michael Bettany and Geoffrey Prime in the early 1980s.

Since it has been untouched by the 1989 Act, section 1 remains important. It is
to be hoped, however, that the Government's failure to reform it does not indicate
a willingness for prosecutions under section 1 to be brought in cases like Chandler
and the A.B. C. case in future, particularly since recent legislation may well prove
(at least in cases like Chandler) adequate to deal with any problems which might
otherwise arise. 13

To turn now to section 2 of the 1911 Act. Until its repeal in May 1989, it
remained one of the most controversial of all statutory provisions. Even some
judges have roundly condemned it; one, in 1970, suggesting it should be
"pensioned off".13 Its reform, however, has been problematical. Indeed, even
when it came about, it took an entire Act of sixteen sections to repeal it!

The main problem with section 2 - though some would arguably see it as a
strength - lay in a single phrase: 'catch-all'. As it stood, section 2 was capable of
making virtually any unauthorised disclosure of any "official" information
whatsoever, by any person whatsoever, a criminal offence. As well as extending
criminal sanctions to persons other than crown servants and contractors, section 2
made the mere receipt of "official" information an offence, unless this occurred

11. Whose consent is necessary for any prosecution under the Act: Official Secrets Act 1911, s.8.
12. [1978J Crim. L.R .. and named because of the defendant's surnames: Aubrey, Berry and Campbell.
The first and last of these were journalists.
13. See, e.g., the Public Order Act 1986, S.38.
14. Per Caulfield J, in connection with the prosecution of Jonathan Aitken (then a parliamentary
candidate and journalist), with others, over the "Nigerian Report" (as Franks puts it).

172



THE OFFICIAL SECRETS ACT 1989

"contrary to the desire" of the receiver. It also narrowed the scope of the "public
interest" defence,15 and created (as amended)16 a further offence, namely failure
"to take reasonable care" of "official" information or documents. The history of
section 2 is a chequered one. It has been used in connection with many different
kinds of unauthorised disclosures, some of which would seem to pose little or no
threat to the public interest, or state security. In 1926, for instance, in the first case
where a section 2 prosecution was brought in connection with "journalistic"
activities, the retired Governor of Pentonville Prison was convicted after he had
written an article, published in the Evening News, and entitled "What Bywaters told
me". The article revealed statements made by a convicted murderer, Frederick
Bywaters, on the eve of his execution. The defendant's claim that these revelations
had not been contrary to the public interest did not avail him. He was fined £250
and ordered to pay costs. The editors of the newspaper were not prosecuted. The
first prosecution of the author of an article based on information supplied to him
by an official in contravention of the Act was not brought until 1963. Other cases
of this kind included those of Jonathan Aitken (and the Editor of the Sunday
Telegraph) in 1970, and the A.B. C. defendants in 1978. Many prosecutions have
been brought against civil servants (in recent years, most notably Sarah Tisdall in
1984, and Clive Ponting in 1985)17 and other officials, including policemen and
ex-policemen, Post Office employees, and even a clerk in the Probate Office, who
"leaked", prior to its official release, information concerning wills.

Whilst the statistics are incomplete, the following figures18 based on Home
Office ones, are fairly reliable as to section 2 prosecutions. In the period
1945-1955, eleven prosecutions were brought, all but one being successful. In the
period 1955-1985, fifty-one individuals and one company were prosecuted,
resulting in thirty-six convictions. In four cases charges were withdrawn.

The reform of section 2
There have been, over the years, several attempts at the reform of section 2.

The first significant proposal was made in 1965, when JUSTICE (the British
'branch' of the International Commission of Jurists) suggested limiting the scope
of section 2 so that disclosures should be unlawful only if they threatened the
national interest, or were made in breach of an undertaking of confidentiality. 19 In
1968, a committee investigating the Civil Service20 reported in general terms to

15. This, in time, received an extremely restricted interpretation by the courts: see, e.g., R. v. Ponting
[1985] Crim. L.R. 318.
16. Official Secrets Act 1920, s.9.
17. See, infra.
18. Based on Home Office figures, as supplied (inter alia) to the Franks Committee.
19. In fact, a reforming measure was taken in 1939, when the extensive powers ofinterrogation (placing
a duty on "every person" to give "any information" relating to an offence under the Act and, if
required, attend at a police station) created by the Official Secrets Act 1920, were limited to section 1
offences: Official Secrets Act 1939. This did not, of course, affect s.2. as such.
20. The Civil Service Cmnd. 3638 (1968) (The Fulton Report).

173



THE DENNING LAW JOURNAL

the effect that administration was "surrounded by too much secrecy", with the
result that the Wilson government published a White Paper, entitled "Information
and the Public Interest", the following year.21 This contained proposals for a
degree of "freedom of information". It did not, however, seek to dismantle, or
otherwise radically reform section 2.

The first major proposal to do that came from the Franks Committee, which
reported in 1972,22 having taken evidence from some 114 persons and bodies,
including many representatives of the press, broadcasting and the Law, as well as
Government Departments. Its proposals were particularly significant because they
formed (in part, at least) the basis of subsequent recommendations,23 as well as a
Government Bill in 1979, and were certainly carefully considered by the
Government in reaching its most recent recommendations (culminating, of course,
in the Official Secrets Act, 1989). The essence of Franks' proposals was that
section 2 should be repealed and replaced by narrower and more specific
provisions,24 the effect of which would have been to apply criminal sanctions to
disclosures of specified types of information only: m:mely, those relating to
defence and internal security; foreign relations; currency; maintenance of law and
order; and information given by private individuals in confidence. In relation to the
first three of these, Franks proposed that the Whitehall system of classification of
information should be adopted, so that a prosecution should only proceed if a
Minister were to certifY that the information was "secret", and that its
unauthorised disclosure would cause serious injury to the interests of the nation.
This requirement Franks described as the "touchstone" of his proposals. It was
also suggested that sanctions should be limited to specified classes of persons, in
that the "primary duty" would have fallen on to identified classes of persons,
including Ministers, Civil Servants, members of the Armed Forces, the Atomic
Energy Authority and Post Office, civilian personnel in the above, and former
members of any of the above classes. Other persons would be guilty of an offence
only if they communicated information, contrary to the proposed Act, which they
knew or had reason to believe had been given to them in contravention of the Act.
Mere receipt of official information should cease to be an offence.

The Official Secrets Act 1989
The White Paper of 1988 pointed out that "The recommendations of the

Franks Committee have naturally dominated discussion of this issue since they
were published in 1972, and have remained a necessary and valuable reference
point for the Government's present consideration of the reform of section 2.,,25

21. Cmnd. 4089 (1969).
22. Report of the Departmental Committee on Sec/ion 2 oflhe Official Secrels Ac/1911, Cmnd. 5104 (1972).
23. See: Home Office Reform of Sec/ion 2 of the Official Secrels Ac/, 1911 (Commons Paper 7285, 1978);
Green Paper Open Guvernmenl (Cmnd. 7520, 1979).
24. Which, Franks proposed, should take the form of an "Official Information Act".
25. Cmnd. 408 (1988), at p.5.

174



THE OFFICIAL SECRETS ACT 1989

Unsurprisingly, therefore, the Official Secrets Act 1989 follows the Franks model
in what are, perhaps, its most important particulars. Thus, the categories of
information covered by the new legislation are limited, though they are not
precisely the same categories as those proposed by Franks. They are information
relating to: security and intelligence (5.1); defence (5.2) and international relations
(5.3); information which might result or assist in the commission of an offence, or
escape from custody, or impede the apprehension or detection of offences or
which concerns interceptions made lawfully under the Interception of Com-
munication Act 1985 or Security Services Act 1989 (5.4); and information relating
to security or intelligence, defence, or international relations, which has been
communicated in confidence to another state or to an international organisation
(5.6). The Act also incorporates something very like the Franks "touchstone", in
that, with some significant exceptions, unauthorised disclosures are criminal only
if some harm results, or is likely to result. The Government did not, however,
adopt Franks' proposed system of Ministerial certification, preferring instead to
define the required harm in connection with each category of information. Thus,
for instance, damage to the capability of the armed forces to carry out their task, or
the risk ofloss of life or injury to servicemen, or the endangering of u.K. interests
abroad, indicate the kind and degree of harm required under section 2.

The exceptions to the above arise under sections 1 and 4. Since unlawful
disclosures under section 4, by their nature, tend to produce obvious harm (such
as the commission of an offence), no additional test of harm was felt necessary.
Section 1(1), on the other hand, singles out members or ex-members of the
security and intelligence services and "notified" persons26 for special treatment,
since unauthorised disclosure by them of any information relating to security or
intelligence obtained by virtue of their work is an absolute offence. In cases of such
disclosures by Crown servants and contractors a modified test of harm applies, in
that a disclosure is damaging if it either causes (or is likely to cause) damage to the
work of the security and intelligence services or is of information which falls within
a class or description of information, documents or articles, the unauthorised
disclosure of which would be likely to cause such damage. One reason, advanced
by the White Paper, for drawing this distinction was that, due to the sensitivity of
security and intelligence matters, the requirement that 'harm' be established in
connection with a disclosure by a member or ex-member might entail the bringing
of evidence which involves a disclosure at least as damaging as that which is the
subject of the prosecution.27 It is hard to resist the impression, however, that a
certain degree of reaction to the "Spycatcher" revelations by ex-MIS officer Peter
Wright is here involved.

26. A person may be "notified by a Minister that he is subject to the provision of s.1.(I) if, in the
Minister's opinion, his work is such that the interests of national security require that he should be
subject to s.I(I), by reason of its being connected with the security and intelligence services": s.1.(6).
(Arguably, there are shades of Franks' "ministerial certification" here.)
27. Cmnd. 408, para. 29.

175



THE DENNING LAW JOURNAL

Like Franks, too, the 1989 Act defines those persons subject to criminal
sanctions by way of "primary" disclosures. These are Crown servants, including
Ministers of the Crown, civil servants, servicemen, policemen and police force
employees, as well as persons holding "prescribed" offices,Z8 government
contractors, and former Crown servants or government contractors. A 'secondary'
disclosure i.e., by a person other than a Crown Servant or government contractor,
or former Crown servant or government contractor, is unlawful if it involves
information disclosed without authority, or entrusted to the discloser or another in
confidence, by a Crown servant or government contractor. A person making such a
disclosure is guilty if he disclosed the information without authority, knowing or
having reasonable cause to believe it was "protected"Z9 information. A special test
of harm applies here, in that the disclosure must be damaging, and the person
making it must have reasonable cause to believe it would be damaging.3o

It is a defence for a person charged under sections 1-4 to show that he did not
know, and had no reasonable cause to believe, that the information disclosed by
him was "protected" or that the disclosure would be "damaging" according to the
definitions given in each section. An exception, however, is made for members and
ex-members of the security services. For such a defendant, the only defence arises
if he did not know and had no reason to believe that the information disclosed
related to security or intelligence. 3 I

The mere receipt of official information ceases to be a criminal offence. On the
other hand, the Act makes it quite clear that a disclosure is made by a Crown
servant with lawful authority "if, and only if, it is made in accordance with [his]
official duty" and, where a government contractor is concerned, if it is made in
accordance with official authorisation, or for the purposes of his functions as a
government contractor.32 Furthermore, unauthorised retention of a "protected"
document is an offence.33

As with the 1911 Act, however, the consent of the Attorney-General remains a
pre-requisite of any prosecution, except for one brought for a section 4 disclosure.
Here, that of the D.P.P. will suffice.34 Sentences for illegal disclosures are, on
indictment, a maximum of two years imprisonment, or a fine, or both, and on
summary conviction, six months, or a fine, or both. Penalties for illegal retention

28.5.12. The Secretary of State may, by order, "prescribe" an office for the purposes of the Act: ss.12;
13.
29. I.e., within the categories given in ss.1-4.
30.5.5.
31. Sections 1(5); 2(3); 3(4); and 4(4). An interesting point of interpretation arises, however. Does the
defence protect only the accused who knows (or suspects) neither that the information was protected,
nor that its disclosure was likely to be damaging. Such an interpretation, in connection with official
secrecy, seems by no means unlikely: see, e.g., R. v. Oakes [1959] 2 QB. 350.
32. Official Secrets Act 1989 S.7.
33.5.8.
34. 5.9(1) and (2).

176



THE OFFICIAL SECRETS ACT 1989

are three months or a fine up to level 5 (on summary conviction).35 Illegal
disclosures are made arrestable offences.36

The right balance?
The question must now be addressed whether the Official Secrets Act 1989

achieves the right balance between the public's right of access to information
concerning public administration,37 and the protection of that official information
which, for reasons of the public interest, should be kept secret. We are not here
concerned with espionage as such. Few would doubt the need for criminal
sanctions there. But the British constitution and system of administration contain
many means whereby confidentiality can be ensured, and of calling to account
those who breach it. Thus, ministerial responsibility, the Civil Service code of
discipline, the law of confidence, and many other measures, both formal and
informal, all have a role to play, and all were considered by Franks in 1972, and by
the Government in 1988. On both occasions, the consensus seems to have been
that criminal sanctions should apply only in respect of certain kinds of disclosures.

Three "yardsticks" are discernible in the Act (as they were in Franks, and even,
to a limited extent, in the old section 2). These are directed at:

1. The type of information disclosed;
2. The position (or office) of the discloser;
3. The degree of harm resulting, or likely to result, from the disclosure.

The "catch-all" nature of section 2, and, to some extent at least, the success (or
otherwise) of the new Act, can be largely assessed by reference to these, given
the declared aim of the Government "to narrow the scope of the present law so
that the limited range of circumstances in which the unauthorised disclosure of
official information needs to be criminal are clearly defined.,,38

Thus, for instance, Frederick Blake, the retired Prison Governor prosecuted for
his "What Bywaters Told Me" article would not be guilty of any offence under the
new Act nor would the Probate Registry clerk who "leaked" information
concerning wills to the newspapers in 1932. In neither case would the information
concerned be "protected" for the purposes of the 1989 Act. More contentiously,
perhaps, it is at least arguable that Clive Ponting (who was, in any case acquitted
under the old section 2) could not be successfully prosecuted by virtue of the
requirement for 'harm', since the information disclosed by him, although
"protected" (it related to defence) concerned events of two years earlier, and had
ceased to be relevant operationally.39

35. S.lO.
36. S.ll.
37. A concept somewhat whimsically captured in the American phrase "Government in The
Sunshine".
38. Cmnd 408, para. 14.
39. The Ponting case is discussed further below, in connection with a "public interest" reference.

177



THE DENNING LAW JOURNAL

Conversely, it might be argued that the scope of "protected" information has
been narrowed too much. "Fiscal" information, and that relating to the currency
(proposed by Franks for protection) is not specificallyincluded, though some such
information, insofar as it is given in confidence to other nations or international
organisations, and relates to international relations, might be "protected" by
section 6. Again, Franks proposed that information given to the Government in
confidence by private individuals should be protected.40 The new Act makes no
such provision. This is a regrettable omission, but not a surprising one. It was
never the intention of the Government to legislate with respect to the protection of
the privacy of the individual. It is notable, however, that Official Information
legislation in many other countries contains just such a provision. The U.S.A.
Freedom of Information Act, for instance, specifically excepts "personal and
medical files, and similar files, the disclosure of which would institute a clearly
unwarranted invasion of privacy", amongst other information, from the general
requirement that information in the possession of Federal Agencies should be
made available to the public.41Similar provisions exist in Australian and Canadian
official information legislation. The absence in the British constitution of any
cogent alternative means of protecting confidential information of a personal
character (unlike, say, cabinet and other governmental documents, in respect of
which disciplinary procedures, conventional rules, and so on, apply) highlights this
omission.

Doubtless, there will be those who feel that, whilst the old section 2 was badly
drafted, and in need of reform, its "catch-all" nature was something of a strength,
since it is not always possible to predict the harm that might result from a
disclosure until it has actually occurred. Limiting the scope of the Act to specified
categories of information and excluding others, therefore, is inherently risky!
Accordingly, for some, a "catch-all" provision, with the safeguard of the
Attorney-General's control over prosecutions, remains the "lesser of two evils".
Much reliance, however, would thereby be placed on the independence of the
Attorney-General. Indeed, Franks commented: "A catch-all provision is saved
from absurdity only by the sparing exercise of the Attorney-General's discretion to
prosecute." Yet ... "the very width of this discretion, and the inevitably selective
way in which it is exercised give rise to considerable unease.,,42

To appreciate this point, we have to go back only as far as 1985, when the I.B.A.
permitted the showing of a programme entitled "MIS's Official Secrets", despite

40. Cmnd. 5104, Vol. 1, paras. 192-206.
41. U.S. Code, Title 5, 5552. (1966). This is exemption 6. The others involve information of the
following kinds: 1. National security; 2. An agency's internal and personal affairs; 3. Specified in a
statute directing non-disclosure; 4. Trade secrets or confidential commercial and financial information;
5. "Inter-agency" memoranda or letters which are not normally available in law by means of discovery;
6. Investigative records compiled by law enforcement agencies; 7. Records of specified agencies,
including The Federal Reserve Board, concerning financial and banking matters. 8. For some reason,
"geological and geophysical information, including maps concerning wells".
42. Cmnd. 5104, Vol. 1, para. 88.

178



THE OFFICIAL SECRETS ACT 1989

advice that the programme involved a clear breach of section 2 of the Official
Secrets Act based as it was in part on information provided by a former MIS
officer, Cathy Massiter. Doubtless not wishing to mount a further unpopular
prosecution in the wake of the Ponting case, the Attorney-General, Sir Michael
Havers, indicated that there would be no prosecution. On receiving this advice, the
I.B.A. lifted the temporary ban it had placed on transmission, and the programme
was shown on 8th March. Needless to say, Ms Massiter would be guilty under
section 1 of the new Act. Furthermore, it seems highly unlikely that such
revelations as those made by Ms Massiter would go unprosecuted. The narrowing
down of the categories of information protected by the Act, and the particular
provisions relating to security and intelligence members or ex-members, would
seem to create a virtual 'mandate' for such prosecutions. Certainly, the resolve to
mount them in future is likely to be far greater than it was in 1985, with the
Attorney-General being able to point to the very narrowness of the new Act as
justification for its more-or-Iess automatic use. A "leaner" Act should prove both
stronger and fitter for its purpose.

So much, then, for the narrowing of the categories of information, disclosure of
which remains criminal under the new Act. What of the circumstances of
disclosures, and in particular, the persons who are liable to prosecution? It is true
that there was room for improvement. Again, the catch-all "flavour" could be
discerned in the phrase "person holding office under His Majesty" in section 2. Its
replacement by the more clearly defined categories of "Crown servants" and
"government contractors,,43 is doubtless an improvement so far as clarity of
draftsmanship is concerned. But will this new clarity, of itself, make any real
difference in terms of actual prosecutions for unauthorised "primary" disclosures?

Paradoxically, this question may be tested by reference to a category of persons
whose members, though clearly "holding office under His Majesty", have never
been prosecuted: Ministers of the Crown. Despite some quite serious leaks of
official information to the press, no Government Minister has ever been charged
under section 2. Franks' explanation for this was that Ministers were largely
"self-authorising"; as the political head of a Government department or Ministry,
a senior Minister, at least, is subject to no higher authority in matters of disclosure,
and therefore has to be regarded as vested with such authority himself. There is, of
course, much in this argument. The assumption which is sometimes made,
however, that Ministers are somehow privileged and therefore immune from
prosecution, does not necessarily follow. Indeed, it has no basis whatever in Law.
Thus, in the event that a Government Minister were to leak information protected
under the 1989 Act, especially if it concerned matters which did not fall within the
scope of his personal responsibility, one might reasonably expect a prosecution to
occur, particularly since Ministers are actually specified in section 12. It is
tempting to prognosticate, however, that, save for the most serious of

43. Official Secrets Act 1989, s.12.

179



THE DENNING LAW JOURNAL

leaks, amounting almost to espionage, Ministers will remain, in practice
immune from prosecutions. Indeed, the lobby system of briefings would seem to
demand it!

The scope of "protected" information, and of the categories of persons likely to
be prosecuted, however, are not all that have been limited by the new Act. The
failure of the Government to incorporate two possible general defences to charges
of unlawful disclosure has been the subject of serious criticism, not least in the
Press. It is to these that we must now turn.

A "public interest" defence
As has been established, the 1911 Act, in section 2, contained at least a form of

"public interest" defence, albeit one which was severely limited in scope. Nowhere
can this be seen to better advantage than in the case of Clive Ponting, the civil
servant who in 1985 was prosecuted under section 2 for leaking ministerial
memoranda to an M.P., Mr Tam Dalyell. The memoranda concerned the sinking
of the Argentine cruiser General Belgrano, by a British submarine, on 2nd May
1982, at the height of the Falklands conflict. Ponting did not argue that the
decision to sink the vessel had been wrong. It was his contention, that the leaked
documents demonstrated the Government's intention to suppress the whole truth
about the episode, and in so doing, mislead a Commons Select Committee which
was investigating the matter.44 He leaked them to Dalyell because, though he was
not a member of the Committee, he was a long-standing M.P. who had himself,
Ponting felt, been deceived by Government Ministers when answering questions
in the Commons.

At Ponting's trial, McCowan] directed the jury in the following terms. First, he
said that section 2 had created a crime of basic intent only. It was not necessary for
the prosecution to establish any particular intention on the accused's part, over and
above the intention to communicate information to someone without authority.
Certainly, it was not necessary to prove that his intention was to harm the interests
of the State. Secondly, he told the jury that whilst section 2 had provided for a
form of "public interest" defence, it should be interpreted narrowly, so that
communication "to a person to whom it is in the interest of the State his duty to
communicate it" meant a communication made, by a Civil Servant, in his official
duty only. Finally, the judge suggested that "the interests of the state" were, for
practical purposes, synonymous with Government policy. Despite this direction,
the jury returned a verdict of "not guilty", and Ponting was duly acquitted.

The Government explained its reasons for rejecting a "public interest" defence
in the White Paper. Whilst acknowledging that "some people who make
unauthorised disclosures do so for what they themselves see as altruistic reasons",
the Government felt that the "general model" in which motive is irrelevant to

44. Such conduct, if substantiated, would amount to a serious breach of convention: see, e.g., Erskine
May's Parliamentary Practice, which makes it clear that any attempt or conspiracy to mislead a
Parliamentary Committee amounts to a contempt of the House.

180



THE OFFICIAL SECRETS ACT 1989

criminal liability should be adhered to. There had been some feeling in
Parliament, secondly, that the inclusion of such a defence would elevate
"whistleblowers" and place them in some "special category", above thieves,
murderers and the like. Thirdly, such a defence, it was argued, would detract from
the clarity sought in the new Act. Accordingly, the Government felt that "any
arguments as to the effects of disclosure on the public interest should take place
within the context of the (proposed) damage tests where applicable."45

Unfortunately, each of these arguments can be challenged. First, the motives
argument fails to take into consideration the possibility of an objective test of
public interest. It fails, seemingly, to distinguish between cases like that ofPonting,
on the one hand, and Sarah Tisdall on the other. Ms Tisdall, unlike Ponting, had
been convicted and sentenced to prison, the first person to be so sentenced for
fifteen years, when she leaked details concerning the disposition of Cruise missiles
to the newspapers. The case is quite distinct from that of Clive Ponting, however,
in that (a) the information she leaked did not suggest impropriety by Government,
or any attempt to mislead Parliament, unlike Ponting's disclosures; and (b) she
leaked it not to an M.P., but to a newspaper (The Guardian). Ms Tisdall's,
arguably, was just the sort of case contemplated in the White Paper; Ponting's was
not.

The provision of some statutory protection from reprisals for "whistleblowers"
is by no means unknown. In the U.S.A., for instance, a civil servant who leaks
information concerning fraud, waste, abuse and other types of wrongdoing in
Government departments is given such protection.46

That such an objective test of public interest is incapable of precise definition is
clearly not true. The concept of the public interest is a fairly commonplace one
elsewhere in English law; in connection with discovery of documents, for instance,
where the courts have grown well used to balancing degrees of harm to the public
interest threatened by disclosures.47 The suggestion that considerations of the
public interest can really best be dealt with in the context of the "damage test",
too, must be closely examined. First, there is the suggestion that a "damaging"
disclosure can never be in the public interest. The White Paper made this point
clearly: "It cannot be acceptable that a person can lawfully disclose information
which he knows may, for example, lead to loss oflife simply because he conceives
that he has a general reason of a public character for doing so." True: but not all
disclosures need to be quite so damaging. Also, as already suggested, an objective
test of "public interest" is perfectly feasible, and it is not necessarily a question of
what the accused "conceives" to be his reasons for disclosing damaging
information. Even the seeming paradox, that a disclosure could be "damaging"
and yet in the public interest, is capable of being resolved. The Obscene

45. Cmnd. 408, para. 61.
46. The Civil Service Amendment Act ("whistleblower" Protection Act 1978, amended 1988, U.S.
Code 200, S.20.).
47. See, e.g., Conway v. Rimmer [19681 A.C. 910; D. v. N.S.P.c.c. [1977J A.C. 171, amongst others.

181



THE DENNING LAW JOURNAL

Publications Act 1959, for instance, provides that a publication, though "obscene"
(and therefore clearly "damaging", in that it "tends to corrupt and deprave"), may
nevertheless be in the public good by reason of its being "in the interests of
science, literature, art or learning or other subject of general concern.,,48 It might
be noted, too, that the test of "public interest" in obscene publications is left to the
jury to apply. Another worrying aspect of the Government's claim arises from the
fact that unless the phrase "public interest" actually appears in the Act, it is
extremely unlikely that any jury will, in practice, be directed to the effect that it
should form any part of their deliberations. It is hard to see, then, how argument as
to the "public interest" would, in fact, take place at all.

The suggestion that a "Public Interest" defence would elevate those charged
with Official Secrets offences simply falls down in the face of what we have already
shown. Indeed, there seems to be every reason to treat offences involving
"publications" of some sort - including disclosures of protected information -
quite differently from those, like murder, rape and theft, which do not. This is, not
uncommonly, already the case.49 Thus, it is suggested that a "Public Interest"
defence could have been successfully included in the Act if, indeed, it was the wish
of the Government that considerations of the "Public Interest" should play some
part in proceedings.

A defence based on "prior publication".
Save in limited form in section 6(5), concerning information communicated to

another state or international organisation, which has been published with the
authority of that state or organisation, no defence based on the fact that information
has been previously published arises under the 1989 Act. Again, the Government
considered such a defence, and gave its reasons for rejecting it, in the White
Paper. These were basically twofold. First, confirmation of some story by a senior
government official might prove considerably more damaging than a prior,
unconfirmed publication. Secondly, information which has been previously
published piecemeal in a number of sources, might prove damaging only when
collated, and published as a single disclosure, e.g., a list of names and addresses
which might prove useful to terrorists. 50

These arguments have undeniable force. Also, as the Government pointed out,
in a case where a publication is unlawful only if it causes harm, the offence would
not be made out if no further harm is likely to result from a second disclosure.

48. Obscene Publications Act 1959, 5.5, used effectively in the "Lady Chatterley's Lover" trial: R. v.
Penguin Books Ltd. [1961] Grim. L.R. 176. Interestingly, and importantly, the judge in that case (Byrne)
ruled that the jury was only required to consider "public good" if it found that the book WOJ obscene,
and not otherwise.
49. In fact, "public interest" is a relevant consideration, or defence, in relation to a number of offences
involving publications of one sort or another: these include, as well as obscene publications, libel and
contempt of court (Contempt of Court Act 1981, s.5.) see, for a discussion on this, Yvonne Cripps: The
Legal Implications of Disclosure in the Public Interest (1986).
50. Cmnd. 408, paras. 62-64.

182



THE OFFICIAL SECRETS ACT 1989

This would not apply, however, to a section 1 offence, which was of course the
subject of the recent Spycatcher case, since the normal requirements for harm
would not here apply. But the implication that a "prior publication" defence would
be wholly novel must be scrutinised. In 1987, the Court of Appeal held that, once
information had passed lawfully into the public domain, a further publication of
such information was not an offence under section 2.51 Insofar as a prior
publication has not satisfied the requirement of harm, it would presumably be a
"lawful" one under the Act. In 1988, the House of Lords took up the issue, at least
indirectly. In A. G. v. Guardian Newspapers and Others,52 their Lordships held that
an interlocutory injunction, restraining the appellants from publishing extracts of
the book Spycatcher by the former MIS officer, Peter Wright, should be lifted, and
not replaced by a permanent injunction, on the grounds that all possible damage to
the Crown's interests which could occur had, in fact, occurred as a consequence of
"prior publication" of the material abroad.

It should be noted that the Government's argument - that most second
publications would not be unlawful, since they would be unlikely to cause damage
- would not apply to even "secondary" disclosures such as those involved in the
Guardian case, under the 1989 Act. In connection with security and intelligence
matters, even "secondary" disclosures do not have to cause damage as such. It is
enough if they involve information of a "class" provided for in section 1(4)(b).
Accordingly, if the Guardian case was to be heard when the new Act comes into
force, a court would, presumably, be bound to make permanent an injunction
restraining publication, despite the fact that publication all over the world has
meant that no further harm to the public (or the Crown's) interests is possible!

So, whilst accepting the Government's argument that an absolute defence of
prior publication might have proved impossible - even undesirable - one may
nevertheless regret that "prior publication" will not be at least a relevant
consideration when assessing the damage caused by any disclosure, even one
involving security or intelligence matters, particularly when such a disclosure is a
"secondary" one. It should have proved perfectly possible to retain strict liability, if
that were felt necessary on the part of security officers and ex-security officers,
whilst maintaining the essence of the Guardian decision, to the effect that, if all
conceivable damage has been done, and can be demonstrably shown to have been
done, it becomes somewhat pointless to 'muzzle' the Press by means of the threat
of criminal sanctions.

Conclusions
Whilst the Government's justification for omitting the two defences discussed

above is not wholly convincing, there is much that is desirable about the Act, and
even those omissions must be considered in context.

51. R. v. Galvin [1987] I QB. 862.
52. [1988] 3 All E.R. 545.

183



THE DENNING LAW JOURNAL

In general terms, the Government was right to seek to align Official Secrets'
offences, so far as possible, with criminal offences generally. Thus, the inclusion of
defences amounting to the absence of mens rea, and the requirement, in
connection with most offences, of harm or likely harm, are commendable. With
regard to the latter, too, the Government was right to reject Franks' system of
conclusive Ministerial certification, correctly observing that such a system would,
or would be seen to, place too much power in the hands of Ministers. Further, by
rejecting such a scheme, the Government, once again, has aligned official secrets
with other areas of recent development, such as the rules of public interest
immunity in relation to discovery. That the mere unauthorised receipt of official
information will no longer be criminal was a long overdue reform. The Act may
well bring about some indirect benefits, too, in relation to public access to
information. The provisions in section 4, protecting information obtained by
means of, or concerned with, interception of communications, could well be such
that two benefits result. First, for a prosecution to be brought, it will presumably
have to be established that such an interception had occurred, thus increasing
public knowledge about such things. Secondly, such an interception.will have to be
a lawful one for a prosecution to succeed. Accordingly, unlawful interceptions may
be brought to light. Under the present law, disclosure of any such interceptions,
whether lawful or unlawful, is potentially criminal (subject to the "public interest"
defence.)

In terms of the Government's objectives, then, the Act may be accounted a
modest success. Its most notable failure, perhaps, is the apparent lack of regard for
the confidentiality of private, personal information, which brings it out ofline with
Official Information legislation in most Western democracies, and with Franks'
proposals. The arguments for rejecting Public Inte'rest and Prior Publication
defences are not wholly convincing, but their absence is likely to be offset
somewhat by the fact that the requirements for harm (in most cases) will,
hopefully, be argued in the normal way, by means of evidence, leading to judicial
determination, and not settled by Ministerial determination.

On balance, then, except for those who regard a catch-all provision as the safer
course, the new Act is, at the very least, likely to bring about a shift in the balance
of interests, and to that extent, anyway, is an improvement on the old.

184