Should we keep our secrets?

WHAT RECORDS should the Security Service, or MI5 as it is often called, keep for eventual release into the public domain? This question, important on constitutional grounds as well as historical, is being looked at afresh by a committee of which I am a member.

Actually we have a rather grand title: the Lord Chancellor's Advisory Council on Public Records. Our normal task is to police the working of the "30-year Rule" under which government records are made available for inspection at the Public Record Office when they are 30 years old.

Now we have been asked by the Home Secretary, Jack Straw, to review the criteria which the Service currently applies in deciding whether files which would otherwise be destroyed should be kept on grounds of historical significance. We would welcome submissions from those with an interest in the subject; it is important to know what people think.

The point about MI5 is that it works primarily in the United Kingdom. Its current work is to investigate terrorism (especially in Northern Ireland), to protect the UK against espionage by foreign intelligence agencies, to help counter the proliferation of weapons of mass destruction and to support the police in the fight against serious crime. However, the part of its remit that raises the most acute problems for the historic record has been its work against so-called subversive organisations of the extreme left or extreme right which, often working behind a respectable facade, were thought to be trying to undermine parliamentary democracy.

This accounted for a major part of the Security Service's activities between 1948, the start of the Cold War, and 1990 when the Soviet Union collapsed. That is why, although anti-subversion work is no longer undertaken, the service holds 290,000 files relating to individuals, among them it is supposed, folders for "Straw, Jack: president of the National Union of Students" and "Mandelson, Peter: trades union official".

The first problem which confronts us is the service's instinctive reluctance to put anything at all into the public domain. Thus the only files which are available at the Public Record Office cover the period from its foundation in 1909 to 1919.

They are, apparently, very bland. However by the millennium it is hoped that material covering the inter-war years and the Second World War will have been released. In short, MI5 will have improved from keeping its files closed for 80 years to a mere 55 years.

Moreover, the service periodically destroys files on a massive scale. This happened following both world wars and again in the early 1990s. The result is that it has preserved about 60 per cent of its records, whereas the Foreign Office, for instance, keeps about 80 per cent of its files and the Cabinet Office and 10 Downing Street retain almost everything. The saving grace is that MI5 has at least preserved its registers, so it will be possible one day to know what was once available if now lost.

The Advisory Council will have to evaluate MI5's arguments for non-disclosure. Like all intelligence services, it feels an absolute duty to protect the identity of its agents or informers. And it sees this duty as extending for long after the agent has ceased working for the Service and possibly after his or her death if family members remain vulnerable. For operational reasons, it keeps the identities of its staff secret, except that in recent years the name of the director general has been published.

It also argues that premature release can put innocent people in a dubious light. And it wants to keep its techniques under wraps because suspects are generally unaware, even if they assume that they are being watched, of how unwittingly they let slip incriminating information. Espionage is like archaeology, putting together fragments to form an impression of the whole.

The question for us is whether these perfectly reasonable considerations are being applied too cautiously in relation to the purpose of disclosure, which is to allow historians and members of the public to understand in a reasonably timely way how the organisation was managed, financed, structured and operated; how it influenced and was influenced by the social, economic and physical culture of which it was part; how decisions were taken and what those decisions were and how they were applied.

The second big question is what retention and disclosure policy is appropriate for the service's huge holding of files on British people, on us. There is substantial and justifiable interest in the interaction between a secret service and the citizen; equally, however, it is scarcely conceivable that personal files could be released during the lifetime of the subject.

Of the 290,000 personal files extant, all but 20,000 of them are closed in the sense that while Security Service officers may consult them, no further enquiries are undertaken on their subjects. Many of these will contain limited information - that "X" is or was a member of, say, the Communist Party together with the proof. No further action will have been taken and there will be thus no private or intrusive information on the file.

There are two ways to go. There is the example of the United States where you may apply to a "freedom of information" officer at the FBI or CIA and ask whether there is a file on you. Subject to safeguards for current investigations, you will be told "yes" or "no" and if the answer is "yes" you will be informed how many pages it comprises, how much is classified and how much you may examine if you so desire. And there is a right of appeal to see the classified material. This approach has two virtues - it satisfies a democratic right to know whether there is a file and it allows for the correction of mistakes and misapprehensions.

The alternative way is to ask what should be known of MI5's investigation of UK citizens from the point of view of the historic record. On whom did the service have files? What sort of people were targets? How did it operate against those targets? What did the files contain? Was the information accurate? What did the service do with what it learnt?

Then, having established the questions, the next consideration would be whether keeping a sample of files would be sufficient to enable authoritative answers to be given. If the answer were in the affirmative, then a policy, perhaps less haphazard than the present arrangements, albeit opposite to the American system, could be proposed.

So far as personal files are concerned, they would only be kept for operational reasons, or for the historic record, or for meeting complaints made to the Security Service Tribunal. If none of these considerations pertained, then the file would be destroyed.

In other words, the argument here is that destruction may be a safeguard for individual liberties. That is perhaps the most difficult question which the Advisory Council, sitting under the chairmanship of the Master of the Rolls, Lord Woolf, will have to decide.

Submissions should be made as soon as possible to Tim Padfield, Secretary of the Advisory Council on Public Records, Public Record Office, Kew, Richmond, Surrey TW9 4DU