[HOUSE OF LORDS]

AND

Crown - Privilege - Objection to produce documents - Oil company action challenging fairness of commercial agreement with Bank of England - Bank acting under government guidance during company crisis - Crown intervention on summons for discovery - Ministerial certificate asserting privilege in public interest - Documents concerned with formulation of national policy and confidential information from business concerns - Whether documents privileged

On January 23, 1975, an agreement was made between the plaintiffs, an oil company, and the Bank of England acting in close contact with and under the direction of the government, with the object of rescuing the company from grave financial difficulties arising out of the international oil crisis on terms consistent with the government's national economic policy. A term of the agreement involved the sale and transfer to the bank of nearly 78 million ordinary stock units in British Petroleum held by the company at £2.30 per unit, the price required by the government.

In October 1976 the oil company began an action against

the bank claiming a declaration that the sale was unconscionable and inequitable and an order that the bank should transfer the stock units back to the company at the price paid in 1975, about £179 million. In March 1977 the company sought an order for discovery of all documents held by the bank relevant to the issues pleaded. The bank disclosed quantities of material relevant to the issues, but on government instructions resisted production of 62 documents in its possession and control. On the hearing of the summons the Crown intervened by the Attorney-General, the bank taking no part in the proceedings. Objection to production was taken by the Chief Secretary to the Treasury in a certificate stating that he had personally read and considered each of the 62 documents listed in a schedule and had formed the opinion that their production would be injurious to the public interest. He particularised the documents in three categories: categories A and B consisted of classes of documents relating to the formulation of government policy in the light of the international consequences of the financial collapse of the oil company and its effect on governmental oil policy, as discussed either at ministerial level or at a lower level with the bank but related to eventual formulation of ministerial policy; category C documents concerned commercial or financial information communicated in confidence to the government or the bank by major business companies and other businessmen; and in respect of those the certificate claimed that the preservation of confidence was in the public interest since if it became known that what was imparted in confidence might be revealed publicly, such information, necessary to policy decisions, would cease to be so readily forthcoming.

By innocent mistake six of the 62 listed documents were sent by the bank's solicitors to the company's solicitors in legible form and were read by them and their counsel. On the hearing of the summons copies of them were handed to Foster J. in a sealed envelope which he did not open. He upheld the Crown's claim for privilege for all the documents without looking at any of them, as he had been invited to do; but he gave the company leave to appeal. At the end of argument the Court of Appeal read the documents in the sealed envelopes before preparing their judgments. The Court of Appeal (by a majority) dismissed the appeal.

On the plaintiffs' appeal to the House of Lords the list of 62 documents discovery of which the company originally sought was reduced to 10 (two in category A and eight in category B): -

Held, (1) (Lord Wilberforce dissenting) that the documents should be produced for inspection by the House of Lords, for the present case was one where without inspection of the documents it was not possible to decide whether the balance of public interest lay for or against disclosure (post, pp. 1120E - 1121B, 1128H - 1129B, 1130C - E, 1134H - 1135A, 1136A - C,E-F, 1142C-F, 1145A-C).

Robinson v. State of South Australia (No. 2) [1931] A.C. 704, P.C.; Glasgow Corporation v. Central Land Board, 1956 S.C. (H.L.) 1, H.L.(Sc.) and Conway v. Rimmer [1968] A.C. 910, H.L.(E.) considered.

(2) That the documents having been inspected (per Lord Salmon and Lord Edmund-Davies; Lord Keith of Kinkel dubitante) none of them contained matter of such evidential value as to make an order for their disclosure, in all the

circumstances, necessary for disposing fairly of the case (post, pp. 1122F, 1130E-F, 1136G); (per Lord Scarman) that they were relevant but their significance was not such as to override the public service objections to their production; and that, accordingly, the appeal must be dismissed (post, p. 1147C).

Per Lord Wilberforce. The case in favour of public interest immunity was made out by the minister's certificate and there was no ground for inspecting the documents (post, p. 1117E-G).

Per Lord Keith of Kinkel and Lord Scarman. Where the court inspects a document for which Crown privilege is claimed the Crown should have a right to appeal before the document is produced (post, pp. 1136E, 1146H - 1147B).

Dictum of Lord Reid in Conway v. Rimmer [1968] A.C. 910, 953, H.L.(E.) applied.

Decision of the Court of Appeal [1979] 1 W.L.R. 473; [1979] 2 All E.R. 461 affirmed on different grounds.

The following cases are referred to in their Lordships' opinions:

Barty-King v. Ministry of Defence (unreported), October 10, 1978, May J.

Compagnie Financière et Commerciale du Pacifique v. Peruvian Guano Co. (1882) 11 Q.B.D. 55, C.A.

Conway v. Rimmer [1968] A.C. 910; [1968] 2 W.L.R. 998; [1968] 1 All E.R. 874, H.L.(E.).

Crompton (Alfred) Amusement Machines Ltd. v. Customs and Excise Commissioners (No. 2) [1972] 2 Q.B. 102; [1972] 2 W.L.R. 835; [1972] 2 All E.R. 353, C.A.; [1974] A.C. 405; [1973] 3 W.L.R. 268; [1973] 2 All E.R. 1169, H.L.(E.).

D. v. National Society for the Prevention of Cruelty to Children [1978] A.C. 171; [1977] 2 W.L.R. 210; [1977] 1 All E.R. 589, H.L.(E.).

Duncan v. Cammell, Laird & Co. Ltd. [1942] A.C. 624; [1942] 1 All E.R. 587, H.L.(E.).

Glasgow Corporation v. Central Land Board, 1956 S.C. (H.L.) 1, H.L.(Sc.).

Grosvenor Hotel, London (No. 2), In re [1965] Ch. 1210; [1964] 3 W.L.R. 992; [1964] 3 All E.R. 354, C.A.

Nixon v. United States (1974) 418 U.S. 683; 41 L.Ed. 2d 1039.

North Ocean Shipping Co. Ltd. v. Hyundai Construction Co. Ltd. [1979] Q.B. 705; [1979] 3 W.L.R. 419; [1978] 3 All E.R. 1170.

Occidental Worldwide Investment Corporation v. Skibs A/S Avanti (The Siboen and The Sibotre) [1976] 1 Lloyd's Rep. 293.

Pao On v. Lau Yiu Long [1980] A.C. 614; [1979] 3 W.L.R. 435; [1979] 3 All E.R. 65, P.C.

Reg. v. Lewes Justices, Ex parte Secretary of State for the Home Department [1973] A.C. 388; [1972] 3 W.L.R. 279; [1972] 2 All E.R. 1057, H.L.(E.).

Robinson v. State of South Australia (No. 2) [1931] A.C. 704, P.C.

Sankey v. Whitlam (1978) 53 A.L.J.R. 11.

Science Research Council v. Nassé [1980] A.C. 265; [1979] 3 W.L.R. 762; [1979] 3 All E.R. 673, H.L.(E.).

Smith v. East India Co. (1841) 1 Ph. 50.

Tito v. Waddell (unreported), March 3, 1975, Walton J.

United States v. Bethlehem Steel Corporation (1942) 315 U.S. 289.

United States v. Reynolds (1953) 345 U.S. 1.

Westminster Airways Ltd. v. Kuwait Oil Co. Ltd. [1951] 1 K.B. 134; [1950] 2 All E.R. 596, C.A.

The following additional cases were cited in argument:

Admiralty v. Aberdeen Steam Trawling and Fishing Co. Ltd., 1909 S.C. 335.

Asiatic Petroleum Co. Ltd. v. Anglo-Persian Oil Co. Ltd. [1916] 1 K.B. 822, C.A.

Attorney-General v. Jonathan Cape Ltd. [1976] Q.B. 752; [1975] 3 W.L.R. 606; [1975] 3 All E.R. 484.

Beatson v. Skene (1860) 5 H. & N. 838.

Birkett v. James [1978] A.C. 297; [1977] 3 W.L.R. 38; [1977] 2 All E.R. 801, C.A. and H.L.(E.).

Carmichael v. Scottish Co-operative Wholesale Society Ltd., 1934 S.L.T. 158.

Hennessy v. Wright (1888) 21 Q.B.D. 509.

Hoffmann-La Roche v. Secretary of State for Trade and Industry, The Times, April 19, 1975.

Rogers v. Orr, 1939 S.C. 492.

Spigelman v. Hocken (1933) 150 L.T. 256.

Wadeer v. East India Co. (1856) 8 De G.M. & G. 182.

APPEAL from the Court of Appeal.

This was an appeal, with leave of the Court of Appeal, from an order dated January 19, 1979, of the Court of Appeal (Bridge and Templeman L.JJ., Lord Denning M.R. dissenting) affirming an order dated July 28, 1978, of Foster J. whereby he refused to order production of the 62 documents listed in Part III of Schedule 1 to the list of documents served by the respondents, the Governor and Company of the Bank of England ("the bank"), in the action between the appellants, the Burmah Oil Co. Ltd. ("Burmah"), and the bank.

The bank had objected to produce the 62 listed documents on the ground that those documents belonged to classes of documents the production of which would be injurious to the public interest. In support of that ground the bank had put forward a certificate dated October 18, 1977, made by the Chief Secretary to Her Majesty's Treasury, in which he stated that he had read and carefully considered all the 62 listed documents and had formed the opinion that their production would be injurious to the public interest.

By a summons dated March 15, 1978, in the action between Burmah and the bank Burmah sought an order for the production of the 62 listed documents. On the hearing of that summons before Foster J. the respondent, Her Majesty's Attorney-General, intervened to argue in support of the claim that production of the scheduled documents would be injurious to the public interest. The Attorney-General also appeared and argued in support of that claim on the hearing of Burmah's appeal before the Court of Appeal. The bank appeared but took no part in the arguments before Foster J. and the Court of Appeal. The Attorney-General had been joined as a respondent to the appeal to the House of Lords in order that he might, again, have the opportunity to argue in support of the public interest claim. The Attorney-General was not a party to the action between Burmah and the bank.

The facts are stated in their Lordships' opinions.

Leonard Hoffmann Q.C. and John Chadwick for Burmah. This appeal arises as the result of interlocutory proceedings brought in an action by Burmah against the bank for a declaration that an agreement dated January 23, 1975, made between them was unconscionable and inequitable and ought to be set aside. Burmah for the purposes of the action seek production of 62 documents listed in the list of documents served by the bank. The bank on the instructions of the Crown have objected to produce these on the ground that they belong to classes of documents the production of which would be injurious to the public interest.

Conway v. Rimmer [1968] A.C. 910 made a great change in the law relating to the production of documents where the public interest was involved and its consequences have not been fully worked out. The House laid down important principles and re-affirmed the principles adumbrated in Duncan v. Cammell, Laird & Co. Ltd. [1942] A.C. 624. The present claim is entirely a class claim. It is not contended by the Crown that there is anything in these documents which if disclosed would be detrimental to the public interest, only that they belong to a class of documents the production of which would be injurious to the public interest and that taken as a whole therefore they should not be produced As to the observations of Lord Denning M.R. below [1979] 1 W.L.R. 473, 486A-C, this is attractive but it is difficult to support his reasoning since, inter alia, it is very difficult to construct a credible class without reference to the contents of the documents, for example, communications from Her Majesty's ambassadors abroad relating to diplomatic communications. That is a viable class but it is a class referable to the contents of the documents.

These proceedings concern a class claim and the basic submissions relate to the principles governing class claims which are to be discovered in three decisions of this House: Conway v. Rimmer [1968] A.C. 910; Reg. v. Lewes Justices, Ex parte Secretary of State the Home Department [1973] A.C. 388 and Alfred Crompton Amusement Machines Ltd. v. Customs and Excise Commissioners (No. 2) [1974] A.C. 405.

Reliance is placed on the following propositions: (1) An authority making a class claim must satisfy the court that there is some public interest which would be adversely affected to a significant extent if there was disclosure of any documents falling within the authority's definition of the class. (a) In deciding the above question, the court will pay great respect to a minister's certificate, particularly if the public interest is one of which the minister has peculiar knowledge, but the certificate is not conclusive. (b) The question is answered by identifying the grounds of public policy alleged to require non-disclosure and deciding what degree of protection is really necessary to give effect to those grounds. (c) The requirements of public policy in this field are not immutable and the classes of protected documents are therefore not static or fixed by precedent. (2) The court will prima facie accept the minister's certificate that specified documents fall within the defined class. (3) If the court answers (1) in favour of the authority, it will then weigh the public interest in preventing disclosure against the public interest in the proper administration of justice.

The following are the passages relied on in the three House of Lords cases mentioned above: Conway v. Rimmer [1968] A.C. 910, 940C-F, 950B et seq., per Lord Reid; 954G - 958G, 971C - 972G, per Lord Morris of Borth-y-Gest; 973, 979, per Lord Hodson; 984-988, per Lord Pearce; 991-995E, per Lord Upjohn. Reg. v. Lewes Justices, Ex parte Secretary of State for the Home Department [1973] A.C. 388, 400 et seq., perLord Reid; 405, per Lord Morris of Borth-y-Gest; 407, per Lord Simon of Glaisdale; 411 et seq., per Lord Salmon. Alfred Crompton Amusement Machines Ltd. v. Customs and Excise Commissioners (No. 2) [1974] A.C. 405, 433E-435, per Lord Cross of Chelsea. The above cases were reviewed by the High Court of Australia in Sankey v. Whitlam (1978) 53 A.L.J.R. 11. The judgment of Mason J. contains one of the clearest analyses of what constitutes public interest in matters of this kind. Reliance is placed on the judgments of Gibbs and Stephens JJ. that the documents in question in the present case are of merely historical interest being four years old.

As to proposition (1), it does not suffice to give the Crown protection if the minister frames his class in such wide terms that the court as a consequence forms the opinion that there are certain sub-classes within that class which need protection because in that case the Crown has not discharged the burden placed upon it. In that event the court can either (i) ask the minister for clarification of the class, or (ii) inspect the documents itself. Not only should the court be satisfied (a) that all documents in the class need protection but also (b) that policy considerations apply to all documents in the class with equal force because otherwise the court is unable to perform the balancing exercise which Conway v. Rimmer [1968] A.C. 910 enjoins. In the present list of documents category A does not disclose whether these documents are of current or only of historical concern.

The time element. There are four headings under this branch of the argument: (i) what is the effect of the passage of time on the policy of protection? (ii) how do the authorities stand on the relevance on the passage of time in the present connection? (iii) how the passage of time affects the balancing process; (iv) reference to those facts which are public knowledge today and which bear on the question whether the passage of time has changed the need for legal protection.

(i) Categories A and B. The Crown relies on the statement of Lord Reid in Conway v. Rimmer [1968] A.C. 910, 952E. The purport of Lord Reid's observations in that passage are that it was not feasible to have the inner working of government exposed whilst the policies in question were still in operation. He was concerned with the effect which ill-informed criticism would have whilst the government was actually performing its task. Lord Reid did not mean to protect governments against political criticism as such but only against ill-informed political criticism which would prejudice actual carrying on of government - the actual policies and business of government. It is conceded that this protection might have to continue for quite some time.

Under this heading two strands of the argument put forward on behalf of the Crown are often fused: (a) the formulation of policy cannot be carried out in public because disclosure might actually hinder the actual

carrying out of the policy, for example, the policy of expenditure cuts. This strand always tends to be linked with (b) the old candour argument, that is, that ministers and civil servants would be less free in their discussions if they knew that subsequently their discussions might become public It is not the appellants' contention that Duncan v. Cammell, Laird & Co. Ltd. [1942] A.C. 624 killed off the candour argument. There may be occasions when that argument is justified. The report of the committee of Privy Counsellors on the disclosure of ministerial memoirs (1976 Cmnd. 6386: Report of the Committee of the Privy Counsellors on Ministerial Memoirs), the Radcliffe Report, is pertinent on this question. Paragraph 30 of the report deals with general principles and paragraphs 49-51 show that the two above policy grounds are intertwined. But there is a great difference between these two grounds for where disclosure would prejudice the carrying out of government policy that is a matter of which the minister has peculiar knowledge and the court would be very reluctant to interfere with his opinion. But it would be limited to that part of government business which is regarded as current. The candour argument is the more weighty of the two but it is one which applies much more selectively and is one which the courts are not incompetent to decide for themselves. The candour argument relates to what persons might be inhibited from saying in the future.

(ii) It is not disputed that in Conway v. Rimmer [1968] A.C. 910, there are numerous dicta that Cabinet minutes are instanced as documents for non-disclosure, and apart from Lord Reid, the question of a time limit was not adumbrated and whilst Lord Reid did include a time qualification he did not elaborate on when such documents were likely to be of no more than historical interest. The test should be that documents of that nature will not be ordered to be disclosed where they are of current interest and concern. On this aspect of the appeal Conway v. Rimmer is neutral. The judgment, however, of Lord Widgery C.J. in Attorney-General v. Jonathan Cape Ltd. [1976] Q.B. 752 is highly relevant and, indeed, central to the present issue. It is a case that certainly does not assist the Crown.

On the question of prejudice to the carrying out of current government business, a leak in a minister's memoirs is likely to be much more damaging, for a publication of that nature is meant to be widely disseminated, but in relation to the production of documents a court can introduce safeguards - it can even sit in camera if that is thought necessary. As to the candour argument, it may be asked: what is likely to be more prohibitive, disclosure by a minister writing his memoirs of the period or, what is likely to be a limited production of documents for the purpose of court proceedings? The effect on the morale of civil servants of the disclosure in memoirs of their advice is a more intractable problem: see paragraph 53 of the Radcliffe Report.

It is emphasised that a disclosure for the purposes of court proceedings is limited to those proceedings. It is a very limited disclosure. The only real difference between Attorney-General v. Jonathan Cape Ltd. [1976] Q.B 752 and the present is that in that case the countervailing factor was freedom of speech whilst here it is the administration of

justice. It would be strange if ministerial disclosures should be put on a higher plane than the administration of justice. Reliance is placed on the Jonathan Cape case and on the Australian case of Sankey v. Whitlam, 53 A.L.J.R. 11, where the documents ordered to be disclosed were four years old.

(iii) In the present case, the effect of the passage of time has been to remove one strand, namely, current government business, from consideration and it leaves only the candour argument and it is only that issue that has to be placed in the balance. It has to be asked; how likely is it that a minister or civil servant would foresee litigation or envisage the possibility of his statement being revealed? It depends on the circumstances. There are cases such as those dealing with quasi-judicial decisions like decisions made by the Gaming Board or the Monopolies Commission where an interested person if he knew of the decision and was dissatisfied would commence litigation. But in a case like the present a minister or civil servant would only foresee litigation if he thought that his actions were open to question. There is nothing here comparable to the example given by Gibbs J. in Sankey v. Whitlam of civil servants giving their opinions as to the suitability of colleagues for promotion.

The factor which underlies Sankey v. Whitlam, 53 A.L.J.R. 11, and Nixon v. United States (1974) 418 U.S. 683 is that where there are allegations against high officers of state or institutions there is public interest that they should be fully investigated and that privilege should not be invoked to prevent the institution of criminal proceedings. In the present case it is true the analogy is much weaker but nevertheless it is alleged that the Bank of England abused its power to extract an unconscionable bargain at the behest of the central government. These are all factors which must be weighed in the balance.

(iv) It is self-evident that these are not current matters nor are they the problems of sterling for sterling is at the present time a strong currency. In so far as North Sea Oil is concerned, the appellants had a 20 per cent. interest in two North Sea Oil fields. In 1976 they disposed of the whole of their interest in one of them and 12 per cent. of their interest in the other.

As to the class C documents, on the time point the same considerations apply as those above. Burmah do not know whether confidentiality should still persist. It is sufficient for businessmen to be assured that matters will only be disclosed where it is necessary for the administration of justice and that such time has elapsed in relation to the matters in question that disclosure would not seriously embarrass the parties concerned.

As to the manner in which the Court of Appeal dealt with the time factor, the Crown relied solely on the judgment of Templeman L.J. [1979] 1 W.L.R. 473, 494-495, who relied on the fact that ministers and civil servants in question are still engaged in public affairs. But this can only go to the candour argument.

Protection for documents no longer of current concern can only be justified on the candour argument. But on the facts of this case the

candour argument is of little weight for the following reasons: (i) litigation is not likely to have been contemplated; (ii) sufficient protection can be given by the safeguards attaching to discovery and the court sitting in camera; (iii) protection for anything really embarrassing could have been covered by a contents claim; (iv) the litigation involved allegations of a serious kind against the government and an important public authority.

Burmah's alternative contention is: On the assumption that still in 1979 a good case can be made out for not disclosing documents relating to high policy mentioned in the minister's certificate, perusal of the course that these proceedings took shows that they must concern matters not of a high concern, for example the fixing of the price of shares. The majority of the Court of Appeal held that the distinction sought to be drawn by the appellants was fallacious. Lord Denning M.R., on the other hand, considered that everything about this transaction was of a commercial nature [1979] 1 W.L.R. 473, 486E et seq. For the purposes of Burmah's argument it is not necessary to go as far as does Lord Denning M.R. in asserting that all the documents relate purely to commercial considerations. It is conceded that there were discussions on the question whether the transaction proposed would affect sterling and it is conceded also that parts of the documents may have to be covered up if discovery is ordered.

At the trial Burmah can put the case on one of two grounds, either that a mortgagee must not take undue advantage of the mortgagor otherwise equity will intervene, or on the wider ground, that here there was inequality of bargaining power.

On the above the following questions arise: When and at what stages did the bank and the Government become aware of Burmah's predicament? How far did they realise that the December agreement of 1974 did nothing to save the situation? To what extent did they contribute to Burmah being in the position that it was for Burmah made no attempt to obtain an outside purchaser between December 23, 1974, and January 22, 1975. To what extent was the bank responsible for Burmah being in the position it was on January 22, 1975? When exactly was the relationship of mortgagor and mortgagee consummated by the bank becoming guarantors of the American banks and therefore needing the B.P. shares as security? This is an important question for (a) if Burmah went insolvent the American banks would call upon the bank and (b) it constituted the relationship of mortgagor and mortgagee in a way which could not be dissolved. Did the bank consider that the price offered for the shares was fair and was the bank conscious that it was exploiting Burmah, that is, acting unconscionably or did the bank consider that as a banker it could make such a deal without acting unconscionably?

Of course Burmah are unaware what are in the sixty-two documents in question but the bank state that they are relevant and Burmah conjecture that eighteen are principal documents and that the remainder are at a lower level. Ultimately the minister must justify non-disclosure like any other litigant on discovery.

The question of discretion. It was said that Foster J. exercised his discretion in favour of the Crown and that that should be conclusive of

the matter. But it is a closely supervised discretion and in Conway v. Rimmer [1968] A.C. 910 this House emphasised that production should not be ordered without the Crown had a right to appeal. This was meant to be a substantive right of appeal. In the interests of justice it should equally be available to the subject. It is difficult to know on what principles Foster J. exercised his discretion for it does not appear that the judge did so after referring to any of the matters adumbrated above and weighing them in the balance.

Donald Rattee Q.C. and Andrew Smith for the bank. The bank's attitude is summarised in paragraph 56 of the printed case of the Attorney-General. The bank will abide by whatever order the House might make.

S. C. Silkin Q.C. and Peter Gibson for the Attorney-General. Foster J.'s exercise of discretion ought to be left undisturbed supported as it is by a majority of the Court of Appeal. The case for non-disclosure on the ground of public interest is made out unweakened by the attacks upon it. Burmah have not made out a case for disclosure on the grounds of the due administration of justice. If, however, the above two factors had to be weighed in the balance then the appellants have not made out a case and the onus is on them to show that the scales are tipped in their favour.

If the documents already available are sufficient to determine the issues in this action then the documents for which protection is claimed should not be disclosed: see R.S.C., Ord. 24, r. 13, and Compagnie Financière et Commerciale du Pacifique v. Peruvian Guano Co. (1882) 11 Q.B.D. 55.

As to Burmah's general propositions, there is little between the parties. Propositions 1 and 2 are not disputed. As to proposition 3, (1) For certain classes of documents the public interest for non-disclosure may be so clear as to make it unnecessary to bring out "the scales," and (2) if the scales are evenly balanced then the public interest for non-disclosure prevails: see Alfred Crompton Amusement Machines Ltd. v. Customs and Excise Commissioners (No. 2) [1974] A.C. 405, 424, perLord Cross of Chelsea, concurred in by the other members of the House. That is not to say that the Crown does not have to show some public interest involved. It is desired to add a fourth proposition: the decisions whether to order or refuse disclosure or whether to order part disclosure are matters for the judge's discretion and an appellate court will not lightly interfere.

A perusal of the certificate of the chief secretary in the present case makes it absolutely plain that he has followed the guidelines laid down by the cases. As to inspection of the documents, the law stands as it was left in Conway v. Rimmer [1968] A.C. 910, namely, that it is only when the court considers that a prima facie case has been made for disclosure that the court will inspect the documents for the protection of the Crown. On the question of inspection the general principle propounded in Westminster Airways Ltd. v. Kuwait Oil Co. Ltd. [1951] 1 K.B. 134 applies here in that the question whether to exercise it is one for the discretion of the judge who should normally accept the affidavit claiming the immunity.

The above contention is attacked in two ways: (a) The historic interest or time lag argument. Burmah rely upon the fact that it is four and a half years since the events in question took place. It is said that this so weakens the public interest in immunity that disclosure should be ordered. If this argument be correct then public interest protection virtually disappears or, at least, there is a very serious erosion indeed of the principles laid down in Conway v. Rimmer [1968] A.C. 910. (b) A commercial transaction. It is said that the documents relate to matters which are in essence financial transactions and that this aspect means that the policy defence is not open to the Bank: see the judgment of Lord Denning M.R. [1979] 1 W.L.R. 473, 486.

The historic interest and time lag issue was not canvassed in Conway v. Rimmer. The withholding of documents on the grounds of public interest of necessity diminishes with the passage of time. In those circumstances what principles should the judges apply? In a class claim Burmah's argument is very difficult to sustain but it is conceded that it is easier in a contents claim. Protection is given to the class as such. Take for example the class of police informants. Without there is a guarantee that protection against disclosure of this class of claims will subsist whilst the informant or even his family are still alive the value of this class of information will be seriously weakened. Observations in the Radcliffe Report relating to cabinet documents and documents of a like kind are of great assistance. It is pertinent to bear in mind the very stringent rules relating to the papers of an outgoing government. The files are closed to a succeeding administration. This is at least what Lord Reid had partly in mind in Conway v. Rimmer [1968] A.C. 910, 952, when he said, "The business of government is difficult enough as it is, and no government could contemplate with equanimity the inner workings of the government machine being exposed to the gaze of those ready to criticise without adequate knowledge of the background and perhaps with some axe to grind."

There is nothing to show in the present case that by reason of the passage of time there is any less need for protection against disclosure of the classes of documents in question here. It is true that Lord Denning M.R. [1979] 1 W.L.R. 473, 488H, was of the opposite opinion but it is very much an ex cathedra statement and he does not rely on any particular principle. If his view be taken the question arises: What is the relevant period of time? It could be argued that these documents passed into history on January 24, 1975, after the agreements were made. In Attorney-General v. Jonathan Cape Ltd. [1976] Q.B. 752, Lord Widgery C.J. based his judgment on (i) the actual contents of the book; (ii) the lapse of 10 years from the last event recorded therein and (iii) the fact that there had been three General Elections since the events which were the subject matter of the book.

It is conceded that where it is a contents claim it is relevant to ask whether protection is still required at the time when production of the documents is requested, but where it is a class claim lapse of time is not relevant. The persons concerned with the inner workings of government know that their views will not become public knowledge, that is why this class of document has to be protected from disclosure.

Reliance is placed on the Radcliffe Committee Report which in coming to its conclusions dealt most comprehensively with the general principles. Especially relevant are those parts of the report concerned with the inner workings of government and the time element: see paragraphs 40 and 41 where there is a quote from Sir Edward Bridge's memorandum, 1946, para. 8. Attention is drawn in particular to the following paragraphs of the report: paragraphs 30, 31, 33, 36, 37, 49-51, 53, 54, 56, 57, 66 and 82. In Attorney-General v. Jonathan Cape Ltd. [1976] Q.B. 752, 762, 764D, Lord Widgery was clearly of the opinion that cabinet papers were exempt from discovery. That judgment, however, ante-dated the Radcliffe Committee Report. The test to be applied is whether the business of government will become more difficult if documents relating to high policy are produced and thus in the future confidential relationships weakened.

As to Sankey v. Whitlam, 53 A.L.J.R. 11, it was alleged that ministers had engaged in criminal activities in the course of carrying on their ministerial duties, in other words, that they indulged in conduct of an unconstitutional nature. If the documents requested had not been ordered to be produced it would have meant that they could have relied on Crown privilege as a cloak to hide their alleged criminal activities. In Tito v. Waddell (unreported), March 3, 1975, Walton J. held that public interest required withholding of the documents in question there for thirty years.

The commercial transaction argument. On this issue Bridge L.J.'s judgment [1979] 1 W.L.R. 473, 492, is adopted. On the assumption that it is open to Burmah to challenge the language of the chief secretary's certificate such authorities as there are assist the respondents: see Smith v. East India Co. (1841) 1 Ph. 50 and Wadeer v. East India Co. (1856) 8 De G.M. & G. 182. Both decisions were discussed in Conway v. Rimmer [1968] A.C. 910. The fact that Smith v. East India Co. is based on the provisions of a statute does not limit the ambit of the decision. Lord Reid in Conway v. Rimmer [1968] A.C. 910, 944G, supported it on the grounds that "disclosure of non-political documents might have afforded political ammunition to those who criticised this system of government." See also Lord Reid at p. 945 where His Lordship refers to Beatson v. Skene (1860) 5 H. & N. 838. These comments are of particular significance in relation to the possible criticism of government policy. It is emphasised that this is not an issue on the pleadings of the present case, namely, criticism of the government or ministers of the government in office at the time. The Government is not a party to the action. In Conway v. Rimmer [1968] A.C. 910, 962, Lord Morris of Borth-y-Gest referred to both the East India cases. Those cases support the proposition that the fact that the subject matter of legal proceedings consists of matters properly described as commercial transactions is not of itself a good reason for allowing documents to be produced where they relate to matters of government policy. Lord Denning M.R.'s opinion in the present case to the contrary is based on Robinson v. State of South Australia (No. 2) [1931] A.C. 704, 715, 716. But Robinson is a very different case from the present. There was nothing in the documents in question there to suggest that they contained anything other than matters relating to ordinary marketing. It merely follows from that

case that if a State acts in precisely the same manner as a trader in the market then it cannot claim any special privilege relating to documents simply because it is a state. If the ambit of that decision is any wider then it flies in the face of Wadeer v. East India Co., 8 De G.M. & G. 182, and In re Grosvenor Hotel, London (No. 2) [1965] Ch. 1210 where Lord Denning M.R. at p. 1247 stated: "... it does appear that these documents concern policy matters at a high level. They relate to decisions taken by the minister in the national interest; ... I cannot think that these documents are in the least necessary in order that justice may be done." Those observations of Lord Denning M.R. apply precisely to the facts of the present case. See also per Salmon L.J. at p. 1255. Whether protection is made out depends on the application to the facts of the case of the accepted principles in this branch of the law. It is impossible to reconcile the observations of Lord Denning M.R. in the Grosvenor Hotel case at p. 1247 with his observations in the present case [1979] 1 W.L.R. 473, 486.

To summarise this branch of the argument. (i) If the State enters into a transaction with another party of a character which is precisely the same as that any private citizen might enter into documents which are relevant are not protected solely because they emanate from the State. (ii) Documents which belong to a class which is subject to a class claim such as matters relating to high policy are not excluded from protection because they also relate to a commercial transaction whether belonging to the State or other parties. (iii) If it is in the minister's certificate that the documents fall within a protected class prima facie the matter will be taken as so certified. The burden is on those who seek production to show that the documents or part of the documents are not in a protected class.

Applying the above principles to the certificate in the present case: (i) It is a fact that the certificate claims that all the documents relate to the formulation of government policy and are of a provenance which places them in a protected class. (ii) The nature of the discussions which were taking place at the time is wholly consistent with that claim. (iii) The decision to give assistance to Burmah on the terms on which it was given was of a nature likely to relate to matters of high policy. (iv) In addition, both decision and terms were likely to contain matters of a highly political nature. (v) Burmah have failed to demonstrate that the documents or any of them do not fall within the protected class as certified.

The relative facts are: (a) It is an example of the necessity to consider great changes of policy of an exceptional nature at exceptionally short notice. (b) The risk being taken was exceptionally great for (i) Burmah was on the verge of insolvency; (ii) it needed at least £220,000,000 for its foreseeable cash requirements and (iii) its United States requirements were highly speculative and there was no guarantee that Burmah might not lose its interests. Burmah's insolvency would have national and international consequences. (c) There was heavy defaulting on United States and other foreign debts. The strength of sterling would be at severe risk. Confidence in other United Kingdom international companies

was likely to be shaken. Burmah had entered into extensive commitments in the North Sea and its insolvency would have left a serious and large gap. (d) The general economic situation was heavily depressed and therefore it was to be expected that other important United Kingdom companies might find themselves in difficulties and have to seek help from the bank and the Government and therefore the precedent that was set was of very high importance. (e) The acquisition of the B.P. shares was a particularly sensitive part of the transaction and the understanding between B.P. and the Government. The increase in the State shareholding to some 70 per cent. would alter the status of B.P. in the eyes of other governments and other oil companies. On the other hand if the shares were to be sold difficult problems of policy would arise. (f) Once the susceptibility to political criticism is considered all these matters could give rise to criticism. It is unthinkable that any responsible minister would first take the decision as a matter of high policy and then consider it as a commercial transaction. Neither of the alternative arguments put forward by Burmah in relation to the commercial transaction argument is sustainable, that is, neither the halfway nor the whole hog argument. Reliance is placed on Conway v. Rimmer [1968] A.C. 910, 957A, 968B-C and In re Grosvenor Hotel, London (No. 2) [1965] Ch. 1210, 1245, 1246, 1249, 1250, 1256, 1260-1261.

To what extent can Burmah demonstrate that any of the documents in question ought to be disclosed in order fairly to dispose of the action? They failed to demonstrate it to the satisfaction of Foster J. or the majority of the Court of Appeal albeit the majority of the Court of Appeal saw some of the documents claimed which the appellants contended were very helpful to their case. It is difficult to see how this aspect of the case, matters of evidence, differs from the ordinary case on matters of weight where the discretion of the judge should prevail if he has applied the correct principles: see D. v. National Society for the Prevention of Cruelty to Children [1978] A.C. 171, 219, per Lord Diplock. Reliance is placed on Glasgow Corporation v. Central Land Board, 1956 S.C.(H.L.) 1, 11, per Viscount Simonds, 16, per Lord Normand and the last sentence of Lord Radcliffe's opinion at p. 20. It is against that background that the House should consider the question of public interest. In conclusion, no case has been made out by the appellants that disclosure of these documents is necessary for fairly disposing of this action. Unless the appellants are able to show that these documents are so essential to their making a case at all that it would be a grave injustice if they were not disclosed, it is only in those circumstances that the court could consider weighing them in the balance as countervailing against the public interest. The appellants must show that there is a very strong likelihood that the documents in question are of the nature alleged.

Gibson following. It is proposed to consider in what circumstances the court will look at documents. The undoubted power of the court to inspect documents in respect of which a claim has been made on the grounds of the public interest should be exercised "sparingly": Conway v. Rimmer [1968] A.C. 910, 971D, per Lord Morris of Borth-y-Gest;

and then only if there are reasons to doubt the accuracy of the certificate or the cogency of the reasons given by the minister, in which event the court if inclined to order production should as a final check inspect the documents (see p. 953B).

There appear to be only six English cases where the courts have exercised the power to inspect documents. Hennessy v. Wright (1888) 21 Q.B.D. 509 was erroneously referred to by Lord Pearce in Conway v. Rimmer [1968] A.C. 910, 981G, as a case where there was an inspection. Field J. stated obiter that he would have inspected the documents if necessary but he held that their nature was so clear that it was unnecessary so to do. Asiatic Petroleum Co. Ltd. v. Anglo-Persian Oil Co. Ltd [1916] 1 K.B. 822 is not very illuminating on the circumstances when the court will order inspection. Robinson v. State of South Australia (No. 2) [1931] A.C. 704 exemplifies the need for the public interest objection to be taken in proper form. There was an inadequate document from the South Australian Government (p. 722) and it appears that but for delay the Government would have been given opportunity to put its objections clearly. In Spigelman v. Hocken (1933) 150 L.T. 256, a road traffic case, there was a ridiculous claim of privilege. The case is of little assistance on the question of inspection. In re Grosvenor Hotel, London (No. 2) [1965] Ch. 1210 the power to inspect was recognised but was not exercised because the class of documents in question related to policy making.

Alfred Crompton Amusement Machines Ltd. v. Customs and Excise Commissioners (No. 2) [1972] 2 Q.B. 102, 133D-E, is an example of the court (the Court of Appeal) doubting the validity of the claim made for immunity and going on to inspect the documents. In this House [1974] A.C. 405, the claim for immunity on the grounds of public interest was upheld without inspection. In Tito v. Waddell (unreported) March 3, 1975, Walton J. after hearing the plaintiffs' submission, considered that he might have to refuse the application for inspection, which might have left the plaintiffs (who were natives of Ocean Island) with the impression that they had been deprived of vital evidence upon a mere technicality he accordingly took the "somewhat unorthodox solution" of seeking the Crown's consent "in the very exceptional circumstances of the case" to his inspection of certain documents. The two factors stressed in that case by the plaintiffs were the two stressed here, namely, the age and commercial nature of the documents. Consent was given and after such inspection the Crown's claim was held valid.

In Hoffmann-La Roche v. Secretary of State for Trade and Industry, The Times, April 19, 1975, a class claim was made in relation to documents concerned with the formulation of government policy. Templeman J. examined the plaintiffs' pleadings carefully and observed that "relevance is a question of degree and many a document which is sufficiently relevant to appear on discovery plays no part in the litigation. It is sufficient for my purpose, that, without inspecting the documents, I am satisfied that they are not essential and probably not important to the plaintiffs' case and that there are good reasons for non-disclosure."

In Barty-King v. Ministry of Defence (unreported), October 10, 1978, May J., in the special circumstances of that case, sought and obtained the

consent of the Crown to inspect four out of the documents in respect of which the claim for privilege had been made in that there was a likelihood that those documents would disclose the fact that was decisive of the whole action; having inspected those four documents he ordered disclosure of two of them. It is to be observed that the judge was careful to differentiate between the different classes of documents in question in that case.

As to the Scottish cases, under Scots law even well before the decision in Glasgow Corporation v. Central Land Board, 1956 S.C.(H.L.) 1, it was recognised that the minister's certificate was not conclusive. There were only two cases where documents were inspected. In Admiralty v. Aberdeen Steam Trawling and Fishing Co. Ltd., 1909 S.C. 335, the documents were inspected but on appeal this was disapproved of. In Carmichael v. Scottish Co-operative Wholesale Society Ltd., 1934 S.L.T. 158 the Lord Ordinary inspected the documents because the minister had issued his certificate without inspection. But in Rogers v. Orr, 1939 S.C. 492, there was express disapproval of the course taken by the Lord Ordinary in the Carmichael case. Glasgow Corporation v. Central Land Board, 1956 S.C. (H.L.) 1, is not helpful on this particular aspect.

The cases subsequent to Conway v. Rimmer [1968] A.C. 910 show that the guidance given by this House has been followed. They show that the power to inspect is a residual and rare power and one to be exercised sparingly. Doubtless if there is reason to question the accuracy of the Minister's certificate or his bona fides then that would be grounds for inspecting documents or if the Minister's reasons were such that the court can weigh those reasons and the public interest on balance requires disclosure. Up to the present time policy making documents have been regarded as a class of such importance that to use the words of May J. in Barty-King v. Ministry of Defence (unreported), October 10, 1978, the court does not even bring out the scales.

It cannot be denied that the applicant and the court are in very considerable difficulty in that the applicant rarely knows what is in any particular document and the court can never be absolutely certain that the document does not contain something of great probative value in the litigation. As the 1967 Law Reform Committee headed by Lord Pearson on Privilege in Civil Proceedings, pointed out, all objections to production of relevant documents may result in injustice being done. It is contended, however, that there are similarities in the court's approach to a claim for legal professional privilege and to the claim of public interest immunity. R.S.C., Ord 24, r. 2 R.S.C., and Ord. 24, r. 5 apply to all documents. Once an objection has been made then different rules apply: see R.S.C., Ord. 24, r. 11 which is governed by R.S.C., Ord. 24, r. 13. Difficult questions may arise in respect of a claim of legal professional privilege where the court may have difficulty in deciding whether the privilege has been properly claimed. For example, at what stage was litigation contemplated? See Westminster Airways Ltd. v. Kuwait Oil Co. Ltd. [1951] 1 K.B. 134. But it is contended that the same approach is adopted and the same burden is placed on an applicant applying for disclosure. It is not sufficient for an applicant to rely on speculation or indulge in a pure fishing exercise.

As to the Category C documents the House is urged to accept the

observations of Bridge L.J. [1979] 1 W.L.R. 473, 492G-H where the relevant principle to be applied is set out with great clarity. One is here dealing with a matter of great importance to the public service. The approach of English law is not that exemplified in Nixon v. United States, 418 U.S. 683.

Hoffmann Q.C. in reply. The Crown's contentions have greatly narrowed the issues. The Crown accepts the appellants' first proposition and footnote (b) to it. It is fundamental to the Crown's case that one should identify the reason why high policy documents need to be protected. The Crown adopts the arguments of policy contained in paragraphs 51 and 53 of the Radcliffe report. Those are valid and important considerations and they are already protected in a number of ways by law and by conventions of the constitution. It must be right to conserve confidence and not to inhibit candour in the tendering of advice and making policy decisions. The Crown underlined this portion of their argument in citing Smith v. East India Co., 1 Ph. 50, which contains the purest judicial expression of the candour argument. The candour argument looks to the future conduct of the administration. There is quite a battery of rules to conserve confidences and candour: see the Franks Report at chapter 5 and the 50 (now 30) years rule contained in the Public Records Act 1958. See also section 2 of the Official Secrets Act 1911. There is also the tradition and discipline of the Civil Service and the convention which prevents a new administration searching the files of the previous administration to ascertain if a civil servant has been too zealous in serving the previous government. Moreover, there is the possibility of the Crown obtaining an injunction for breach of confidence by an ex minister or retired civil servant. There is also now the adoption of the recommendation made by the Radcliffe Report which protects a civil servant from being identified by name for his views for fifteen years.

The question in the present case is: Does the preservation of candour require that there will be no disclosure in a court of law despite the safeguards that a court of law can impose? If a balancing process is involved then as part of that process the House can take into account whether candour in government would be impaired by disclosure in a court of law. Of course there are special relationships such as police informers, and quasi judicial tribunals. That is not disputed. That is not the Crown's case here. The Crown puts forward a proposition of the widest generality. Paragraph 106 of the Franks Report is relevant in the present connection. See also paragraphs 176 and 179.

What is the strength of the argument that candour in government requires public interest protection? Conway v. Rimmer [1968] A.C. 910 shows (i) that the necessity for protection is an issue that the House is able to assess for itself and (ii) that the House place small weight on the candour argument: see for example per Lord Upjohn at p. 994A. Further, although reasons of policy are put forward by the Crown as a valid issue it is not a strong argument for giving public interest protection: see the speech of Lord Radcliffe in Glasgow Corporation v. Central Land Board, 1956 S.C. (H.L.) 1. There is no conflict between what he states there and what is stated in the Radcliffe Report.

Burmah concede that the time factor does not operate strongly in

relation to the candour argument. As to the grounds which Viscount Simon stated in Duncan v. Cammell, Laird & Co. Ltd. [1942] A.C. 624, 642, would not afford a minister adequate justification for objecting to production of documents, it is difficult to believe that Lord Reid would have wished to dissent from those observations. If the candour argument is put on one side, Lord Reid in Conway v. Rimmer [1968] A.C. 910, 952, must have meant that whilst persons are carrying on government it would be very difficult if at the same time they had to fend off criticism. This is how the matter is put by May J. in Sankey v. Whitlam, 53 A.L.J.R. 11. But this reason ceases to apply once the policy has been carried into execution. Further, neither the Crown nor the Court of Appeal in the present case have placed reliance on the time element. There are no state secrets in question here, it is purely a candour argument.

As to Lord Reid's treatment in Conway v. Rimmer [1968] A.C. 910, 944, 945, of Smith v. East India Co., 1 Ph. 50, and Beatson v. Skene, 5 H. and N. 838, the House is invited to reject the Crown's argument that prospective political criticism is itself a ground for protection apart from the candour argument.

As against the generalised claim for protection, there is a public interest in the due administration of justice. It may be thought that justice has not been done if some documents are not disclosed especially where the action is against an emanation of the government and where no governmental act is involved and where the proceedings are akin to a commercial transaction which could be brought against an ordinary commercial institution. One reason for stating that the bank acted unconscionably is that on the relevant day they were acting as their own customers. It was not one act but a series of acts which resulted in the transaction complained of.

As to the commercial nature of the transaction, it is important where there is a commercial transaction involving the government or a governmental institution that the Crown should be in the same situation as a private litigant. It is on these grounds that the Gaming Board case (Reg. v. Lewes Justices, Ex parte Secretary of State for the Home Department) [1973] A.C. 388, Wadeer v. East India Co., 8 De G.M. & G. 182, and the Hoffmann-La Roche case, The Times, April 19, 1975 are distinguished. Thus in the Le Roche case there was plainly a challenge to governmental action and policy.

On the burden of proof, the Crown has read too much into the speech of Lord Cross of Chelsea in Alfred Crompton Amusement Machines Ltd. v. Customs and Excise Commissioners (No. 2) [1974] A.C. 405. His Lordship's observation should not be erected into a general doctrine on the burden of proof. In the Crompton case it was held that the Customs & Excise Commissioners were the best judges of how best to protect the information that they received. As to the cases relating to inspection referred to by Mr. Gibson, they are helpful as far as they go but those that were decided before Conway v. Rimmer [1968] A.C. 910, are not helpful for the law was changed by that case. Further, Tito v. Waddell (unreported), March 3, 1975, is of little assistance since it was a special case where the documents were inspected with the agreement of the Crown. There is no indication of the course that Walton J. would have taken if the Crown had been non-co-operative. The general principle relating to inspection was

that enunciated by Lord Upjohn in Conway v. Rimmer [1968] A.C. 910, 995F - 996D It is difficult to see how the public interest is damaged if judges in private inspect the documents in question. To contend that if there is a class of documents which the minister states in his opinion should not be disclosed the court is therefore precluded from inspecting them even in private is a pre-Conway v. Rimmer argument. In Conway v. Rimmer it was held that the minister's certificate was not ipso facto conclusive.

Their Lordships took time for consideration.

November 1, 1979.LORD WILBERFORCE. My Lords, in this action the appellant the Burmah Oil Co. Ltd. ("Burmah") is suing the Governor and Company of the Bank of England ("the bank") for relief in respect of the sale to the bank by Burmah in 1975 of 77,817,507 ordinary stock units of £1 each of the British Petroleum Co. Ltd. ("B.P.") at a price of approximately £179 million. Burmah claims, in brief, that this price represented a substantial undervalue of the stock and that the bargain was unconscionable, inequitable and unreasonable. It is important to understand that this action, and these issues, arise exclusively between Burmah and the bank.

The present appeal arises out of an application by Burmah for production of 62 documents listed in the list of documents served by the bank. The bank on the instructions of the Crown have objected to produce these on the ground that they belong to classes of documents production of which would be injurious to the public interest. They have put forward a certificate dated October 18, 1977, signed by the Chief Secretary to the Treasury supporting this objection. On the interlocutory hearing of the objection in the High Court Her Majesty's Attorney-General intervened in order to argue the case in support of it, and it was upheld by Foster J. On appeal by Burmah to the Court of Appeal [1979] 1 W.L.R. 473, the Attorney-General took a similar course, and that court, by majority, affirmed the judge. On a further appeal to this House, the Attorney-General was joined as a respondent and as such argued the case against production; the bank, as in the High Court and the Court of Appeal, took no part in the argument. But, I repeat, the only defendant in the action is the bank.

My Lords, in an interlocutory matter involving a large element of discretion, which has been concurrently decided by both courts below, I apprehend that your Lordships should be reluctant to intervene and indeed should only do so if of opinion that some different principle of law from that accepted below ought to be applied. On the view which I take of the case, it is not one for a different exercise of the court's discretion, indeed I regard it as a straightforward one, more so than is usually found in this class of case. I shall deal with the suggestion that some extension or change in the existing law ought to be made.

The starting point in the discussion must be the certificate of the Chief Secretary. This is a lengthy and detailed document to which justice cannot be done without setting it out in full. It is perfectly clear that this document

represents the result of careful and responsible consideration: that the minister has read and applied his mind to each of the documents: that, to adopt language used by the courts in other cases, the minister has not merely repeated a mechanical formula, that the certificate is not "amorphous" or of a blanket character, but is specific and motivated. Further, the minister has not contented himself with a general assertion that production would be injurious to the public interest, he has stated very fully the reasons why this would in his opinion be so: in summary that they concern discussions at a very high level, as to one category at ministerial level, and as to another the highest official level, as to the formulation of government policy. He has not even contented himself with a general reference to government policy. He has specified this as concerned with (a) the possible effect of a collapse of Burmah upon the £ sterling, upon other British companies with large overseas borrowings, upon the government's North Sea oil policy, and the future production of North Sea oil and correspondingly on the expectation which might be aroused on the part of other private borrowers if Burmah were to receive assistance, (b) the international and other consequences of a sale of the B.P. stock to the bank - which would bring the government shareholding up to 70 per cent. - and (c) as regards possible further financial support to Burmah after January 1975 having regard to the possible consequences of a financial collapse by Burmah.

It is apparent that these identified matters of policy were of the highest national and political importance and that they called for formulation of policy at the highest governmental levels, including the Cabinet, involving directly several ministers in the Treasury, the Department of Energy and the Paymaster General, and, in the first two mentioned departments, handled by the Permanent Under Secretary of State.

Omitting some formal passages, the certificate is as follows:

"3. I have personally read and carefully considered all the documents listed in the schedule and I have formed the opinion that their production would be injurious to the public interest for the reasons hereinafter set out. 4. The documents listed in the schedule fall within three categories described below. There is or are shown in the schedule against each document listed the appropriate category or, where a document falls within more than one category, the appropriate categories. The three categories are as follows: -

"Category A

"These consist of communications between, to and from ministers (including ministers' personal secretaries acting on behalf of ministers) and minutes and briefs for ministers and memoranda of meetings attended by ministers. All such documents relate to the formulation of the policy of the government - (a) in face of the financial difficulties of the Burmah Oil Co. Ltd. (hereinafter called 'Burmah') in December 1974 and January 1975, and having regard especially to: - (i) the likely effect of the default of Burmah in respect of a large dollar loan upon: - (a) The £ sterling, (b) Other British companies with large overseas borrowings; (ii) the possible effect of a financial collapse by

Burmah upon the government's North Sea oil policy and upon the future production of North Sea oil; (iii) the expectations which would be aroused on the part of other private borrowers defaulting on dollar debts if Burmah were to receive assistance; (b) in consequence of the measures taken in response to Burmah's said financial difficulties and in particular as to what was to be done with the B.P. stock sold by Burmah to the bank in January 1975 having regard especially to the international consequences of a sale by the bank of that stock; (c) in connection with the giving of further support to Burmah after January 1975, having regard particularly to the international consequences of a financial collapse by Burmah and the effect of such a collapse on the government's North Sea oil policy.

"Category B

"These consist of communications between, to and from senior officials of the Department of Energy, of the Treasury and of the bank including memoranda of meetings of and discussions between such officials, and drafts prepared by such officials (including drafts of minutes and briefs comprised in category A), all such communications and drafts relating to the formulation of one or more aspects of the policy described in category A.

[His Lordship said that the category C documents did not call for separate consideration.]

"5. Many of the documents listed in the schedule though in the possession, custody or power of the bank, were not brought into existence by the bank or addressed to the bank. The bank occupies a unique position in relation to the government. Though distinct from the government, it is the principal banker to the government and, inter alia, performs the function of advising the government in the field of economic and financial affairs. The bank is frequently consulted by the government particularly when policy decisions in that field fall to be taken. Through the Governor, Deputy Governor and other of its officials it often takes part together with officers of the Treasury and other government departments in the process of briefing and advising ministers. To assist the bank in the performance of its functions it is supplied by the government with many confidential documents. In addition the bank brings into existence and itself receives documents in the course of its participation in the process of the formulation of government policy. These are as much a part of the decision-making process as the internal documents of government departments relating to the formulation of policy.

"6. It is, in my opinion, necessary for the proper functioning of the public service that the documents in category A and category B should be withheld from production. They are all documents falling within the class of documents relating to the formulation of government policy. Such policy was decided at a very high level, involving as it did matters of major economic importance to the United Kingdom. The documents in question cannot properly be described as

routine documents. Those in category A are all documents passing at a very high level, including communications intended for the guidance and recording the views of the Prime Minister or recording discussions at a very high level. The documents in category B though passing at a lower level or recording discussions at a lower level, nevertheless all relate to the policy decisions to be taken at a higher level. Decisions made by ministers are frequently preceded by detailed discussion within and between government departments (and in appropriate cases, of which the present is one, within the bank and between the bank and government departments) and by consideration of the various possibilities open to ministers. It is out of such discussion and consideration that the advice to be tendered to ministers is often formulated (frequently, initially, in the form of drafts of documents intended for the consideration and approval of ministers) and the decisions of ministers are often reflected in departmental documents passing at a lower level. This is true of the present case. More generally, it would, in my view, be against the public interest that documents revealing the process of providing for ministers honest and candid advice on matters of high level policy should be subject to disclosure. In this connection, I would respectfully agree with the reasoning of Lord Reid in Conway v. Rimmer [1968] A.C. 910. 952, to whose remarks ... my attention has been drawn, as regards the effect on the inner workings of the government machine of the public disclosure of documents concerned with policy.

[His Lordship omitted 7 as dealing with category C.]

"8. I understand that oral evidence may be given in these proceedings. If oral evidence were sought to be given of the contents of any of the documents to the production of which I have in this certificate objected, I would wish to object to such evidence on the same grounds as those hereinbefore set out in relation to the documents in question.

There followed a schedule listing 62 documents and specifying into which category or categories they fell.

The claim to "public interest immunity" in respect of these documents is clearly what has come under a rough but accepted categorisation to be known as a "class" claim, not a "contents" claim, the distinction between them being that with a class claim it is immaterial whether the disclosure of the particular contents of particular documents would be injurious to the public interest - the point being that it is the maintenance of the immunity of the class from disclosure in litigation that is important; whereas in a contents claim the protection is claimed for particular contents in a particular document. A claim remains a class even though something may be known about the contents: it remains a class even if parts of documents are revealed and parts disclosed. The appellant did not, I think, dispute this. And, the claim being a class claim, I must state with emphasis that there is not the slightest ground for doubting that the documents in question fall within the class described: indeed the descriptions themselves and references in disclosed documents make it clear that they do. So this is not

one of those cases, which anyway are exceptional, where the court feels it necessary to look at the documents in order to verify that fact. We start with a strong and well-fortified basis for an immunity claim.

I now deal with the two main arguments used by the appellants. The first is to seek to make a distinction between a decision to allow the bank to buy the B.P. stock and a decision as to price: the first, it is said, may be "policy," the second is something less than policy. I have to reject this distinction. The whole course of negotiation of which, as I shall explain, we know a great deal, shows that these two matters were indissolubly linked as part of one decision. It is indeed inconceivable that any responsible minister or civil servant would regard the only matter of policy to be decided to be the purchase of the stock in principle and would leave over the matter of price as one merely of "nuts and bolts."

The second argument is perhaps more plausible, it is to say that. whatever may have been the need to protect governmental policy from disclosure at the time (1975) all is now past history: the decision has been made; the sale has gone through; Burmah has been saved from collapse. So what is the public interest in keeping up the protective screen?

I think that there are several answers to this. The first (and easiest) is that all is not past history - at least we do not know that it is. Government policy as to supporting private firms in danger of collapse: as to ownership of B.P. stock: as to the development of North Sea oil is ongoing policy; the documents are not yet for the Record Office. They are not, to use a phrase picked out of Lord Reid's speech in Conway v. Rimmer [1968] A.C. 910, 952, of purely historical interest. Secondly the grounds on which public interest immunity is claimed for this class of document are, no doubt within limits, independent of time. One such ground is the need for candour in communication between those concerned with policy making. It seems now rather fashionable to decry this, but if as a ground it may at one time have been exaggerated, it has now, in my opinion, received an excessive dose of cold water. I am certainly not prepared - against the view of the minister - to discount the need, in the formation of such very controversial policy as that with which we are here involved, for frank and uninhibited advice from the bank to the government, from and between civil servants and between ministers. It does not require much imagination to suppose that some of those concerned took different views as to the right policy and expressed them. The documents indeed show that they did. To remove protection from revelation in court in this case at least could well deter frank and full expression in similar cases in the future.

Another such ground is to protect from inspection by possible critics the inner working of government while forming important governmental policy. I do not believe that scepticism has invaded this, or that it is for the courts to assume the role of advocates for open government. If, as I believe, this is a valid ground for protection, it must continue to operate beyond the time span of a particular episode. Concretely, to reveal what advice was then sought and given and the mechanism for seeking and considering such advice, might well make the process of government more difficult now. On this point too I am certainly not prepared to be wiser than the minister. So I think that the "time factor" argument must fail.

The basis for an immunity claim, then, having been laid, it is next necessary to consider whether there is any other element of public interest telling in favour of production. The interest of the proper and fair administration of justice falls under this description. It is hardly necessary to state that the mere fact that the documents are or may be "relevant" to the issues, within the extended meaning of relevance in relation to discovery, is not material. The question of privilege or immunity only arises n relation to "relevant" documents and itself depends on other considerations, viz., whether production of these documents (admittedly relevant) is necessary for the due administration of justice. In considering how these two elements are to be weighed one against the other, the proper starting point must be the decision of this House in Conway v. Rimmer [1968] A.C. 910. That case established the law in line with that thought to exist in Scotland, and it is also well in line with Commonwealth authority: see Sankey v. Whitlam (1978) 53 A.L.J.R. 11 and Robinson v. State of South Australia (No. 2) [1931] A.C. 704. The latter case - strongly relied on by Lord Denning M.R. - does not in my view lay down any principle diverging from or extending beyond Conway v. Rimmer and was indeed absorbed in that case. Of course Conway v. Rimmer, as the speeches of their Lordships show, does not profess to cover every case, nor as it frozen the law, but it does provide a solid basis for progress as regards the point now under discussion.

It may well be arguable whether, when one is faced with a claim for immunity from production on "public interest" grounds, and when the relevant public interest is shown to be of a high, or the highest, level of importance, that fact is of itself conclusive, and nothing which relates to the interest in the administration of justice can prevail against it. As Lord Pearce said in Conway v. Rimmer [1968] A.C. 910, 987: "Obviously production would never be ordered of fairly wide classes of documents at a high level" and see Reg. v. Lewes Justices, Ex parte Secretary of State for the Home Department [1973] A.C. 388, 412 per Lord Salmon. In the words of May J. in Barty-King v. Ministry of Defence (unreported), October 10, 1978 (concerned with internal thinking and policy at a high civil service level), it is not even necessary to bring out the scales. Mr. Silkin for the Attorney-General did not contend for any such rigorous proposition, i.e. that a high level public interest can never, in any circumstances, be outweighed. In this I think that he was in line with the middle of the road position taken by Lord Reid in Conway v. Rimmer and also with the median views of the members of the High Court of Australia in Sankey v. Whitlam, 53 A.L.J.R. 11 - see particularly the judgment of Gibbs A.C.J. I am therefore quite prepared to deal with this case on the basis that the courts may, in a suitable case, decide that a high level governmental public interest must give way to the interests of the administration of justice.

But it must be clear what this involves. A claim for public interest immunity having been made, on manifestly solid grounds, it is necessary for those who seek to overcome it to demonstrate the existence of a counteracting interest calling for disclosure of particular documents. When this is demonstrated, but only then, may the court proceed to a balancing

process. In Conway v. Rimmer [1968] A.C. 910 itself it was known that there were in existence probationary reports on the plaintiff as to which an obviously strong argument could be made that their disclosure was necessary if the plaintiff's claim were to have any hope of succeeding (in the end they turned out to be far from helpful to him): so the court had something very definite to go upon which it could put into the scales against the (minor) public interest of not revealing routine reports. So, too, in Sankey v. Whitlam, 53 A.L.J.R. 11, the High Court thought it scarcely credible that the documents in question would not reveal factual material as to Commonwealth borrowing. But the present case is quite different. There is not, and I firmly assert this, the slightest ground, apart from pure speculation, for supposing that there is any document in existence. among those which it is sought to withhold, or anything in a document which could outweigh the public interest claim for immunity. I make this assertion good under two heads.

1. A very full and careful disclosure has been made of all documents bearing upon negotiations between Burmah and the bank leading to the sale of the stock. As was said by Templeman L.J. [1979] 1 W.L.R. 473, 495: "... very great care has been taken to conceal the minimum and to produce consistently with the public interest every document which is relevant to the action." Thus, disclosure has been made of all documents relating to occasions and discussions at which representatives of Burmah were present. Disclosure has been made of all factual reports of meetings with Burmah and of Burmah's statements to the bank and vice versa. There is a mass of these documents which your Lordships have read and carefully considered. What are withheld are documents, or in some cases parts of documents, recording discussions either between the bank and the government or public officials, or internally within the government or the civil service, when Burmah was not present. Since the proceedings are between Burmah and the bank, there must be a strong argument for believing that the dividing line has been correctly drawn; that what is necessary to Burmah's case has been disclosed, and that what has not been disclosed cannot help Burmah's case. However, I shall explore this further in relation to the actual issues.

2. The exact nature of Burmah's claim against the bank is not very clear, but I need not, indeed should not, analyse it for present purposes. I must not be taken as holding that there is any support for it in law; or the contrary: the claim must at this stage be treated as valid. There are very detailed pleadings in which issues of fact and also matters of evidence are stated. The basic claim is that the sale of the B.P. stock at the price of £179 million ought to be reopened. For this four grounds are stated, namely that the sale:

"(a) was unconscionable, inequitable and unreasonable; (b) was procured by the bank acting in breach of its duty of fair dealing and taking an unfair and unconscionable advantage of Burmah; (c) wrongfully purported to render nugatory Burmah's right to redeem; (d) involved the bank obtaining an improper collateral advantage in connection with the taking of security."

In support of these Burmah [in paragraph 54 of its statement of claim] relies on nine points:

"(1) the inequality of bargaining power of the bank and Burmah; (2) the advantage taken by the bank of Burmah's temporary financial predicament; (3) the sale of the B.P. stock at an undervalue; (4) the absence of profit sharing; (5) the delay by the bank in stating its proposals; (6) the failure of the bank to provide guidelines for sale to third parties; (7) the inability of Burmah to seek assistance elsewhere; (8) the duty of the bank referred to in paragraph 4 above [i.e., a 'duty of fair dealing']; (9) the loss of dividend income from the B.P. stock."

Now it is clear from this presentation of the case that the success or failure of Burmah's claim must rest upon the objective nature of the bargain, the negotiations between Burmah and the bank, the relevant circumstances or bargaining power of the parties and any pressure which the bank may be shown to have exercised upon Burmah. As to these matters, much is common ground and is admitted on the pleadings: the mass of documents disclosed provides part of what is necessary to make good the remainder: part may depend upon oral evidence given by representatives of Burmah, representatives of the bank and by representatives of the important financial institutions which were advising Burmah in the negotiations. How then can anything said between the bank and the government or its officials help to resolve these issues? Whether the bank acted at the behest of the government or not, seems to me (with all respect to some views expressed in the Court of Appeal) irrelevant. After a lengthy and thorough argument I was only able to perceive two as to which a case of any kind could be made. The first relates to issue (4) above: the absence of profit sharing. The fact here was that the bank, at one time in the negotiations, had suggested that the sale of the B.P. stock should involve an agreement that if (as might have seemed likely) the value of the stock were to rise, Burmah should share in the profit. It is clear already from disclosed documents (a) that the bank thought that a profit sharing term would be fair, (b) that the bank, and later Burmah itself, so represented to the government, (c) that the government refused to agree: all this is as plain as can be from the documents, I cannot see therefore that any document passing between the bank and the government, or recording any discussion between the bank and the government, or internally to the government could assist Burmah's case. If any such reference at all were made in any such document to profit sharing (a matter of pure speculation) why should it be in terms different from what the bank had said to Burmah? If it were in the same terms it would add nothing to this case. The second relates to the general allegation that the sale was unconscionable. In my opinion establishment of this must depend upon the objective facts as indicated above. But Burmah submits otherwise. Their claim is rather remarkable. It is "very likely," to use their own words, that some of the withheld documents (they specify 18 out of the 62) may record a statement by a representative of the bank to the government, or at some meeting, that in the opinion of the bank the sale, on terms fixed by the

government was "unfair" - even, in forensic hyperbole, "grossly unfair" and that this would greatly assist Burmah's case.

My Lords, I am willing to leave aside the question whether, on the pleaded issues, such a statement - made as it would have been in the course of fluid discussions - would assist Burmah to prove its case which as I have said depends on objective considerations and on action between Burmah and the bank. For, in my opinion, the claim is on its face totally unjustified. The phrase "very likely" is a pure ipse dixit of Burmah and unsupported by any evidence: there is nothing anywhere which affords the slightest support for it: any finding that such a statement exists is "very likely," "likely" or "not unlikely," or as to its possible terms, can not judicially be made. That such a document may exist, and what it may contain, is the purest speculation.

This brings me to the issue of inspection. For now it is said, "Well, let us look at the documents and see - to do so cannot do any harm. If there is nothing there no damage will be done: if there is, we can weigh its importance." As presented (and to be fair to Burmah's very able counsel, such a submission occupied a far from prominent place in their argument) this may appear to have some attraction. But with all respect to those who think otherwise, I am firmly of opinion that we should not yield to this siren song. The existing state of the authorities is against it: and no good case can be made for changing the law. Indeed, to do so would not in my opinion be progress.

As to authority, before Conway v. Rimmer [1968] A.C. 910, although the court had power to inspect any document, the question whether to exercise it was treated as one for the discretion of the judge, who, it was said, should normally accept the affidavit claiming the immunity: see Westminster Airways Ltd. v. Kuwait Oil Co. Ltd. [1951] 1 K.B. 134. In In re Grosvenor Hotel, London (No. 2) [1965] Ch. 1210 a number of conflicting opinions were expressed both as to the law and the facts. I think that the Court of Appeal regarded the objection of the minister as defective and considered therefore that the documents could be inspected for the court to form its own opinion as to the public interest. In Conway v. Rimmer [1968] A.C. 910 itself, it was said that the power should be exercised "sparingly" (per Lord Morris of Borth-y-Gest, p. 971), and then only if there are reasons to doubt the accuracy of the certificate or the cogency of the minister's reasons. Inspection should be by way of final check. Or, as Lord Upjohn put it, inspection should be made if the judge "feels any doubt about the reason for [the document's] inclusion as a class document" (p. 995). In Alfred Crompton Amusement Machines Ltd. v. Customs and Excise Commissioners (No. 2) [1974] A.C. 405 this House upheld the claim to public interest immunity without inspecting the documents, although that course had been taken by the Court of Appeal. In first instance cases, the judges have treated the power to inspect as an exceptional one, to be rarely used: in two instances they sought and obtained the Crown's consent to inspect selected documents: Tito v. Waddell (unreported), March 3, 1975, and Barty-King v. Ministry of Defence (unreported), October 10, 1978. This is inconsistent with the recognition of a general right or duty to inspect.

As to principle, I cannot think that it is desirable that the courts should assume the task of inspection except in rare instances where a strong positive case is made out, certainly not upon a bare unsupported assertion by the party seeking production that something to help him may be found, or upon some unsupported - viz., speculative - hunch of its own. In the first place it is necessary to draw a reasonably clear line between the responsibility of ministers on the one hand, and those of the courts on the other. Each has its proper contribution to make towards solution of the problem where the public interest lies - judicial review is not a "bonum in se" it is a part - and a valuable one - of democratic government in which other responsibilities coexist. Existing cases, from Conway v. Rimmer onwards, have drawn this line carefully and suitably. It is for the minister to define the public interest and the grounds on which he considers that production would affect it. Similarly, the court, responsible for the administration of justice, should, before it decides that the minister's view must give way, have something positive or identifiable to put into the scales. To override the minister's opinion by "amorphous" phrases, or unsupported contentions, would be to do precisely what the courts will not countenance in the actions of ministers. Secondly, decisions on grounds of public interest privilege fall to be made at first instance, by judges or masters in chambers. They should be able to make these decisions according to simple rules: these are provided by the law as it stands. To invite a general procedure of inspection is to embark the courts on a dangerous course: they have not in general the time nor the experience, to carry out in every case a careful inspection of documents and thereafter a weighing process. The results of such a process may, indeed are likely, to be variable from court to court and from case to case. This case provides an example of opposite conclusions come to upon identical materials: see [1979] 1 W.L.R. 473. This inevitable uncertainty is not likely to do credit to the administration of justice and is bound to encourage appeals.

In the end, I regard this as a plain case: of public interest immunity properly claimed on grounds of high policy on the one hand in terms which cannot be called in question; of nothing of any substance to put in the scale on the other. I return to the point that both courts below have refused to exercise a discretionary power to order production of these documents, or to inspect them. Their decision can only be reversed if they erred in law. To say that they erred in law in not inspecting the documents involves the proposition that there is a duty, either in all cases or at least in such a case as this, to inspect. In my opinion it is not the law, and ought not to be the law, that there is any such duty. In saying this and in the previous discussion I have done no more than adopt, with greater prolixity, the completely convincing judgment of Bridge L.J. (with whom in substance Foster J. and Templeman L.J. agreed) in the Court of Appeal.

I would dismiss the appeal.

LORD SALMON. My Lords, in December 1974 the Burmah Oil Co. Ltd. (which I shall refer to as "Burmah") was facing ruin and needed massive financial assistance to survive. Burmah had borrowed £54,000,000 on loan stock in England and $650,000,000 loans in the United States of

America to finance its business. In the normal course, the maturity date for these loans would have been 1991/6. Towards the end of 1974 there had been a very serious slump - one of the worst since 1929. Having regard to factors I need not recite Burmah's creditors could have become entitled to require immediate repayment of the loans. There was a great risk that unless somebody came to Burmah's rescue, Burmah would be forced into liquidation and a receiver and manager would be appointed. This would have put in jeopardy all Burmah's assets which included their most important interests in the North Sea oil fields.

When towards the end of December 1974, the Bank of England (which I shall call "the bank") and the government heard of the alarming debacle facing Burmah, which, unless avoided, could do serious harm to sterling and other important national interests, it was decided to come to Burmah's rescue.

According to Burmah, it might well have been possible for it to obtain the necessary financial support from abroad. This support, however, might have had conditions attached to it which would have seriously prejudiced the national economy; so the bank asked Burmah not to look for help abroad until it had discovered what help the bank could offer. Between December 23, 1974, and January 23, 1975, there were constant negotiations between Burmah, the bank, the Treasury and the Department of Energy.

Burmah's most valuable assets were about 78 million stock units in the British Petroleum Co. Ltd. (which I shall call "B.P."). By December 31, 1974, outline proposals had been provisionally agreed between Burmah and the bank to the effect that the bank should produce the necessary cash and guarantees to save Burmah, and Burmah should mortgage its 78 million stock units in B.P. to the bank and transfer 51 per cent. of its interests in the North Sea oil fields to the government for a price to be negotiated later. The market value of the B.P. stock units had been about £6 per unit in July 1974 but, owing to the slump, had fallen to below £2 towards the end of December 1974.

By January 10, 1975, it had become apparent that the position of Burmah was even more precarious and that even more help was needed to save it than had previously been anticipated. On about January 10 the bank told Burmah that the previously suggested mortgage of Burmah's B.P. stock would not meet the situation and suggested that Burmah should sell its B.P. stocks to the bank at its current market price and that any profit made by the bank out of this transaction should be shared between the bank and Burmah. At this time, Burmah thought that the B.P. stock units were likely to recover to the price they had commanded in July 1974.

According to Burmah, five days went by before it learned from the bank that the government did not agree to any profit sharing scheme with Burmah. Seven days later on January 22, the Governor and the Deputy Governor of the bank told Burmah that it was and always had been their view that a profit sharing scheme between Burmah and the bank was reasonable and should be arranged, but that the government would not agree to any such scheme. It was made plain to Burmah that the government was not prepared to rescue Burmah unless its B.P. stock units were sold to the bank at £2.30 per unit, i.e. 38 pence below the then rising

market price and without a profit sharing scheme of any kind between the bank and Burmah.

Burmah had no choice but to accept these terms. The selling price amounted to about £179,000,000. By July 1975 the market price per unit had recovered to about £5.50. Accordingly the stock which had been bought by the bank only a few months previously for about £179,000,000 was already again worth about £480,000,000 in July 1975.

In the present action, Burmah claims against the bank that the purchase of the B.P. stock at the price of £179,000,000 should be re-opened mainly on the grounds that the purchase: "(a) was unconscionable, inequitable and unreasonable; (b) was procured by the bank acting in breach of its duty of fair dealing and taking an unfair and unconscionable advantage of Burmah; ..." My Lords, it has been conceded, but only for the purpose f this appeal, that Burmah has a good cause of action if it can establish its allegations against the bank.

The present appeal is however concerned only with discovery of documents. The bank disclosed a very long list of documents but never itself objected to producing any of them for inspection by Burmah. The government, however, objected to the production of 62 of these documents on the ground of public interest immunity. On the interlocutory hearing at first instance, the Attorney-General successfully intervened to support the objection to the production of the 62 documents in question. On the unsuccessful appeal by Burmah to the Court of Appeal the Attorney-General took a similar course; and on appeal to this House the then Attorney-General was joined as a respondent and counsel appearing for him argued the case against production. The bank, in this House, as in both the courts below, took no part in the argument.

I am in no way criticising the very full and carefully prepared certificate of the Chief Secretary to the Treasury clearly explaining why in his view it would be contrary to the public interest if any part of the classes of documents were to be produced for inspection which he has listed in his certificate under categories A, B and C. The Chief Secretary's certificate is set out verbatim in the speech of my noble and learned friend Lord Wilberforce, and I will not repeat it. [His Lordship read the paragraph headed "Category A" from the certificate and continued:] Category B consists of recorded communications between senior officials relating to the formulation of one or more of the aspects of government policy described in category A.

Category C consists of recorded communications made in confidence by important companies or businessmen to ministers or senior officials in relation to commercial or financial information. In my view, the classes of documents in categories A and B should be treated alike. The class of documents in category C seems to me to be irrelevant and could, in any event, be protected from production on the principles laid down in Reg. v. Lewes Justices, Ex parte Secretary of State for the Home Department [1973] A.C. 388 and D. v. National Society for the Prevention of Cruelty to Children [1978] A.C. 171.

In my opinion, this case has nothing whatever to do with whether the bank was right or wrong in deciding to rescue Burmah from the disaster

which faced it - still less with the reasons which prompted the bank to arrive at this decision. I entirely agree that the reasons for rescuing Burmah were of the highest national and political importance and that such parts of the documents in categories A and B (if any) which dealt with these reasons were immune from production and irrelevant.

It is common ground that the bank decided to rescue Burmah. There is no issue as to why it did so. The only issues of fact are whether the bank's insistence on buying Burmah's B.P. stock units below the market price and with no profit sharing scheme of any kind was "unconscionable, inequitable and unreasonable" and whether this transaction was "procured by the bank acting in breach of its duty of fair dealing and taking an unfair and unconscionable advantage of Burmah."

When the decision was taken that Burmah should be saved, it must have been obvious to the bank and to the government that this rescue operation might involve the expenditure of a great deal of money. In these circumstances, I find it difficult to imagine how £179,000,000 would have been paid to Burmah for their B.P. stock units unless it had been confidently expected that this stock might well rise to a price yielding a surplus well above £179,000,000 which would more than cover the expense of the rescue operation. In other words, apparently the bank considered that the stock might well rise to a value which would leave a substantial residue after debiting it with the cost of the rescue operation. This residue is probably what, in all the circumstances, the bank thought it would be fair to share with Burmah, in proportions to be discussed, and strongly advised the government to accept this profit sharing as fair. This, I think, is also the kind of profit sharing which my noble and learned friend Lord Wilberforce considered might be shown to be fair, objectively, by the documents already released for inspection.

What is not fair is, I suppose, unfair; and there may be a very fine line or a chasm dividing the fair from the unfair. If the bank took what they believed to be an unfair advantage of Burmah, this might point to the chasm rather than to the fine line. If the chasm exists, this could point to unconscionable conduct by the bank for which it would be responsible, even although it may have been ordered so to behave by the government, and was incapable of disobeying such an order. Whether unconscionable conduct of this kind would carry any legal liability can be decided only at the trial of the action. Whether the bank's conduct was unconscionable may well, in the long run, depend upon objective tests alone. If, however, during the bank's discussions with the government immediately prior to the conclusion of the agreement for the purchase of the B.P. stock, the bank had said anything to suggest that in its view the terms of the purchase insisted on by the government were unconscionable, this would strongly support Burmah's assertion that those terms were indeed unconscionable. The known fact that the bank had already stated, in effect, that the terms were unfair and unreasonable suggests that it may well be that it stated that the terms were also unconscionable. If the bank did so, it would be strong evidence to support Burmah's case on the facts. No one can tell without looking at the documents referring to the discussions between the bank and the government

what was said by the bank at these discussions. I recognise also that these documents may not necessarily record everything that was said.

Nevertheless, in the circumstances to which I have referred, I certainly consider, and I understand that the majority of your Lordships also consider, that 10 out of the 62 documents which the Attorney-General does not wish to be produced should be examined by your Lordships to see whether they may contain evidence "necessary for disposing fairly" of the instant action. These 10 documents cover the period from January 10 to 28, 1975, and are numbered 16, 20, 21, 22, 24, 34, 26, 32, 35 and 36 in the appellants' case. The Attorney-General contends that these documents belong to a class of documents which must be immune from inspection because the immunity is "necessary for the proper functioning of the public service." There are, no doubt, classes of documents which are immune from production because their production would imperil the safety of the state or diplomatic relations, and also classes of documents such as Cabinet minutes and others whose immunity from production is considered necessary for the proper functioning of the public service.

I would agree that the documents disclosed in this case might be included amongst the latter classes of documents to the extent which they related to the reasons which persuaded the government and the bank, in the national interest, to rescue Burmah from the debacle with which it was faced in January 1975. In such circumstances, the lapse of time since 1975 upon which Burmah relies would, in my opinion, in no way affect the immunity.

The instant case, however, as I have already indicated, is in no way concerned with the reasons for the decision to save Burmah. It is solely concerned with the question whether the bank acted unconscionably when it, in effect, made Burmah sell its B.P. stock units upon the terms dictated by the government. If these terms were unconscionable, the fact that the bank was obliged by the government to insist upon them would not, I think, release the bank from responsibility for what it did - nor reflect any credit upon the government.

Had any of the 10 documents which I have mentioned (a) referred to the reasons why Burmah had been saved from ruin in the national interest, and (b) contained evidence that the terms upon which Burmah was obliged to sell its B.P. stock units were unconscionable, then in my opinion the documents should have been produced for inspection but with everything under (a) completely covered up.

I entirely agree that the final decision as to whether or not the national interest made it necessary for Burmah to be saved from ruin was certainly a most important piece of policy making which probably required many delicate and secret factors to be considered. The terms upon which Burmah was to be obliged to sell its B.P. stock units to the bank, however, seem to me to have little to do with policy making in the ordinary sense of that term but only with the making of a hard and perhaps questionable business bargain. A very strong Board of the Privy Council in Robinson v. State of South Australia (No. 2) [1931] A.C. 704, said, at pp. 715-716:

"In view of the increasing extension of state activities into the spheres of trading business and commerce, and of the claim of privilege in relation to liabilities arising therefrom now apparently freely put forward, [Turner L.J.'s observations in Wadeer v. East India Co. (1856) 8 De G.M. & G. 182, 189] stand on record to remind the courts, that while they must duly safeguard genuine public interests they must see to it that the scope of the admitted privilege is not, in such litigation, extended. Particularly must it be remembered in this connection that the fact that production of the documents might in the particular litigation prejudice the Crown's own case or assist that of the other side is no such 'plain overruling principle of public interest' as to justify any claim of privilege. The zealous champion of Crown rights may frequently be tempted to take the opposite view, particularly in cases where the claim against the Crown seems to him to be harsh or unfair. But such an opposite view is without justification. In truth the fact that the documents, if produced, might have any such effect upon the fortunes of the litigation is of itself a compelling reason for their production - one only to be overborne by the gravest considerations of state policy or security."

Robinson's case, as Lord Denning M.R. [1979] 1 W.L.R. 473, 487E points out, was commended in Conway v. Rimmer [1968] A.C. 910 by my noble and learned friends Lord Morris of Borth-y-Gest at p. 970, Lord Hodson at p. 979 and Lord Pearce at p. 983. Robinson's case also laid down that in a case in what used to be called Crown privilege but which is now referred to as public interest immunity, the courts might in a proper case inspect the documents privately and decide whether or not they should be produced. This was also approved in Conway v. Rimmer by my noble and learned friend Lord Pearce where he said, at p. 983:

"The court has always had an inherent power to inspect and order the production of a document or classes of document if in its view the documents, to quote Martin B.'s words [in Beatson v. Skene (1860) 5 H. & N. 838, 854], 'may be made public without prejudice to the public service'."

My Lords, I have privately inspected the 10 documents to which I have referred. In my opinion, none of them throws much, if any, light on what is necessary for fairly disposing of this case; and I would accordingly dismiss the appeal.

LORD EDMUND-DAVIES. My Lords, whatever doubts may or may not assail one, this House is required to deal with this interlocutory appeal on the basis (a) that the plaintiffs have a good cause of action, and (b) that the likelihood or otherwise of their succeeding in their claim is at present irrelevant.

Expressed in the broadest terms, the action, brought against the Bank of England alone, is for a declaration and ancillary relief in respect of an agreement entered into by the parties on January 23, 1975. Its most important term for present purpose was that Burmah undertook to sell

to the bank its largest single asset, consisting of 77,817,507 ordinary stock units of £1 each of the British Petroleum Co. Ltd. ("B.P. stock") for £178,980,266, this representing a price of £2.30 for each unit. Burmah seeks to set aside the agreement on several grounds, the chief being that it was unconscionable, inequitable and unreasonable.

The appeal relates solely to the discovery of documents. It is common ground that the government played an active part in the negotiations leading up to and culminating in the agreement, and the documents held by the bank naturally include a number which reveal something of the important role played by the government. The existence of such documents has been disclosed in a long list prepared by the bank, but on the instructions of the government they have resisted production of 62 of their number. In his certificate the Chief Secretary to the Treasury says that he has read and carefully considered all of them and has formed the opinion that their production would be injurious to the public interest. He divides them into Categories A, B and C, which he describes in the following terms:

"Category A

"These consist of communications between, to and from ministers (including ministers' personal secretaries acting on behalf of ministers) and minutes and briefs for ministers and memoranda of meetings attended by ministers. All such documents relate to the formulation of the policy of the government ..." The minister thereafter sets out various aspects of government policy in relation to the financial difficulties of Burmah.

"Category B

"These consist of communications between, to and from senior officials of the Department of Energy, of the Treasury, and of the bank including memoranda of meetings of and discussions between such officials, and drafts prepared by such officials (including drafts of minutes and briefs comprised in category A), all such communications and drafts relating to the formulation of one or more aspects of the policy described in category A."

"Category C

"These consist of memoranda of telephone conversations and meetings between senior representatives of major companies and other businessmen on the one hand and a minister or senior officials of government departments and of the bank on the other and memoranda of meetings of such officials and briefs for ministers and drafts of such briefs, all recording or otherwise referring to commercial or financial information communicated in confidence by such company representatives and businessmen."

In explanation of the government's opposition to the disclosure of documents falling within these categories, the Chief Secretary stated:

"6. It is, in my opinion, necessary for the proper functioning of the public service that the documents in category A and category B

should be withheld from production. They are all documents falling within the class of documents relating to the formulation of government policy. Such policy was decided at a very high level, involving as it did matters of major economic importance to the United Kingdom ..."

"7. It is, further, in my opinion, necessary for the proper functioning of the public service that the documents in category C should be withheld from production. All the documents in this category record or otherwise refer to commercial or financial information communicated by businessmen outside government (including senior officers of other oil companies) in confidence to ministers or senior officials in government departments or to the bank in its capacity as adviser to the government ... If the documents in this category were produced, those supplying the information could be seriously embarrassed. In my opinion, once it was known that what was imparted in confidence might be revealed publicly there would be a grave danger that such information would cease to be as readily forthcoming as it now is. I have no doubt that this would be detrimental to the public interest."

My Lords, let me say as once that, in the light particularly of the decision of this House in D. v. National Society for the Prevention of Cruelty to Children [1978] A.C. 171, I am not satisfied that the claim made in respect of category C documents could be maintained. But, having regard to the conclusion I have arrived at regarding the outcome of this interlocutory appeal, the validity of that particular claim need not now be discussed.

The wording of the claim advanced has obviously been chosen in the light of certain observations of this House in Conway v. Rimmer [1968] A.C. 910. It is a "class" claim, in contradistinction to a "contents" claim. In other words, the Chief Secretary avers that all 62 documents belong to a class of documents which by their very nature ought to be withheld, and this regardless of whether there is any thing in the contents of any or all of them the disclosure of which would be against public interest. For a class claim may legitimately be advanced even in respect of documents having no contents which it would prejudice the public interest to disclose. A "contents" claim, on the other hand, is self-explanatory, and in Conway v. Rimmer [1968] A.C. 910, Lord Reid said, at p. 943:

"It does not appear that any serious difficulties have arisen or are likely to arise with regard to [a 'contents' claim]. However wide the power of the court may be held to be, cases would be very rare in which it could be proper to question the view of the responsible minister that it would be contrary to the public interest to make public the contents of a particular document."

Even then, however, as the noble and learned Lord added, the question might arise of selecting for disclosure parts of a document the subject of a "contents" claim.

A party to litigation who seeks, as here, to withhold from disclosure

to the other party documents which, being included in their list or affidavit of documents, are ex concessis relevant to the litigation has, as this House made clear in Conway v. Rimmer and Reg. v. Lewes Justices, Ex parte Secretary of State for the Home Department [1973] A.C. 388, per Lord Reid at p. 400H, a heavy burden of proof. But it is not contended by the plaintiffs that the Chief Secretary has failed to establish a good prima facie case for withholding all the listed documents. Indeed, it does not stop there, for Burmah (which nevertheless seeks to inspect 18 of them) accepts that it is now for them in their turn to establish that those 18 documents are "very likely to contain evidence which is highly material" to the issues arising in the case. And they must go further, for the court has no power to order disclosure unless it is "of opinion that the order is necessary either for disposing fairly of the cause or matter or for saving costs": see R.S.C., Ord. 24, r. 13 (1). By no means all relevant documents satisfy such a test.

Indeed, many documents may be of merely vestigial importance, and therefore not the proper subject matter of an order for disclosure. But it is by no means unknown in litigation - and perhaps understandable - to ask for more documents than one hopes to get. And that, in my view, has happened in this case. It is not necessary for me to relate the events leading up to the "December agreement" arrived at between the parties by Burmah's letter to the bank dated January 3, 1975, accepting certain "outline proposals" advanced by the latter on December 31. It should be sufficient to state that the agreement effected an equitable charge to the bank of Burmah's B.P. stock, that it provided that in the event of a sale of such stock Burmah would be consulted beforehand, and that the proceeds of sale would have benefited Burmah by being applied in full in discharge of its liability to the bank. But, for the purposes of this interlocutory appeal, based upon the alleged unconscionability not of the December agreement but of the later agreement concluded on January 23, 1975, I should have thought that no question of disclosure could properly arise in relation to any document earlier than about January 10, 1975, and I proceed to deal with the appeal on that basis.

Summarising the later events, Burmah's case is that on January 10 the bank expressed the view that the only solution of their financial problems was for them to sell B.P. stock to the bank at the current quoted price and on terms that Burmah would share in any profit enuring on the bank's re-sale of the B.P. stock. And, as the appeal proceeded before this House, it became Burmah's case that not until January 15 did it learn from the bank that the government was opposed to any sharing of profit made on the re-sale, and that not until January 22 did the bank inform Burmah that: (a) H.M. Government had decided that the price to be paid by the bank for the B.P. stock was a mere £2.30 per ordinary stock unit; (b) H.M. Government was not prepared to accept any profit-sharing formula; (c) Subject to (a) and (b), H.M. Government agreed to the bank buying B.P.'s Burmah stock. Burmah asserts that, despite its protests against the unconscionability of these terms, they were in due course obliged to accept them, as by that time they had no other course open to them. In their defence the bank expressly admit (a) that on January 22 their Governor and Deputy Governor told Burmah that

they remained of the view that profit-sharing was reasonable and that they had done their best to represent this view to the government, and further (b) that the quoted price of B.P. ordinary stock units had been rising since January 14. But they deny the unconscionability which is at the root of the plaintiffs' claim, and the object of the discovery now sought from them is to destroy that denial.

My Lords, I have already indicated that the plaintiffs have asked for too much. But it was urged by the Attorney-General that, so expansive were the numerous admissions made in the bank's defence and so liberal had the bank been in supplying documents that Burmah had all the material necessary for the presentation of their case. I do not think that is right. In the face of the bank's umbrella denial of any inequality of bargaining power, the sale of B.P. stock at an undervalue, and all other forms of unconscionable conduct on their part, it could, as I think, prove a valuable reinforcement of Burmah's case if they could establish by means of some of the withheld documents that the bank had itself committed themselves to the view that the terms finally presented to Burmah were tainted by those unconscionable features of which Burmah complained.

What are the probabilities of such documentary support being in existence? Is it merely pure conjecture? If so, applying the plaintiffs' own test, production should be refused. But in my judgment, there is more to it than that. It is, at the very least, "on the cards" that, in the light of the bank's known support and advocacy of profit-sharing, they expressed their unequivocal dislike when the government expressed determination to impose its final terms upon Burmah. It was, I think, an over-simplification for the Attorney-General to submit that the only issue is whether the January agreement was in fact inequitable, and not whether the bank regarded it as inequitable. For if, faced by government obduracy despite its strong representations, the bank insisted upon the proposed contractual terms, an arguable foundation for the appellants' allegations of unconscionability against the bank itself could be laid. Then is all this merely "on the cards," simply a "fishing expedition"? If that is all there is to it, discovery should be refused. But in my judgment the existence of such documentary material is likely. And that, in my judgment, is sufficient. For although, as the noble and learned Lord, Lord Wilberforce, has pointed out, it was known in Conway v. Rimmer [1968] A.C. 910 that there were in existence probationary reports on the plaintiff, positive knowledge of that sort is not, in my view, a sine qua non before discovery may be ordered. Nevertheless, as I have already indicated, I think it is very unlikely to have come into existence before January 10, 1975, and (if it exists at all) it will probably be found in the 10 documents numbered 16, 20, 21, 22, 24, 34, 26, 32, 35 and 36 in the appellants' case.

And so, as I see it, the position is reached that, on the one hand, the appellants seek disclosure of 10 documents which may well contain material "necessary ... for disposing fairly of the cause or matter or for saving costs," while, on the other hand, the Attorney-General by his intervention asserts that the withholding of these 10 documents (2 in category A and 8 in category B) is "necessary for the proper functioning

of the public service." In these circumstances, the balancing exercise with which the courts of this country have become increasingly familiar since Conway v. Rimmer [1968] A.C. 910 is called for, and if the appellants are to succeed the scales must come down decisively in their favour: see Alfred Crompton Amusement Machines Ltd. v. Customs and Excise Commissioners (No. 2) [1974] A.C. 405, per Lord Cross of Chelsea at p. 434F.

Despite the strong claims advanced by the Chief Secretary, none of the 10 documents belong to those categories (such as Cabinet decisions and papers) hitherto largely regarded as totally immune from production. And even were such a claim made good, it would not necessarily follow that discharge was precluded. For, as Lord Radcliffe said in Glasgow Corporation v. Central Land Board, 1956 S.C. (H.L.) 1, 18-19:

"The power reserved to the court is ... a power to order production even though the public interest is to some extent affected prejudicially. ... The interests of government, for which the minister should speak with full authority, do not exhaust the public interest. Another aspect of that interest is seen in the need that impartial justice should be done in the courts of law, not least between citizen and Crown, and that a litigant who has a case to maintain should not be deprived of the means of its proper presentation by anything less than a weighty public reason. It does not seem to me unreasonable to expect that the court would be better qualified than the minister to measure the importance of such principles in application to the particular case that is before it.

My Lords, it follows, as I think, that the respondents were wrong in submitting that, if the appellants are to succeed in this interlocutory appeal, they must establish that the Chief Secretary's certificate is probably inaccurate. On the contrary, disclosure may well be ordered even though its accuracy is not impugned, for the minister's view is one-sided and may be correct as far as it goes but is yet not to be regarded as decisive of the matter of disclosure. For, as Lord Reid said in Conway v. Rimmer [1968] A.C. 910, 943:

"The minister who withholds production of a 'class' document has no duty to consider the degree of public interest involved in a particular case by frustrating in that way the due administration of justice. If it is in the public interest in his view to withhold documents of that class, then it matters not whether the result of withholding a document is merely to deprive a litigant of some evidence on a minor issue in a case of little importance or, on the other hand, is to make it impossible to do justice at all in a case of the greatest importance."

There is a further feature in this case which it would be pusillanimous to ignore. It consists in the fact that this is not one of those cases where the complete detachment of the party resisting disclosure is beyond doubt. It is true that the government is not a party to these proceedings, but it would be unrealistic to think that the conduct of government's servants and advisers nowise enters into this case. Not only is it the fact that,

whereas the defendants, left to their own devices, would have complied in full with the plaintiffs' request for discovery, but it is also clear that its only opponent (through the intervention of the Attorney-General) is the government, whose own role must inevitably and inescapably be scrutinised and may be subjected to criticism. Accordingly, since not only justice itself but also the appearance of justice is of considerable importance, the balancing exercise is bound to be affected to some degree where the party objecting to discovery is not a wholly detached observer of events in which it was in no way involved. It cannot realistically be thought that the government is wholly devoid of interest in the outcome of these proceedings. On the contrary, it has a very real and lively interest, for were the plaintiffs to succeed it could only be on the basis that the bank behaved unconscionably, and the evidence indicates that the bank was acting throughout in accordance with government instructions. This fact renders apposite certain observations of Lord Blanesburgh, in Robinson v. State of South Australia (No. 2); [1931] A.C. 704, 715-716:

"In view of the increasing extension of state activities into the spheres of trading business and commerce, and of the claim of privilege in relation to liabilities arising therefrom now apparently freely put forward, [Turner L.J.'s] observations [in Wadeer v. East India Co., 8 De G.M. & G. 182, 189] stand on record to remind the courts, that while they must duly safeguard genuine public interests they must see to it that the scope of the admitted privilege is not, in such litigation, extended. Particularly must it be remembered in this connection that the fact that the production of the documents might in the particular litigation prejudice the Crown's own case or assist that of the other side is no such 'plain overruling principle of public interest' as to justify any claim of privilege. The zealous champion of Crown rights may frequently be tempted to take the opposite view, particularly in cases where the claim against the Crown seems to him to be harsh or unfair. But such an opposite view is without justification. In truth the fact that the documents, if produced, might have any such effect upon the fortunes of the litigation is of itself a compelling reason for their production - one only to be overborne by the gravest considerations of state policy or security."

My Lords, I should add that, on the state of the available material, I have difficulty in regarding it as conceivably prejudicial to state policy were it revealed by such discovery of documents as I have earlier referred to that the bank themselves regarded as unconscionable the terms dictated by the government before the January agreement could be concluded. That the government might well be considerably embarrassed by the revelation is readily understandable, but that is without relevance in this interlocutory appeal.

Yet, when all is said and done and even accepting that the withheld documents are likely to contain material supportive of the allegation of unconscionability, this House is at present completely in the dark as to the cogency of such material. For example, does it clearly and substantially support the allegation, or only to an insignificant degree? Unless its evidentiary value is clear and cogent, the balancing exercise may well

lead to the conclusion that the public interest would best be served by upholding the Chief Secretary's objection to disclosure. On the other hand, if the material provides strong and striking support of the plaintiffs' claim, the court may conclude that, when this is set against such prejudice to the public interest as is likely to arise were any disclosure made in late 1979 regarding even high-policy commercial negotiations conducted in January 1975, the interests of justice demand that disclosure (complete or partial) should be ordered. A judge conducting the balancing exercise needs to know (see per Lord Pearce in Conway v. Rimmer [1968] A.C. 910, 987):

"... whether the documents in question are of much or little weight in the litigation, whether their absence will result in a complete or partial denial of justice to one or other of the parties or perhaps to both, and what is the importance of the particular litigation to the parties and the public. All these are matters which should be considered if the court is to decide where the public interest lies."

A judge may well feel that he cannot profitably embark on such a balancing exercise without himself seeing the disputed documents. May he take a peep? In Conway v. Rimmer, Lord Reid said, at p. 953:

"It appears to me that, if the minister's reasons are such that a judge can properly weigh them, he must, on the other hand, consider what is the probable importance in the case before him of the documents or other evidence sought to be withheld. If he decides that on balance the documents probably ought to be produced, I think that it would generally be best that he should see them before ordering production ... I can see nothing wrong in the judge seeing documents without their being shown to the parties ... If on reading the document he ... thinks that it ought to be produced he will order its production. But it is important that the minister should have a right to appeal before the document is produced."

Lord Upjohn said, at p. 995:

"... if privilege is claimed for a document upon the ground of 'class,' the judge, if he feels any doubt about the reason for its inclusion as a class document, should not hesitate to call for its production for his private inspection, and to order its production if he thinks fit."

But it has been suggested that the position is otherwise where the "class" claim is not challenged. I see no reason why this should be so, once it is postulated that the withheld "class" documents are "likely" to contain material substantially useful to the party seeking discovery. That qualification is necessary, for what is no more than a "fishing expedition" ought not to be advanced by the judge's having a peep to see whether they contain an attractive catch. But, provided such reservation is rigidly adhered to, a judicial peep seems to be justifiable in both cases and may, indeed, prove vital in each if the judge is to be enabled to arrive at a just conclusion in the matter of discovery. Indeed, though not so intended, the further observations of Lord Upjohn in Conway v. Rimmer

[1968] A.C. 910 seem to be entirely apposite to both cases. He said, at pp. 995-996:

"There is a lis between A & B; the Crown may be A or B or, as in this case, a third party ... But when the judge demands to see the documents for which privilege is claimed he is not considering that lis but quite a different lis, that is whether the public interest in withholding the document outweighs the public interest that all relevant documents not otherwise privileged should be disclosed in litigation. The judge's duty is to decide that lis; if he decides it in favour of disclosure, cadit quaestio; if he decides it in favour of non-disclosure he banishes its contents from his mind for the purposes of the main lis."

In my judgment, such material as is presently available leads me to the conclusion that this Appellate Committee of your Lordships' House should now privately inspect the 10 documents earlier referred to. What they contain remains to be seen. All three members of the Court of Appeal themselves inspected certain documents and expressed widely varying conclusions as to their materiality, but your Lordships have no knowledge of which of the 62 listed documents they inspected, and, even if you had, it is still for your Lordships to form your own views regarding the 10 documents earlier specified. If the conclusion is that they contain nothing of any significance, this appeal should be dismissed. But if, on the other hand, material evidence comes to light, your Lordships will be called upon to adjudicate whether, in balancing the competing public interests, disclosure of any of the 10 documents or any parts of them should be ordered. The final disposal of this appeal must await and will depend upon the nature of that adjudication.

My Lords, it was some days after I had completed and passed for typing the foregoing observations that I privately inspected the 10 documents earlier referred to. Having done so, in my judgment disclosure of none of them can be described as "necessary either for disposing fairly of the cause or matter or for saving costs." It follows that I would uphold in its entirety the objection to production advanced by the Chief Secretary to the Treasury and dismiss this interlocutory appeal.

LORD KEITH OF KINKEL. My Lords, this appeal is concerned with the legal topic known formerly as Crown privilege and now as public interest immunity. The topic embraces those rules of law which are applicable for the purpose of determining whether a litigant is entitled to the discovery of documents claimed to be relevant to his case, the production of which is resisted by the Crown on the ground that it would be detrimental to the public interest, that ground being vouched by a ministerial certificate.

The matter was last considered comprehensively by this House in Conway v. Rimmer [1968] A.C. 910. It was there held, contrary to certain dicta in Duncan v. Cammell, Laird & Co. Ltd. [1942] A.C. 624, that a minister's certificate was not conclusive, but that it was a proper function of the court to weigh against each other the aspect of public

interest appearing from the reasons stated in the certificate and the public interest that the administration of justice should not be frustrated. This result was regarded as being in line with the law of Scotland as affirmed in Glasgow Corporation v. Central Land Board, 1956 S.C. (H.L.) 1, and also with that of most Commonwealth countries on the lines expounded in Robinson v. State of South Australia (No. 2) [1931] A.C. 704 and that of the United States as appearing from United State. v. Reynolds (1953) 345 U.S. 1.

Having held that the minister's certificate was not conclusive and laid down the principle that the decision for or against discovery was to be reached by the balancing process described, the House had to go on to consider whether discovery was appropriate in the circumstances of the case before it. The documents sought to be produced were reports of a routine character relating to the plaintiff during his period of probation as a police constable, and by their nature were of extreme importance to the proper disposal of his action for malicious prosecution against his former superintendent. The House had no difficulty in deciding that prima facie they ought to be produced, but thought it proper, before so ordering, to inspect the documents in order to make sure that there was nothing in their contents the disclosure of which would prejudice the public interest. For this purpose it was necessary to enter to some extent upon a general consideration of the type of circumstances under which discovery might correctly be ordered despite Crown objection backed by a ministerial certificate. But in my opinion no definitive body of binding rules universally applicable to future cases in the field is to be gathered from the speeches delivered, and the sound development of the law now requires that it be examined afresh. As was said by Lord Upjohn [1968] A.C. 910, 991-992:

"... in this field the courts are entitled from time to time to make a re-appraisal in relation to particular documents of just what it is that the public interest demands in shielding them from production."

It is convenient to start with the points of distinction between what are commonly called "class" and "contents" claims to immunity. In Conway v. Rimmer [1968] A.C. 910, Lord Reid said, at pp. 943-944:

"A minister's certificate may be given on one or other of two grounds: either because it would be against the public interest to disclose the contents of the particular document or documents in question, or because the document belongs to a class of documents which ought to be withheld, whether or not there is anything in the particular document in question disclosure of which would be against the public interest. It does not appear that any serious difficulties have arisen or are likely to arise with regard to the first class. However wide the power of the court may be held to be, cases would be very rare in which it could be proper to question the view of the responsible minister that it would be contrary to the public interest to make public the contents of a particular document. A question might arise whether it would be possible to separate those parts of a document of which disclosure would be innocuous from those parts which ought not to be made public,

but I need not pursue that question now. In the present case your Lordships are directly concerned with the second class of documents."

Lord Hodson in the same case at p. 979 said that he did not regard the classification which places all documents under the heading either of contents or class as being wholly satisfactory. I agree with him. What really matters is the specific ground of public interest upon which the ministerial objection is based, and it scarcely needs to be said that the more clearly this ground is stated the easier will be the task of the court in weighing it against the public interest in the administration of justice. The weight of a contents claim is capable of being very readily measured. Obvious instances are documents relating to defence of the realm or relations with other states. It might be said that such documents constitute a class defined by reference to the nature of their contents. But I would prefer to regard the claim in regard to such a document as being in substance a contents claim, in relation to which Lord Upjohn said in Conway v. Rimmer, at p. 993:

"A claim made by a minister on the basis that the disclosure of the contents would be prejudicial to the public interest must receive the greatest weight; but even here I am of opinion that the minister should go as far as he properly can without prejudicing the public interest in saying why the contents require protection. In such cases it would be rare indeed for the court to overrule the minister but it has the legal power to do so, first inspecting the document itself and then, if he thinks proper to do so, ordering its production."

Claims to immunity on class grounds stand in a different category because the reasons of public interest upon which they are based may appear to some minds debatable or even nebulous. In Duncan v. Cammell, Laird & Co. Ltd. [1942] A.C. 624, Viscount Simon L.C. at p. 642 referred to cases "where the practice of keeping a class of documents secret is necessary for the proper functioning of the public service." These words have been seized on as convenient for inclusion in many a ministerial certificate, including the one under consideration in the present case. But they inevitably stimulate the query "why is the concealment necessary for that purpose?" and unless it is answered there is nothing tangible to put in the balance against the public interest in the proper administration of justice.

Over a considerable period it was maintained, not without success, that the prospect of the disclosure in litigation of correspondence or other communications within government departments would inhibit a desirable degree of candour in the making of such documents, with results detrimental to the proper functioning of the public service. As mentioned by Lord Reid in Conway v. Rimmer [1968] A.C. 910, 945, the fashion for this was set by Lord Lyndhurst L.C. through the reasons, possibly oblique, which he gave for refusing production of communications between the directors of the East India Co. and the Board of Control in Smith v. East India Co. (1841) 1 Ph. 50. This contention must now be treated as having little weight, if any. In Conway v. Rimmer [1968]

A.C. 910, Lord Morris of Borth-y-Gest, at p. 957, referred to it as being of doubtful validity. Lord Hodson, at p. 976, thought it impossible at the present day to justify the doctrine in its widest term. Lord Pearce, at p. 986, considered that a general blanket protection of wide classes led to a complete lack of common sense. Lord Upjohn, at p. 995, expressed himself as finding it difficult to justify the doctrine "when those in other walks of life which give rise to equally important matters of confidence in relation to security and personnel matters as in the public service can claim no such privilege." The notion that any competent and conscientious public servant would be inhibited at all in the candour of his writings by consideration of the off-chance that they might have to be produced in a litigation is in my opinion grotesque. To represent that the possibility of it might significantly impair the public service is even more so. Nowadays the state in multifarious manifestations impinges closely upon the lives and activities of individual citizens. Where this has involved a citizen in litigation with the state or one of its agencies, the candour argument is an utterly insubstantial ground for denying him access to relevant documents. I would add that the candour doctrine stands in a different category from that aspect of public interest which in appropriate circumstances may require that the sources and nature of information confidentially tendered should be withheld from disclosure. Reg. v. Lewes Justices, Ex parte Secretary of State for the Home Department [1973] A.C. 388 and D. v. National Society for the Prevention of Cruelty to Children [1978] A.C. 171 are cases in point on that matter.

I turn to what was clearly regarded in Conway v. Rimmer [1968] A.C. 910 as the really important reason for protecting from disclosure certain categories of documents on a class basis. It was thus expressed by Lord Reid, at p. 952:

"I do not doubt that there are certain classes of documents which ought not to be disclosed whatever their content may be. Virtually everyone agrees that Cabinet minutes and the like ought not to be disclosed until such time as they are only of historical interest. But I do not think that many people would give as the reason that premature disclosure would prevent candour in the Cabinet. To my mind the most important reason is that such disclosure would create or fan ill-informed or captious public or political criticism. The business of government is difficult enough as it is, and no government could contemplate with equanimity the inner workings of the government machine being exposed to the gaze of those ready to criticise without adequate knowledge of the background and perhaps with some axe to grind. And that must, in my view, also apply to all documents concerned with policy making within departments including, it may be, minutes and the like by quite junior officials and correspondence with outside bodies. Further it may be that deliberations about a particular case require protection as much as deliberations about policy. I do not think that it is possible to limit such documents by any definition."

Lord Hodson at p. 973 referred to classes of documents which from their very character ought to be withheld from production, such as

Cabinet minutes, dispatches from ambassadors abroad and minutes of discussions between heads of departments. Lord Pearce at p. 987 said that obviously production would never be considered of fairly wide classes of documents at a high level such as Cabinet correspondence, letters or reports on appointments to office of importance and the like. Lord Upjohn spoke to similar effect at p. 993, saying that the reason for the privilege was that it would be wrong and entirely inimical to the proper functioning of the public service if the public were to learn of these high level communications, however innocent of prejudice to the state the actual contents of any particular document might be, and that this was obvious.

In my opinion, it would be going too far to lay down that no document in any particular one of the categories mentioned should never in any circumstances be ordered to be produced, and indeed I did not understand counsel for the Attorney-General to pitch his submission that high before this House. Something must turn upon the nature of the subject matter, the persons who dealt with it, and the manner in which they did so. In so far as a matter of government policy is concerned, it may be relevant to know the extent to which the policy remains unfulfilled, so that its success might be prejudiced by disclosure of the considerations which led to it. In that context the time element enters into the equation. Details of an affair which is stale and no longer of topical significance might be capable of disclosure without risk of damage to the public interest. The ministerial certificate should offer all practicable assistance on these aspects. But the nature of the litigation and the apparent importance to it of the documents in question may in extreme cases demand production even of the most sensitive communications at the highest level. Such a case might fortunately be unlikely to arise in this country, but in circumstances such as those of Sankey v. Whitlam, 53 A.L.J.R. 11 or Nixon v. United States (1974) 418 U.S. 683 to which reference is made in the speech of my noble and learned friend Lord Scarman, I do not doubt that the principles there expounded would fall to be applied. There can be discerned in modern times a trend towards more open governmental methods than were prevalent in the past. No doubt it is for Parliament and not for courts of law to say how far that trend should go. The courts are, however, concerned with the consideration that it is in the public interest that justice should be done and should be publicly recognised as having been done. This may demand, though no doubt only in a very limited number of cases, that the inner workings of government should be exposed to public gaze, and there may be some who would regard this as likely to lead, not to captious or ill-informed criticism, but to criticism calculated to improve the nature of that working as affecting the individual citizen. I think that considerations of that nature were present in the mind of Lord Denning M.R. when delivering his dissenting judgment in the Court of Appeal in this case, and in my opinion they correctly reflect what the trend of the law should be.

There are cases where consideration of the terms of the ministerial certificate and of the nature of the issues in the case before it as revealed by the pleadings, taken with the description of the documents sought to be recovered, will make it clear to the court that the balance of public interest lies against disclosure. In other cases the position will be the

reverse. But there may be situations where grave doubt arises, and the court feels that it cannot properly decide upon which side the balance falls without privately inspecting the documents. In my opinion the present is such a case. Those of your Lordships who have already spoken have set out comprehensively the circumstances of the case, the nature and the claimed grounds of the cause of action upon which the appellants found and the terms of the ministerial certificate. I need not rehearse these matters. The appellants maintain that the bargain struck between them and the Bank of England on January 23, 1975, whereby inter alia the latter acquired the appellants' holding of B.P. stock at a very low price, was an unconscionable one, impetrated by abuse of the bank's superior bargaining power, and that it should therefore be set aside. The particular aspect of the bargain which the appellants claim to have been unconscionable was that it made no provision for them participating in any gain which might accrue from a future rise in the value of the B.P. stock units. The bank in relation to the transaction acted in substance as an agency of the government and was able to offer only the terms which the government required it to offer. It is common ground that the Governor and Deputy Governor of the bank favoured an arrangement whereby the appellants would participate in any future profit on the B.P. stock units, but that the government were not prepared to agree to this. Naturally there were meetings and discussions about this matter between representatives of the government departments concerned and of the Bank of England. The list of 62 documents discovery of which was originally sought by the appellants is now reduced to 10. These all relate to such meetings and discussions between January 10 and 22, 1975. The Secretary of State for Energy and the Paymaster General were present at one of these meetings. The Deputy Governor of the Bank of England was present at all of them. Certain of them constituted a follow-on of meetings for the earlier part of which representatives of the appellants had been present.

The 10 documents in question are certified by the minister as relating to the formulation of the policy of the government in face of the appellants' financial difficulties in the light of certain considerations which he sets out. Two of them involve ministers, the remainder senior officials of various government departments. There are of course a substantial number of documents among the other 52 which are similarly certified. While it must be accepted that 10 documents have to do with the formulation of policy, I consider it to be a proper inference that they are not concerned solely with that. Having regard to the timing and the context, they must clearly deal to some extent with the application of policy, with the expression by the Deputy Governor of the Bank of England of the bank's views about what would be reasonable terms for any agreement to be entered into with the appellants, and with the instructions ultimately given to the bank by the governmental representatives. It has to be kept in mind that the court need not order disclosure of the whole terms of any particular document. Parts of it relating to any politically sensitive aspects of the formulation of policy may be withheld from disclosure, while disclosing other parts of clear evidential importance in the litigation. Having carefully considered all the circumstances, I have come to the conclusion that a reasonable probability exists of finding the documents in question

to contain a record of the views of the responsible officials of the Bank of England expressed in such terms as to lend substantial support to the contention that the bargain eventually concluded with the appellants was unconscionable. I do not agree that the issue of unconscionability is to be treated entirely objectively. If it were to be proved, for example, that the Deputy Governor of the bank strongly protested that the terms of the bargain were unconscionable but was overborne by the government, that would, in my view, be important evidence in the appellants' favour. There can be no doubt that the court has power to inspect the documents privately. This was clearly laid down in Conway v. Rimmer [1968] A.C. 910. I do not consider that exercise of such power, in cases responsibly regarded by the court as doubtful, can be treated as itself detrimental to the public interest. Indeed, I am of opinion that it is calculated to promote the public interest, by adding to public confidence in the administration of justice. Whatever the merits of the present action, there can be no doubt that the numerous shareholders of the Burmah Oil Co. Ltd. suffered a grievous blow as a result of the transaction which is the subject of it. Whether they might have suffered worse had the transaction not been entered into is neither here nor there. They might not unreasonably feel a sense of grievance were the court not even to inspect the documents privately to operate the balancing exercise which it is the court's duty to carry through. Such considerations were in the mind of Walton J. in Tito v. Waddell (unreported), March 3, 1975, when he asked and obtained the consent of counsel for the Crown to his inspection of certain documents, lest the Banaban islanders might feel they had received less than justice. In my opinion he could and might appropriately have inspected the documents without any such consent. Apprehension has on occasion been expressed lest the power of inspection might be irresponsibly exercised, perhaps by one of the lower courts. As a safeguard against this, an appeal should always be available, as indicated in Conway v. Rimmer [1968] A.C. 910, per Lord Reid at p. 953.

For these reasons I am in agreement with the majority of your Lordships that this is a proper case for the court to require the 10 documents in question to be made available for private inspection. I do not consider that the discretion to order or refuse production of the documents was capable of being exercised soundly and with due regard to principle in the absence of such inspection. Accordingly I see no difficulty in differing from Foster J. and the majority of the Court of Appeal.

Having inspected the documents, I agree with the majority of your Lordships, though with some hesitation, that none of them contains matter of such evidential value as to make an order for their disclosure, in all the circumstances, necessary for disposing fairly of the case.

It follows that I would dismiss the appeal.

LORD SCARMAN. My Lords, at the beginning of his long and careful argument on behalf of the Attorney-General, Mr. Silkin Q.C. reminded the House that this is an interlocutory appeal from a decision of the Court of Appeal refusing (by a majority) to interfere with decisions reached by Foster J. in the exercise of his discretion. Of course, he was correct. But it would be wrong to infer either that the appeal lacks importance

because it is interlocutory or that no question of law arises because upon questions as to discovery of documents the judge exercises his discretion. The truth is that the appeal raises a question of law of great importance. Your Lordships are asked to determine the respective spheres of the executive and the judiciary where the issue is whether documents for which "public interest immunity" is claimed are to be withheld from disclosure in litigation to which they are relevant. More specifically, the House has to decide whether Conway v. Rimmer [1968] A.C. 910 is definitive of the law, i.e., sets limits statute-wise to the power of the court, or is an illustration upon its particular facts of a broader principle of judicial review.

Others of your Lordships have already told the story of Burmah's disaster. Briefly, Burmah was in appalling financial trouble at the end of 1974. The company was on the brink of default in respect of £54,627,900, 8½ per cent. unsecured loan stock. By the trust deed which constituted the stock, Burmah was liable to repay the whole if the trustees were to declare that the undischarged borrowings by Burmah and its United Kingdom wholly owned subsidiaries exceeded the total of their share capital plus capital and revenue reserves. Such a declaration was believed to be imminent. Some informed observers believed that the undischarged debt already exceeded Burmah's capital and reserves. This spectre of insolvency arose from the misfortunes, and perhaps the mistakes, of 1974. In order to extend its North American operations Burmah had in January and July 1974 arranged loan facilities up to $650 million with some 30 banks, mostly overseas. Burmah had drawn some $625 million on those facilities. In the event of default under any of the dollar loan agreements, or of any other indebtedness becoming prematurely due by reason of default (e.g., under the trust deed, to which I have already referred), Burmah must repay all the dollar loans. Burmah was also seriously at risk of default under arrangements it had made for financing the construction or chartering of tankers. The company's two financial advisers, Flemings and Barings, saw no way out of these difficulties for Burmah, unaided. And aid could only come from the market (including the big oil companies and the financial institutions of the City of London) or the British Government. In any negotiation for financial help Burmah had two assets of special importance - "the B.P. stock" and a substantial interest in two North Sea oilfields, "Thistle" and "Ninian." Burmah owned some 20 per cent. of the ordinary stock of B.P.: but at the end of 1974 the market value of this stock had fallen very low - much lower, in fact, than its "asset value." The stock was, however, unencumbered. It was, therefore, a source of confidence to Burmah's unsecured creditors and available to Burmah in any financial negotiation the company might be advised to undertake. The two oilfields were assets of great potential value; but in 1974 required heavy development expenditure.

Clearly much else was at stake besides the loss of shareholders' capital if Burmah were to be forced by insolvency into liquidation. Failure by a large British company to meet its dollar obligations and the disruptive effect of the company's failure upon the exploitation of North Sea oil would be matters of national importance. Burmah's financial advisers, therefore, approached the Bank of England ("the bank"). The bank immediately informed the government.

The litigation which gives rise to the question which the House now has to decide is concerned with the negotiations between Burmah, the bank and the government, which began at the very end of December 1974 and ended on January 23, 1975, when Burmah agreed to sell the B.P. stock to the bank at a price of £2.30 a stock unit. It is Burmah's case that the hank, under instructions from the government, took an unconscionable advantage of Burmah's weakness by insisting on a price which, while reflecting the current market depression, was a gross undervalue, and without offering Burmah, as the Governor and Deputy Governor thought it would have been reasonable to do, a share in any profit on a re-sale of the stock by the bank. All expected the market price to recover: and, indeed, its recovery began in January 1975. By July 1975 the price had risen appreciably: and by July 1978 it was over £8 a unit.

There were two phases of the negotiation. The first ended with the "December agreement" (made on December 31, 1975, and confirmed in writing by Burmah on January 3, 1975). The relevant terms were that the bank would guarantee the dollar loans for 12 months and would cover the repayment of the unsecured loan stock by subscribing in cash at par a new £54 million loan stock; in consideration of this assistance, Burmah was to transfer to the bank by way of security, with full power of sale, the B.P. stock, and agreed to the transfer, whenever required by the government, of 51 per cent. of its interest in the oilfields of the Continental shelf. Burmah also agreed to the appointment of Peat, Marwick, Mitchell & Co. to report on its finances.

One obvious consequence of the December agreement was that Burmah lost the power to dispose of the B.P. stock as it thought fit. The stock was charged to the bank, who had full powers of sale. The power of disposal of this vital asset had, therefore, passed from Burmah to the bank. A not so obvious, but very serious, consequence was that the agreement did nothing to prevent default under either the loan stock trust deed or the dollar loan agreements. The December agreement offered a cure after default: but it was prevention of default which alone could save Burmah. A third consequence was that the confidence of Burmah's unsecured creditors was profoundly shaken by the loss, as they saw it, of the B.P. stock without any compensating benefit being offered to them.

The second phase of the negotiations began on January 10, 1975, by which time Barings had become aware of the dangerous implications of the December agreement for Burmah. Burmah put forward a proposal which would not commit the company to a sale of the B.P. stock. But the bank, reflecting the views of the government and, in effect, acting upon the instructions of the government, rejected it. Cash, which was what Burmah needed to avoid default on its borrowings, was on offer only for a sale of the stock at a price determined by the government; and that price, £2.30 per unit, was very low indeed. Burmah, having lost the right to dispose of the stock without the bank's consent and having no chance of raising money elsewhere in the time available, were in no position to resist. The government, in the revealing words of the Attorney-General's written case, para. 6 (20):

"had decided that the price to be paid for the B.P. stock should be £2.30 per unit, the government was not prepared to accept any profitsharing

sharing formula and subject thereto the government agreed to the bank purchasing the B.P. stock from Burmah."

Burmah pleaded with the bank. The Governor and Deputy Governor were sympathetic: but the government was adamant and the bank was not prepared to commit itself to any terms other than those acceptable to the government. At a meeting on January 22, 1975, the Governor and Deputy Governor said that they remained of the view that the suggestion of a profit-sharing arrangement was reasonable and that they had done what they could to represent this view to the government. Later the same day H.M. Paymaster General told the representatives of Burmah that it was not possible for the government to improve the terms of sale.

On January 23, 1975, Burmah accepted the terms proposed by the government and the bank. The acceptance constituted the "January agreement" which took the place of the December agreement. Burmah now agreed with the bank (so far as relevant) that: (i) the bank would purchase the B.P. stock at £2.30 per unit for a cash sum of £178,980,266.10; (ii) the bank would guarantee Burmah's dollar borrowings up to $650m.; (iii) the bank would provide a stand-by facility of £75m. The B.P. stock was transferred to the bank on the same day. Shortly afterwards Barings wrote a letter to the bank protesting against the terms of sale.

A feature of the negotiations was the participation of the government. The Permanent Secretary to the Department of Energy and senior officials of H.M. Treasury took a prominent part. When at their very end, on January 22, 1975 (Burmah then being within eight days of a catastrophic default on its obligations), Burmah protested at the terms being offered, the Deputy Governor said the bank would not object to Burmah having direct discussions with the Paymaster General who was handling the matter for the government: and later on the same day it was the Paymaster General who in answer to Burmah's representations replied that it was not possible to improve the terms of the sale. The role of the bank is clear: whatever the opinion of the Governor or Deputy Governor as to the fairness of the terms on offer, the bank was not prepared to negotiate terms of sale other than those acceptable to the government. The government's attitude created the pressure (whether reasonable or not); the bank exerted it upon Burmah by refusing to modify the government's terms.

By a writ, issued on October 6, 1976, Burmah claimed against the bank an order that it cause 77,817,507 ordinary stock units of £1 each of the British Petroleum Co. Ltd. (the B.P. stock) to be transferred back to Burmah against repayment of the purchase price of £178,980,266.10. To succeed, Burmah has to make good one or more of the allegations contained in paragraph 53 of its statement of claim. Burmah alleges that the sale of the B.P. stock to the bank:

"(a) was unconscionable, inequitable and unreasonable; (b) was procured by the bank acting in breach of its duty of fair dealing and taking an unfair and unconscionable advantage of Burmah; (c) [by its terms] wrongfully purported to render nugatory Burmah's right to redeem; (d) involved the bank obtaining an improper collateral advantage in connection with the taking of security."

Mr. Hoffmann for Burmah provided a brief but helpful analysis of the cause of action. He described it as one "not yet fully developed in English law." In its narrow formulation his case is based on, or analogous to, the mortgagor's equity of redemption; the bank, having a charge on the B.P. stock, rendered Burmah's right of redemption nugatory, it is said, by forcing Burmah to sell on unreasonable terms. In its broadest formulation his case is one of "economic duress," as developed in the case law of the United States of America. This development is said to spring from equitable principles established and recognised by the English Courts of Chancery, and to have a common law, or Admiralty, analogue in the law of salvage at sea. He relies on United States v. Bethlehem Steel Corporation (1942) 315 U.S. 289 (in particular the concurring opinion of Murphy J. at p. 311 and the discussion of the English and American case law in the dissenting opinion of Frankfurter J. beginning at p. 325).

Despite its boldness, the cause of action is to be assumed for the purpose of this appeal to be one recognised by English law. No application has been made to strike out the statement of claim: nor, in my opinion, could such an application succeed. Indeed, there are indications in the modern case law that economic duress in a commercial setting may well constitute a good cause of action: see Occidental Worldwide Investment Corporation v. Skibs A/S Avanti (The Siboen and The Sibotre) [1976] 1 Lloyd's Rep. 293 (Kerr J.); North Ocean Shipping Co. Ltd. v. Hyundai Construction Co. Ltd. [1979] Q.B. 705 (Mocatta J.) and the Privy Council decision, Pao On v. Lau Yiu Long now reported [1980] A.C. 614 delivered on April 9, 1979.

The bank has pleaded fully to the statement of claim, admitting much but denying specifically the allegations of unfair pressure, unconscionability and abuse of its bargaining power. On December 21, 1977, the bank served on Burmah its list of documents. Full discovery was given of all minutes or notes of meetings at which Burmah was represented. But in Part III of Schedule 1 of the list the bank described and enumerated 62 documents which it objected to produce on the ground that they "belong to classes of documents, the production of which would be injurious to the public interest." Burmah issued a summons for production of the 62 documents, which after considering a certificate of objection given by the Chief Secretary to the Treasury Foster J. dismissed. On January 19, 1979, the Court of Appeal (Bridge and Templeman L.JJ., Lord Denning M.R. dissenting) dismissed Burmah's appeal, but gave leave to appeal. The judge and the Court of Appeal allowed the Attorney-General to intervene. In your Lordships' House the Attorney-General has been made a party to the appeal. In the event the case against production of the documents has been argued at all stages by counsel for the Attorney-General, the bank being content to abide by whatever order is ultimately made.

In his certificate, dated October 18, 1977, the Chief Secretary to the Treasury recognises that the 62 documents, disclosure of which Burmah seeks, relate to the matters in question in this action. He expresses the opinion that their production would be injurious to the public interest. The reason given for his opinion is that it is necessary for the proper functioning of the public service that production of the documents should

be withheld. His objection is what has become known as a "class" objection. The Chief Secretary grounds his opinion not upon the contents but upon the class of the documents. He places them in three classes: categories A, B and C. The documents in categories A and B relate to the formulation of government policy. Category C is a class of documents which refer to commercial or financial information communicated by businessmen in confidence to ministers or senior officials of the government or to the bank in its capacity as advisers to the government. None of the 62 documents, he says, can be called "routine." He makes clear that, if it were sought to give oral evidence of the contents of any of the documents, the government would object.

The appeal, as it turns out, concerns only categories A and B. In its written case Burmah concedes that of the 62 listed documents only 18 are likely to contain evidence highly material to the issues in the action. These 18 documents are in categories A and B, which relate to the formulation of government policy. During argument it became clear that only 10 of the 18 could contain information likely to assist on the critical issue - the bank's conduct of the negotiations of the terms of sale of the B.P. stock in the sensitive areas of price and a possible profitsharing formula. The question for the House is, therefore, whether these 10 documents, which cover discussions and communications between the bank and the government during the second phase of the negotiations, i.e., the negotiation of the January agreement, are to be withheld from production.

Although in the High Court discovery of documents is automatic in most civil litigation, this is no more than a convenient practice ordered and regulated by rules of court: see R.S.C., Ord. 24, and the recent decision of this House in Science Research Council v. Nassé [1980] A.C. 265; [1979] 3 W.L.R. 762. Discovery of documents remains, ultimately, a matter for the discretion of the court. It is a discretion governed by two general rules of law. The first is that discovery is not to be ordered unless necessary for fairly disposing of the case or for saving costs: see R.S.C., Ord. 24.

The second is that only documents in a party's possession and control which relate to the matters in issue are required to be disclosed, but all such documents, subject to certain exceptions, are to be disclosed, whether or not admissible in evidence. Public interest immunity is, of course, an exception. The case law has given an extended meaning to the qualifying clause "which relate to the matters in issue." It embraces not only documents directly relevant but also documents which may well lead to a relevant train of inquiry: see Compagnie Financière et Commerciale du Pacifique v. Peruvian Guano Co. (1882) 11 Q.B.D. 55. This extended meaning is a vital part of the law of discovery, enabling justice to be done where one party knows the facts and possesses the documents and the other does not.

Foster J. based his decision on the view which he formed that production of the documents for which immunity is claimed would not materially assist the plaintiffs' case at trial. He was. I think, right, when faced with the public interest immunity objection to disclosure, to ask himself whether

production could be said to be necessary for fairly disposing of the case. For, if it be shown that production was not necessary, it becomes unnecessary to balance the interest of justice against the interest of the public service to which the minister refers in his certificate. But the judge formed his view without himself inspecting the documents. It is said - and this view commended itself to the majority of the Court of Appeal - that the bank has given very full discovery of the documents directly relevant to the critical issue in the action, namely, the conduct by the bank of the negotiations with Burmah: that Burmah knows as much about this issue as does the bank: and that it can be fully investigated and decided upon the documents disclosed and the evidence available to Burmah without recourse to documents noting or recording the private discussions between the bank and the government. Upon this view, Burmah's attempt to see these documents is no more than a fishing expedition.

I totally reject this view of the case. First, as a matter of law, the documents for which immunity is claimed relate to the issues in the action and, according to the Peruvian Guano formulation, 11 Q.B.D. 55, may well assist towards a fair disposal of the case. It is unthinkable that in the absence of a public immunity objection and without a judicial inspection of the documents disclosure would have been refused. Secondly, common sense must be allowed to creep into the picture. Burmah's case is not merely that the bank exerted pressure: it is that the bank acted unreasonably, abusing its power and taking an unconscionable advantage of the weakness of Burmah. Upon these questions the withheld documents may be very revealing. This is not "pure speculation." The government was creating the pressure: the bank was exerting it upon the government's instructions. Is a court to assume that such documents will not assist towards an understanding of the nature of the pressure exerted? The assumption seems to me as unreal as the proverbial folly of attempting to understand Hamlet without reference to his position as the Prince of Denmark. I do not understand how a court could properly reach the judge's conclusion without inspecting the documents: and this he refused to do. The judge in my opinion wrongly exercised his discretion when he refused to inspect unless public policy (of which public interest immunity is a manifestation) required him to refuse.

It becomes necessary, therefore, to analyse closely the public interest immunity objection made by the minister and to determine the correct approach of the court to a situation in which there may be a clash of two interests - that of the public service and that of justice.

In Conway v. Rimmer [1968] A.C. 910 this House had to consider two questions. They were formulated by Lord Reid in these terms, at p. 943:

"... first, whether the court is to have any right to question the finality of a minister's certificate and, secondly, if it has such a right, how and in what circumstances that right is to be exercised and made effective."

The House answered the first question, but did not, in my judgment, provide, nor was it required to provide, a complete answer to the second.

As I read the speeches in Conway v. Rimmer the House answered the first question by establishing the principle of judicial review. The minister's certificate is not final. The immunity is a rule of law: its scope is a question of law: and its applicability to the facts of a particular case is for the court, not the minister, to determine. The statement of Lord Kilmuir L.C. of June 6, 1956 (all that is relevant is quoted in Conway v. Rimmer at p. 922) that: "The minister's certificate on affidavit setting out the ground of the claim must in England be accepted by the court ..." is no longer a correct statement of the law. Whether Conway v. Rimmer be seen as a development of or a departure from previous English case law is a matter of no importance. What is important is that it aligned English law with the law of Scotland and of the Commonwealth. It is the heir apparent not of Duncan v. Cammell, Laird & Co. Ltd. [1942] A.C. 624 but of Robinson v. State of South Australia (No. 2) [1931] A.C. 704 and of Glasgow Corporation v. Central Land Board, 1956 S.C. (H.L.) 1.

Having established the principle of judicial review, the House had in Conway v. Rimmer [1968] A.C. 910 a simple case on the facts to decide. The question was whether routine reports, albeit of a confidential character, upon a former probationary police constable should in the interests of justice be disclosed in an action brought by him against his former superintendent in which he claimed damages for alleged malicious prosecution. There was a public interest in the confidentiality of such reports, but the Home Secretary, in his affidavit objecting to production on the ground of injury to the public interest, did not go so far as to say that it was necessary for the proper functioning of the public service to withhold production. On the other hand, the reports might be of critical importance in the litigation. Granted the existence of judicial review, here was a justiciable issue of no great difficulty. The House decided itself to inspect the documents, and, having done so, ordered production.

In reaching its decision the House did indicate what it considered to be the correct approach to the clash of interests which arises whenever there is a question of public interest immunity. The approach is to be found stated in two passages of Lord Reid's speech: pp. 940C-F and 952C-G. The essence of the matter is a weighing, on balance, of the two public interests, that of the nation or the public service in non-disclosure and that of justice in the production of the documents. A good working, but not logically perfect, distinction is recognised between the contents and the classes of documents. If a minister of the Crown asserts that to disclose the contents of a document would, or might, do the nation or the public service a grave injury, the court will be slow to question his opinion or to allow any interest, even that of justice, to prevail over it. Unless there can be shown to exist some factor suggesting either a lack of good faith (which is not likely) or an error of judgment or an error of law on the minister's part, the court should not (the House held) even go so far as itself to inspect the document. In this sense, the minister's assertion may be said to be conclusive. It is, however, for the judge to determine whether the minister's opinion is to be treated as conclusive. I do not understand the House to have denied that even in "contents" cases the court retains its power to inspect or to balance the injury to the public service against the risk of injustice, before reaching its decision.

In "class" cases the House clearly considered the minister's certificate to be more likely to be open to challenge. Undoubtedly, however, the House thought that there were certain classes of documents, which ought not to be disclosed however harmless the disclosure of their contents might be, and however important their disclosure might be in the interest of justice. Cabinet minutes were cited as an example. But the point did not arise for decision. For the documents in Conway v. Rimmer [1968] A.C. 910, though confidential, were "routine," in no way concerned with the inner working of the government at a high level; and their production might well be indispensable to the doing of justice in the litigation.

The point does arise in the present case. The documents are "high level." They are concerned with the formulation of policy. They are part of the inner working of the government machine. They contain information which the court knows does relate to matters in issue in the action, and which may, on inspection, prove to be highly material. In such circumstances the minister may well be right in his view that the public service would be injured by disclosure. But is the court bound by his view that it is necessary for the proper functioning of the public service that they be withheld from production? And, if non-disclosure is necessary for that purpose, is the court bound to hold that the interest in the proper functioning of the public service is to prevail over the requirements of justice?

If the answer to these two questions is to be in the affirmative as Lord Reid appears to suggest in Conway v. Rimmer, I think the law reverts to the statement of Lord Kilmuir. A properly drawn minister's certificate, which is a bona fide expression of his opinion, becomes final. But the advance made in the law by Conway v. Rimmer was that the certificate is not final. I think, therefore, that it would now be inconsistent with principle to hold that the court may not - even in a case like the present - review the certificate and balance the public interest of government to which alone it refers, against the public interest of justice, which is the concern of the court.

I do not therefore accept that there are any classes of document which, however harmless their contents and however strong the requirement of justice, may never be disclosed until they are only of historical interest. In this respect I think there may well be a difference between a "class" objection and a "contents" objection - though the residual power to inspect and to order disclosure must remain in both instances. A Cabinet minute, it is said, must be withheld from production. Documents relating to the formulation of policy at a high level are also to be withheld. But is the secrecy of the "inner workings of the government machine" so vital a public interest that it must prevail over even the most imperative demands of justice? If the contents of a document concern the national safety, affect diplomatic relations or relate to some state secret of high importance, I can understand an affirmative answer. But if they do not (and it is not claimed in this case that they do), what is so important about secret government that it must be protected even at the price of injustice in our courts?

The reasons given for protecting the secrecy of government at the level of policy-making are two. The first is the need for candour in the advice offered to ministers: the second is that disclosure "would create or fan ill-informed or captious public or political criticism." Lord Reid in Conway v. Rimmer [1968] A.C. 910, 952, thought the second "the most important reason." Indeed, he was inclined to discount the candour argument.

I think both reasons are factors legitimately to be put into the balance which has to be struck between the public interest in the proper functioning of the public service (i.e., the executive arm of government) and the public interest in the administration of justice. Sometimes the public service reasons will be decisive of the issue: but they should never prevent the court from weighing them against the injury which would be suffered in the administration of justice if the document was not to be disclosed. And the likely injury to the cause of justice must also be assessed and weighed. Its weight will vary according to the nature of the proceedings in which disclosure is sought, the relevance of the documents, and the degree of likelihood that the document will be of importance in the litigation. In striking the balance, the court may always, if it thinks it necessary, itself inspect the documents.

Inspection by the court is, I accept, a power to be exercised only if the court is in doubt, after considering the certificate, the issues in the case and the relevance of the documents whose disclosure is sought. Where documents are relevant (as in this case they are), I would think a pure "class" objection would by itself seldom quieten judicial doubts - particularly if, as here, a substantial case can be made out for saying that disclosure is needed in the interest of justice.

I am fortified in the opinion which I have expressed by the trend towards inspection and disclosure to be found both in the United States and in Commonwealth countries. Of course, the United States have a written constitution and a Bill of Rights. Nevertheless both derive from the common law and British political philosophy. Mutatis mutandis, I would adopt the principle accepted by the Supreme Court in Nixon v. United States, 418 U.S. 683 which is summarised in 41 L.Ed. 2d 1039, 1046:

"Neither the doctrine of separation of powers, nor the need for confidentiality of high level communications, without more, can sustain an absolute unqualified presidential privilege of immunity from judicial process under all circumstances; although the President's need for complete candor and objectivity from advisers calls for great deference from the courts, nevertheless when the privilege depends solely on the broad, undifferentiated claim of public interest in the confidentiality of such conversations, a confrontation with other values arises; absent a claim of need to protect military, diplomatic or sensitive national security secrets, it is difficult to accept the argument that even the very important interest in confidentiality of Presidential communications is significantly diminished by production of such material for in camera inspection with all the protection that a United States District Court will be obliged to provide."

In Australia the High Court had to consider the problem in a recent ease where the facts were, admittedly, exceptional. In Sankey v. Whitlam, 53 A.L.J.R. 11 the plaintiff sought declarations that certain papers and documents, to which the magistrate in criminal proceedings instituted by the plaintiff against the defendants had accorded privilege, should be produced. The offences alleged against Mr. Whitlam, a former Prime Minister, and others were serious - conspiracies to act unlawfully in the conduct of official business. Gibbs A.C.J. dealt with the issue of Crown privilege as follows, at p. 23:

"For these reasons I consider that although there is a class of documents whose members are entitled to protection from disclosure irrespective of their contents, the protection is not absolute, and it does not endure for ever. The fundamental and governing principle is that documents in the class may be withheld from production only when this is necessary in the public interest. In a particular case the court must balance the general desirability that documents of that kind should not be disclosed against the need to produce them in the interests of justice. The court will of course examine the question with especial care, giving full weight to the reasons for preserving the secrecy of documents of this class, but it will not treat all such documents as entitled to the same measure of protection - the extent of protection required will depend to some extent on the general subject matter with which the documents are concerned. If a strong case has been made out for the production of the documents, and the court concludes that their disclosure would not really be detrimental to the public interest, an order for production will be made. In view of the danger to which the indiscriminate disclosure of documents of this class might give rise, it is desirable that the government concerned, Commonwealth or State, should have an opportunity to intervene and be heard before any order for disclosure is made. Moreover no such order should be enforced until the government concerned has had an opportunity to appeal against it, or test its correctness by some other process, if it wishes to do so (cf. Conway v. Rimmer [1968] A.C. 910, 953)."

Both Nixon's case, 418 U.S. 683 and Sankey v. Whitlam, 53 A.L.J.R. ll are far closer to the Scottish and Commonwealth stream of authority than to the English. In the Glasgow Corporation case, 1956 S.C. (H.L.) 1, Viscount Simonds said, at p. 11:

"that there always has been and is now in the law of Scotland an inherent power of the court to override the Crown's objection to produce documents on the ground that it would injure the public interest to do so."

In Robinson v. State of South Australia (No. 2) [1931] A.C. 704 the Privy Council reminded the Supreme Court of South Australia of the existence of this power. The power must be exercised judicially, and all due weight must be given to the objections of the Crown: that is all.

Something was made in argument about the risk to the nation or the public service of an error at first instance. Injury to the public interest - perhaps

perhaps even very serious injury - could be done by production of documents which should be immune from disclosure before an appellate court could correct the error. This risk is inherent in the principle of judicial review. The House in Conway v. Rimmer [1968] A.C. 910 recognised its existence, but, nevertheless, established the principle as part of our law. Gibbs J. also mentioned it in Sankey v. Whitlam, 53 A.L.J.R. 11. I would respectfully agree with Lord Reid's observations on the point in Conway v. Rimmer [1968] A.C. 910, 953D: "... it is important that the minister should have a right to appeal before the document is produced."

In cases where the Crown is not a party - as in the present case - the court should ensure that the Attorney-General has the opportunity to intervene before disclosure is ordered.

For these reasons I was one of a majority of your Lordships who thought it necessary to inspect the 10 documents. Having done so, I have no doubt that they are relevant and, but for the immunity claim, would have to be disclosed, but their significance is not such as to override the public service objections to their production. Burmah will not suffer injustice by their non-disclosure, while their disclosure would be, in the opinion of the responsible minister, injurious to the public service. I would, therefore, dismiss the appeal.

By way of tail-piece I mention the strange affair of the edited documents. The bank, claiming immunity for part, but not the whole, of certain documents, covered up the parts to the disclosure of which it objected. Burmah's advisers were able to penetrate the cover and read their contents. They did not tell their client what they had seen. Should they now be disclosed, the cover having been blown? The issue evaporated because it became clear in argument that Burmah were ultimately fighting to see only the 10 documents, which a majority of your Lordships has now inspected. But the accident of an insufficient cover cannot weaken the objection of public interest immunity. Even if the parties allow discovery, the court must take the objection of its own motion: and this may have to be done even before the Crown intervenes. There was a difference of opinion as to the importance of the covered up parts. But in view of the course taken by the parties in argument in this House the question does not arise. Burmah's advisers acted with propriety in the handling of the incident.

Solicitors: Allen & Overy; Freshfields; Treasury Solicitor.

J. A. G.