D v National Society for the Prevention of Cruelty to Children (NSPCC) [1977] UKHL 1

House of Lords

Between

DAppellant

National Society for the Prevention of Cruelty to Children (NSPCC)Respondent

Before

Lord Diplock

Lord Hailsham of St

Lord Simon of Glaisdale

Lord Kilbrandon

Lord Edmund-Davies

LORD DIPLOCK

My Lords,

In form this is an interlocutory appeal upon a summons relating to the discovery of documents by the National Society for the Prevention of Cruelty to Children (N.S.P.C.C.) in a pending action brought against them by the respondent (Mrs. D.). In substance, the question for your Lordships is whether the N.S.P.C.C. can be compelled in legal proceedings to disclose the identity of persons who give them information that a child is being ill-treated or neglected.

The N.S.P.C.C. is a voluntary society founded in 1889 and incorporated by Royal Charter in 1895. Its purposes are: —

To prevent the public and private wrongs of children, and the corruption of their morals ;

To take action for the enforcement of laws for their protection;

To provide and maintain an organisation for the above objects ;

To do all other such lawful things as are incidental or conducive to the attainment of the above objects.

Action for the enforcement of laws for the protection of children may take two forms:

The prosecution of offenders for criminal offences against children, and

The institution of care proceedings in a juvenile court under section 1 of the Children and Young Persons Act, 1969.

That section authorises a juvenile court to make a variety of different Orders for the protection of children or young persons who are being ill- treated or neglected. By subsection (1) care proceedings may only be brought by a local authority, a constable or an " authorised person ". By the Children and Young Persons Act 1969 (Authorisation for the purposes of section D Order 1970, the N.S.P.C.C. is the only authorised person under the Act. It had also been an authorised person for the purposes of taking proceedings under the corresponding section, section 62 of the previous statute of the Children and Young Persons Act 1933.

Although empowered to bring care proceedings, neither constables nor the N.S.P.C.C. are under any statutory duty to do so. In this respect their position differs from that of a local authority. By section 2(2) where it appears to a local authority that there are grounds for bringing care proceedings in respect of a child who resides or is found in their area it is the duty of that authority to bring such proceedings unless they are satisfied that it is neither in the child's interest or the public interest to do so, or that some other person is about to do so or to charge him with an offence. By section 2(3) constables and the N.S.P.C.C. are required to give notice to the local authority before beginning care proceedings themselves.

The detailed events which gave rise to the action brought by Mrs. D. against the N.S.P.C.C. are vividly recounted in the judgment of the Master of the Rolls as reported at [1976] 3 W.L.R. 124, to which reference may be made. For present purposes it is sufficient to summarise them as follows: —

In the afternoon of 13th December, 1973, somebody told the N.S.P.C.C. that the 14-month old daughter of Mrs. D. had been beaten and ill treated

over the past six weeks. On receipt of this information an inspector of the N.S.P.C.C. called upon Mrs. D. at her home in order to see the condition of the child. The information turned out to be untrue. The child showed no signs of ill-treatment. She was healthy and well-cared for.

Mrs. D. was naturally very upset by this visit and to learn of the false accusation against her. As a result of this her health was affected. She wanted to know the name of the N.S.P.C.C.'s informant; but this was refused. After an unsuccessful attempt under Order 24, Rule 7(a) to obtain discovery of documents from the N.S.P.C.C. before commencing any proceedings, she issued a writ and statement of claim on 19th June, 1974, claiming relief of two different kinds against the N.S.P.C.C.

Damages for failure to exercise reasonable care in investigating the complaint that had been made about her child before repeating it to her; and

An order that the N.S.P.C.C. disclose to her all documents in their custody, possession or power relating to the complaint and the identity of the complainant.

The particulars of negligence included an allegation that the N.S.P.C.C.:

" failed to make or cause to be made any or any proper sufficient in- " quiries of the complainant having regard to the identity and/or status " and/or means of knowledge of the complainant and/or the nature " and/or substance of the complaint in order to check that the complaint " was made bona fide and not maliciously, before visiting the Plaintiff."

At the present stage of the proceedings your Lordships are not concerned with the question whether the statement of claim discloses a good cause of action on the part of the plaintiff against the N.S.P.C.C. for damages for breach of duty or for negligence. That is a matter which will fall to be decided at the trial of the action, if it ever comes to trial. The present appeal must be disposed of on the basis that the facts pleaded in the state- ment of claim, if true, do constitute a good cause of action ; and accord- ingly that any documents which disclose the identity of the person who gave the false information relate to matters in question in the action and are prima facie subject to disclosure under Order 24, Rule 1(1).

On the 26th September, 1974, the N.S.P.C.C. took out a summons under Order 24 (Rule 2(5)) for an order that there should be no discovery by the defendants of documents where such documents reveal or are capable of revealing the identity of the N.S.P.C.C.'s informant.

On llth December, 1974, Master Jacob dismissed this application. He ordered discovery and inspection in the usual form. On 26th June, 1975, Mr. Justice Croom-Johnson allowed an appeal from Master Jacob's order and ordered that there should be no discovery or inspection of documents by the defendants where or to the extent that such documents reveal or are capable of revealing the identity of the N.S.P.C.C.'s informant. From this order Mrs. D., with leave of the judge, appealed to the Court of Appeal. On 6th May, 1975, the Court of Appeal by a majority, Lord Denning M.R. dissenting, allowed the appeal and reinstated the Master's order for discovery.

Before this House the claim of the N.S.P.C.C. to refuse discovery of docu- ments which could reveal the identity of their informant was based squarely upon the public interest in maintaining the confidentiality of information given to the Society so that it may take steps to promote the welfare of a child, whether, as happens in the great majority of cases, by giving support, advice and guidance to the family of which the child is a member or, if this be necessary in the interest of the child, by instituting care proceedings in respect of him or prosecuting those who have committed offences against him.

To assist them to carry out the purposes of their charter and their functions as a person authorised to take care proceedings under S.1. of the Children and Young Persons Act 1969, the N.S.P.C.C. invite the help of the general public in telling the Society's officers of any child of whom they know who

may be suffering because of misfortune, ignorance, neglect or ill-treatment. The leaflets, which the Society distributes widely to enlist the public's aid, contain the promise " Your name and the information you give for the " purpose of helping children will be treated as confidential." The uncontra- dicted evidence of the Director of the N.S.P.C.C. is that the work of the Society is dependent upon its receiving prompt information of suspected child abuse and that, as might be expected, the principal sources of such information are neighbours of the child's family or doctors, school-teachers, health visitors and the like who will continue to be neighbours or to maintain the same relationship with the suspected person after the matter has been investigated and dealt with by the N.S.P.C.C. The evidence of the Director is that without an effective promise of confidentiality neighbours and others would be very hesitant to pass on to the Society information about suspected child abuse. There is an understandable reluctance to " get involved " in something that is likely to arouse the resentment of the person whose suspected neglect or ill-treatment of a child has been reported by the informant, however true the information may be. Unless the N.S.P.C.C. can guarantee the anonymity of its informants, its ability to learn of cases where children are at risk would be drastically reduced.

The fact that information has been communicated by one person to another in confidence, however, is not of itself a sufficient ground for pro- tecting from disclosure in a court of law the nature of the information or the identity of the informant if either of these matters would assist the court to ascertain facts which are revelant to an issue upon which it is adjudicating. (Alfred Crompton Amusement Machines Ltd. v. Customs and Excise Com- missioners (No. 2) [1974] A.C. 405 at pp. 433/4). The private promise of con- fidentiality must yield to the general public interest that in the administration of justice truth will out, unless by reason of the character of the information or the relationship of the recipient of the information to the informant, a more important public interest is served by protecting the information or the identity of the informant from disclosure in a court of law.

The public interest which the N.S.P.C.C. relies upon as obliging it to withhold from the plaintiff and from the court itself material that could disclose the identity of the Society's informant is analogous to the public interest that is protected by the well established rule of law that the identity of police informers may not be disclosed in a civil action, whether by the process of discovery or by oral evidence at the trial. (Marks v. Beyfus [1890] 25 Q.B.D. 494.)

The rationale of the rule as it applies to police informers is plain. If their identity were liable to be disclosed in a court of law, these sources of information would dry up and the police would be hindered in

of preventing and detecting crime. So the public interest in preserving the anonymity of police informers had to be weighed against the public interest that information which might assist a judicial tribunal to ascertain facts relevant to an issue upon which it is required to adjudicate should be withheld from that tribunal. By the uniform practice of the judges which by the time of Marks v. Beyfus had already hardened into a rule of law, the balance has fallen upon the side of non-disclosure except where upon the trial of a defendant for a criminal offence, disclosure of the identity of the informer could help to show that the defendant was innocent of the offence. In that case, and in that case only, the balance falls upon the side of disclosure.

My Lords,

in Reg. v. Lewes Justices [1973] A.C. 388 this House did not hesitate to extend to persons from whom the Gaming Board received information for the purposes of the exercise of their statutory functions under the Gaming Act 1968 immunity from disclosure of their identity analogous to that which the law had previously accorded to police informers. Your Lordships' sense of values might well be open to reproach if this House were to treat the confidentiality of information given to those who are authorised by statute to institute proceedings for the protection of neglected or ill-treated children as entitled to less favourable treatment in a court of law than information given to the Gaming Board so that gaming may be 306983 A 2

kept clean. There are three categories of persons authorised to bring care proceedings in respect of neglected or ill-treated children: local authorities, constables and the N.S.P.C.C. The anonymity of those who tell the police of their suspicions of neglect or ill-treatment of a child would be preserved without any extention of the existing law. To draw a distinction in this respect between information given to the police and that passed on directly to a local authority or to the N.S.P.C.C. would seem much too irrational a consequence to have been within the contemplation of parliament when enacting the Children and Young Persons Act 1969. The local authority is under an express statutory duty to bring care proceedings in cases where this is necessary if neither the police nor the N.S.P.C.C. have started them; while, as respects the N.S.P.C.C., the evidence shows that, presumably because it is not associated in the public mind with officialdom, the public are readier to bring information to it than to the police or the welfare services of the local authority itself.

Upon the summons by the N.S.P.C.C. for an order withholding discovery of documents to the extent that they were capable of revealing the identity of the Society's informant, it was for the judge to weigh the competing public interests involved in disclosure and non-disclosure and to form his opinion as to the side on which the balance fell. In a careful judgment in which he reviewed the relevant authorities Croom-Johnson J. ordered that disclosure should not be given. Upon an interlocutory summons relating to discovery this was a matter upon which the judge had a discretion with which an appellate court would not lightly interfere; but the reasoning by which his decision was supported is of wider application. It would also rule out any attempt to ascertain the identity of the N.S.P.C.C.'s informant by questions put to witnesses at the trial and would dispose of the plaintiff's claim to dis- closure of the informant's identity as part, and perhaps to her the most important part of the substantive relief the plaintiff seeks. The interlocutory judgment thus raises matters of principle fit for the consideration of this House.

For my part I would uphold the decision of Croom-Johnson J. and reverse that of the Court of Appeal. I would do so upon what in argument has been referred to as the " narrow " submission made on behalf of the N.S.P.C.C. I would extend to those who give information about neglect or ill-treatment of children to a local authority or the N.S.P.C.C. a similar immunity from disclosure of their identity in legal proceedings to that which the law accords to police informers. The public interests served by preserving the anonymity of both classes or informants are analogous: they are of no less weight in the case of the former than of the latter class, and in my judgment are of greater weight than in the case of informers of the Gaming Board to whom immunity from disclosure of their identity has recently been extended by this House.

In the Court of Appeal, as in this House, counsel for the N.S.P.C.C. advanced, as well as what I have referred to as the narrow submission, a broad submission that wherever a party to legal proceedings claims that there is a public interest to be served by witholding documents or information from disclosure in those proceedings, it is the duty of the court to weigh that interest against the countervailing public interest in the administration of justice in the particular case and to refuse disclosure if the balance tilts that way. This broad submission, or something rather like it confined to information imparted in confidence, was adopted in his dissenting judgment by the Master of the Rolls, but as I have already indicated there is the authority of this House that confidentiality of itself does not provide a ground of non-disclosure; nor am I able to accept the proposition that the basis of all privilege from disclosure of documents or information in legal proceed- ings is to prevent the breaking of a confidence. For my part, I think this House would be unwise to base its decision in the instant case upon a proposition so much broader than is necessary to resolve the issue between the parties.

The majority of the Court of Appeal rejected both the broad and the narrow submissions. In essence their ground for doing so was that " public interest" as a ground for witholding disclosure of documents or information was but another term for what had before Conway v. Rimmer [1968] A.C.910 been called "Crown privilege" and was available only where the ... public interest involved was the effective functioning of departments or other organs of central government. " Crown privilege " they regarded as having always been so confined; Conway v. Rimmer

[1968] AC 910

did not extend the ambit of Crown privilege: all it did was to decide that a claim by a Minister of the Crown that documents were of a class which in the public interest ought not to be disclosed was not conclusive but that it was for the court itself to decide whether the public interest which would be protected by on-disclosure outweighed the public interest in making available to the court information that might assist it in doing justice between the litigants in the particular case.

This narrow view as to the scope of public interest as a ground for protecting documents and information from disclosure was supported in argument before this House by copious citations of passages taken from judgments in previous cases in the course of which documents for which a claim to non-disclosure had been described as relating to essential functions of government, to the performance of statutory duties, to the public service or to the interests of the State. From this your Lordships were invited to infer that the document in question would not have been entitled to pro- tection from disclosure unless it fell within the description used in the particu- lar case.

My Lords,

the maxim expressio unius, exclusio alterius is not a canon of construction that is applicable to judgments. To construe a judgment as if its function were to lay down a code of law is a common error into which the English reliance upon precedent makes it easy to fall. A cautious judge expresses a proposition of law in terms that are wide enough to cover the issue in the case under consideration; the fact that they are not also wide enough to cover an issue that may arise in some subsequent case does not make his judgment an authority against any wider proposition.

I see no reason and I know of no authority for confining public interest as a ground for non-disclosure of documents or information to the effective functioning of departments or organs of central government. In Conway v. Rimmer the public interest to be protected was the effective functioning of a county police force, in In re D [1970] 1 W.L.R. 599 the interest to be protected was the effective functioning of a local authority in relation to the welfare of boarded-out children. In the instant case the public interest to be protected is the effective functioning of an organisation authorised under an Act of Parliament to bring legal proceedings for the welfare of children. I agree with Croom-Johnson J. that this is a public interest which the court is entitled to take into consideration in deciding when the identity of the N.S.P.C.C.'s informants ought to be disclosed. I also agree that the balance of public interest falls on the side of non-disclosure.

I would allow this appeal.

The Rt. Hon. Lord Hailsham of St. Marylebone

My Lords,

The question for decision in this appeal is the extent to which, if at all, the appellants, that is the National Society for the Prevention of Cruelty to Children, are entitled to refuse to disclose the identity of informants who have supplied information to the Society making allegations of possible child abuse against a third party. The question comes before the House on a summons under 0.24 r.2(5) requesting an order limiting discovery. But in other circumstances it could well have arisen on a summons relating to interrogatories or the sufficiency of an answer to interrogatories, or on a question posed at the trial on behalf of a party to a witness. It is, in short,

primarily a question of the law of evidence rather than discovery, although in different circumstances the result might depend to some extent on the stage which the proceedings had reached, or the type of proceeding on which the issue had been raised.

Both parties have pursued their position as a matter of principle, with pertinacity, and with manifest sincerity, and there are powerful arguments on both sides. In order to appreciate their force it is necessary to rehearse me salient facts.

The story begins at 8.30 p.m. on the night of 13 December 1973. The respondent to this appeal, plaintiff in these proceedings, was alone in her suburban house with a Norwegian au pair girl and her baby. This baby, a girl of fourteen months, was upstairs asleep in her cot. There was a ring at the door, and, when the door was opened there appeared a stranger, a Mr. Jenkins, who identified himself as what he is, an officer of the Society and craved admittance. He was admitted with the somewhat reserved cordiality accorded to those who are expected to solicit alms or support for a well-known charity. But it speedily became apparent that he was present on a more serious, and, it may be thought, a more sinister mission. He was the recipient of a circumstantial complaint against the respondent alleging mal- treatment of the baby girl upstairs. According to what he said, the complaint, which would have to some extent have been investigated before his visit. was that the baby's head and stomach were bruised, that she was curiously immobile for her age, and that she received no stimulation at home. Accord- ing to Mr. Jenkins, the informant claimed to have seen the bruises and the maltreatment of the baby was alleged to have taken place over the previous six weeks, though the information had been in his own possession only from 4;30 p.m. on the same day. Mr. Jenkins said that there could be no mistake in the identity of the respondent or her house.

All this information, it must now be assumed for the purpose of this appeal, except of course the identity of the respondent, was entirely without founda- tion. The baby was brought downstairs, undressed, and showed no marks. At the respondent's insistence, the family doctor, a Dr. Basden, was sent for. He examined the baby in the presence of Mr. Jenkins, and said, and subse- quently certified, that there was nothing whatever wrong with her. She was in fact " a perfect baby ". She had been seen by himself or his partner three times since the 23rd July. There was no indication of any kind that on any of these occasions the baby was being maltreated or was backward. Dr. Basden's impression, subsequently verified by affidavit, was that she was a baby " well cared for by a loving mother ". I mention these details because in his dissenting judgment, the Master of the Rolls [1976] 3 W.L.R. at p. 132, claimed to make the assumption that the unknown informant had been making his (or her) allegations in good faith. I can make no such assumption. The circumstantiality of the allegations was such that if, as I must now assume, they were in fact erroneous, it is difficult to reconcile them with good faith and a sound mind. It may, of course, be that at the trial, where, if it takes place, these questions may be canvassed, some perfectly innocent explanation may emerge. In the meantime, however, I make no assumption either as to the good or bad faith of the informant. But, whether I am right or wrong, what is plain is that, if the Appellant Society's claim to withhold disclosure is upheld, the non-disclosure would serve to protect a malicious or reckless as well as a bona fide informant.

It is evident that such an encounter as that which I have described would cause resentment, even anger, on the part of the respondent, and it is clear that it did so in the present case. On the 21st December 1973 the respondent's present solicitors wrote to the appellants a long circumstantial letter ending with a demand, amongst other things, that the appellants disclose the name of the informant and the nature of the information, and on the 31st Decem- ber 1973 this letter was answered by the appellants' present solicitors as categorically refusing on behalf of their clients to make the disclosure required. From these uncompromising attitudes the parties have never resiled. By originating summons dated the 22nd April 1974 the respondent sought

disclosure under section 31 of the Administration of Justice Act 1970 and O.24 r. 7A of the Rules of the Supreme Court. After considering the cases of Norwich Pharmacal Co. v. Commissioners of Customs & Excise [1974] A.C. 133, and Dunning v. United Liverpool Hospitals [1973] 1 W.L.R. 586, this summons was dismissed by Master Jacob on the 18th June 1974. The dismissal was accepted without appeal by the respondent. On the following day, 19th June 1974, the respondent commenced the present proceedings by writ accompanied by a statement of claim. This statement of claim both in its original and amended form (delivered on the 15th May 1975) is in substance an action for personal injury alleged to be due to the negligence of the Society alleging severe shock, depression with a suicidal tendency, and continuing insomnia. For the purposes of the present proceedings we are asked to assume, and I do assume without deciding, that the statement of claim discloses a cause of action. The relief claimed in the statement of claim includes: damages and, unusually, in proceedings for personal injuries, an order " that the defendants disclose to the plaintiff all documents in their custody, possession or power relating to the said complaint and the identity of the complainant". The next relevant step was the application under O.24 r.2 (5) by the appellants which gives rise to this appeal. It asked that " there should be no discovery or inspection of documents under O.24 r.2(l) of the RSC 1965 by the defendants where or (by an amendment) to the extent that, such documents reveal or are capable of revealing the identity of any person . . . who made complaint of the defendants concerning the conduct of the plaintiff towards the child . . .".

This summons has given rise to a considerable degree of judicial disagree- ment. Master Jacob dismissed the summons and made the familiar type of order for discovery against the appellants. Croom-Johnson J. in Cham- bers allowed an appeal from Master Jacob and made an order to the effect requested by the appellants. By a majority of two to one (Scarman L.J. and Sir John Pennycuick, Lord Denning M.R. dissenting) the Court of Appeal reinstated the order of Master Jacob, but granted leave to appeal to the House of Lords. So the matter comes before your Lordships' House. I do not believe that the question involved has been decided in this precise form before, and therefore whichever way the appeal be decided it must to some extent break new ground.

I start with the assumption that every court of law must begin with a determination not as a general rule to permit either party deliberately to withhold relevant and admissible evidence about the matters in dispute. Every exception to this rule must run the risk that because of the withholding of relevant facts, justice between the parties may not be achieved. Any attempt to withhold relevant evidence therefore must be justified and requires to be jealously scrutinised. It is in this frame of mind that I approach the question at issue.

At the same time I utter a word of caution. The facts, disclosure of which is required, must be required for the purpose of deciding the dispute. A collateral purpose is not justified and must be disregarded. It is impossible to recite the sequence of facts which I have rehearsed without agreeing with the observation of Croom-Johnson J. in the course of a careful judgment when he said:

" I do not think that there is really any doubt that what she (the " respondent) is most interested in is discovering who the informer was."

To the extent that the respondent may have a collateral purpose in eliciting the information that purpose must be disregarded.

However that may be, this is not the same thing as saying that the respondent's only interest is the elicitation of the informant's name for the purpose of other proceedings and I proceed on the assumption which I make, and to which both parties adhered, that these are bona fide proceedings arguably disclosing a good cause of action which the respondent genuinely desires to pursue for their own sake, and not simply for discovering the name

of the informant for the purpose of initiating fresh proceedings against him (or her). It is on this basis that I proceed to discuss the questions of this appeal.

Before any other question arises, one must first consider whether the information which the respondent demands is relevant to her action against the appellants. This matter, though fiercely contested up to and including the proceedings in the Court of Appeal, is now conceded. Apart from allegations of aggressiveness and other misbehaviour or tactlessness during his visit against Mr. Jenkins, the respondent's case consists in the assertion that, before authorising his visit, which, she says, was certain to cause distress, the Society should have checked more carefully the accuracy of the informant's credentials. It may be open to question whether this claim gives adequate weight to the Society's obligation to protect the interests of a child supposedly at risk. It may also be questioned whether there were any pre- liminary steps open to the Society to check the information preferable to an actual visit to the mother. But it seems obvious that these matters can hardly be fully canvassed without tracing the information in the possession of the Society to its source, identifying the informant, examining his (or her) bona fides, and assessing the value of the whole. By what right can the courts, the respondent might well ask your Lordships rhetorically, permit a defendant in such proceedings to withhold relevant information, which in the vast majority of cases the courts would have to insist should be laid before the tribunal in order that a just result might be obtained in accordance with law? There are, it is conceded by both parties, some notable exceptions to the general principles concerned. But, says the respondent, the need for complete disclosure in the interest of truth for the purposes of a just result must be paramount, except in a few well-established classes of information, which, says the respondent, should by no means be extended. She divides these exceptions into cases where the public interest is involved, and cases where a privilege, like legal professional privilege can be said to exist.

This has led to a wide discussion by both parties of the nature and limitation of the exceptions, with particular reference to the decisions in your Lordships' House from 1967 onwards beginning with the case of Conway v. Rimmer [l968] A.C. 910.

The appellant Society argued, in effect, for a general extension in range of the nature of the exceptions to the rule in favour of disclosure. This, it was suggested, could be summarised in a number of broad propositions, all in support of the view that, where an identifiable public interest in non-disclosure can be established, either there is a firm rule against disclosure (e.g., legal professional privilege or State secrets) or the court has a discretion whether or not to order disclosure, and that this discretion must be exercised against disclosure in all cases where, after balancing the relevant considerations, the court decides that the public interest in non-disclosure outweighs the ordinary public interest in disclosure. The appellants contended that new cases will arise from time to time calling for a protection from disclosure in classes of case on which it was not previously extended, and that the courts had in practice shown great flexibility in adapting these principles to new situations as and when these arise. The appellants contended that some of those entitled to the benefits of protection had, and some had not, been subject to statutory or common law duties or been clothed with Government authority or been answerable to Parliament or the executive. This contention was aimed at the majority judgments in the Court of Appeal which in substance disallowed the appellants' claim to immunity on the grounds that they are a private Society clothed arguably with authority to fulfil a function but not a duty which they are compelled to perform, and that they are not in any sense either an organ of central Government or part of the public service. The appellants noted that the dissenting judgment of Lord Denning M.R. which was in their favour, largely relied on the confidentiality which the appellants had pledged to potential informants. Their own contention was that, while the mere fact that a communication was made in confidence did not of itself justify non-disclosure, the fact of confidentiality was relevant to reinforce the view that disclosure would be against the public interest. In

this connection the appellants cited Alfred Crompton Amusement Machines Lid. v. Customs and Excise Commissioners (No. 2) [19741 A.C. 405. Lastly the appellants contended that there was no reported case in which the court, once it had identified a public interest in non-disclosure, had ever regarded itself as debarred from taking it into consideration or from weighing its importance against the damage to be apprehended from excluding relevant evidence.

The contentions have at least the merit of propounding a lucid and coherent system. Nevertheless, I am compelled to say that, in the breadth and generality with which they were put forward, I do not find them acceptable.

They seem to me to give far too little weight to the general importance of the principle that, in all cases before them, the courts should insist on parties and witnesses disclosing the truth, the whole truth, and nothing but the truth, where this would assist the decision of the matters in dispute. In the second place, I consider that the acceptance of these principles would lead both to uncertainty and inconsistency in the administration of justice. If they were to be accepted, we should remember that we should be laying down a large innovation not merely in the law of discovery, but equally in the law of evidence, which has to be administered not merely in the High Court, but in the Crown court, the county courts, and the magistrates' courts throughout the land. What is the public interest to be identified? On what principles can it be defined? On what principles is the weighing-up process to proceed? To what extent, if at all, can the right to non-disclosure be waived? Can secondary or extraneous evidence of the facts not disclosed be permitted? To what extent should the Crown be notified of the fact that the issue has been raised? These questions are all manageable if the categories of privilege from disclosure and public interest are considered to be limited. Indeed, reported authority, which is voluminous, shows that largely they have been solved. But to yield to the appellants' argument on this part of the case would be to set the whole question once more at large, not merely over the admitted categories and the existing field but over a much wider, indeed over an undefined, field.

Thirdly, and perhaps more important, the invitation of the appellants seems to me to run counter to the general tradition of the development of doctrine preferred by the English courts. This proceeds through evolution by extension or analogy of recognised principles and reported precedents. Bold statements of general principle based on a review of the total field are more appropriate to legislation by Parliament which has at its command techniques of enquiry, sources of information and a width of worldly-wise experience, far less restricted from those available to the courts in the course of contested litigation between adversaries.

On the other hand, I find equally unattractive the more restricted and even, occasionally, pedantic, view of the authorities advanced on behalf of the respondent. This was based on a rigid distinction, for some purposes valuble, between privilege and public interest, and an insistence on a narrow view of the nature of the interest of the public, reflected in the reasoning of the majority in the Court of Appeal, which would virtually have restricted the public interest cases to the narrower interests of the central organs of the State, or what might be strictly called the public service. The effect of the argument would not merely limit the ambit of possible categories of exception to the general rule. In my view, it would virtually ensure that the categories would now have to be regarded as effectively closed. In her printed case the respondent contended that:

" No party is protected from his obligation to disclose documents on " the grounds of public interest unless there is some connection between " the claim for protection and the functions of central government or " the public service of the State . . . The expression ' Crown privilege' " has been criticised but ... it accurately reflects the basic requirement " that there must be a connection with the Crown or the public service " of the State."

In support of this contention the respondent referred inter alia to Conway v Rimmer (supra), to Reg. v. Lefes Justices ex parte the Home Secretary [1973] A.C. 388 and to Alfred Crompton Amusement Machines Ltd. v. Customs &. Excise Commissioners (No. 2) (supra). There is, of course, a sense, which will become apparent as I proceed, in which the appellants' claim can be brought squarely within the respondent's principle. But the principle is itself, as I shall show, open to criticism. In particular the argument was based on what was described as a fundamental principle that the exceptions to the general rule requiring disclosure all come within one or the other of two rigidly confined categories, one described as privilege, when secondary evidence could be given or the privilege could be waived, and the other as " Public interest" where these possible escapes were excluded. But this, it was contended, was virtually restricted to the category formerly, but inaccurately, referred to as " Crown privilege ".

The result of this is that I approach the problem with a caution greater than that contended for the appellants, but with a willingness to extend established principles by analogy and legitimate extrapolation more flexible than that contended for by the respondent.

I am emboldened to do so by the reflection that, quite apart from legislation like the Civil Evidence Act 1968, the law of evidence has steadily developed since my own practice at the Bar began in 1932. This can be seen by a consideration of cases like McTaggart v. McTaggart [1949] P. 94, Mole v. Mole [1951] P. 21, Theodorapoulas v. Theodoropoulas [1964] P. 311, which undoubtedly developed from the long recognised category of " without prejudice " negotiations but which in my opinion has now developed into a new category of a public interest exception based on the public interest in the stability of marriage. I think the case, widely canvassed in argument of Reg. v. Lewes Justices (supra) was a clear extension of the previous " Crown privilege " type of case by which, for the first time, communications to the Gaming Board were recognised as a suitable object of such " privilege ". Possibly In re D [1970] 1 W.L.R. 599 is another example, for it decided, I think, for the first time, that local authority records of child care investigations were immune from disclosure in wardship proceedings to which they would otherwise be relevant. I believe that traces of similar evolution for instance in the field of legal professional privilege can be found in the nineteenth century authorities.

I find it also interesting to note that the report (16th report of the Law Reform Committee Cmnd. 3471) to which judges of every Division of the High Court were signatories, which was referred to extensively by counsel for both sides, shows a definite development in the law and practice in the precise field now under discussion from what it was generally considered to be when I entered the profession in 1932.

According to paragraph 1 of that report, which is before us, but which represents no more than contemporary text book authority:

" Privilege in the main is the creation of the common law whose " policy, pragmatic as always, has been to limit to a minimum the " categories of privileges (sic) which a person has an absolute right to " claim, but to accord to the judge a wide discretion to permit a witness, " whether a party to the proceedings or not, to refuse to disclose " information where disclosure would be a breach of some ethical or " social value and non disclosure would be unlikely to result in serious " injustice in the particular case in which it is claimed." (emphasis mine)

This doctrine was not merely an incidental statement at the beginning of the report. It runs right through it, and forms the basis of some of the most notable conclusions (see e.g. paragraph 3, paragraph 7, paragraphs 36, 37, paragraphs 41, 43, paragraph 48-52).

Counsel for the respondent, who was himself, as he candidly confessed, signatory to the report, was constrained to argue that the report, the authors of which included Lords Pearson and Diplock, Winn and Buckley L.JJ., Orr J. and the present Vice Chancellor, was an inaccurate representation

of the then existing state of the law, and that the two cases (Attorney-General v. Clough [1963] 1 Q.B. 773 and Attorney-General v. Mulholland [1963] 2 Q.B. 477) cited in the report to support the proposition, did not in truth do so, were wrong if they did, and, being modern, departed from legal principle. Speaking for myself, I am sure that the law has in fact developed in this field during my life time, and I find it incredible that paragraph (1) of the report cited bearing the weight of judicial authority I have described does not represent the current practice of the courts in 1967, although in fact it goes plainly beyond the current practice of my youth.

For these reasons, I feel convinced that I am entitled to proceed more boldly than counsel for the respondent argued, though more timidly than the robust counsels of the appellants' counsel urged.

The authorities, therefore, seem to me to establish beyond doubt that the courts have developed their doctrine in this field of evidence. An example of this is seen in the privilege extended to editors of newspapers in the nineteenth century, before the present O.82 r.6 was passed, to refuse to answer interrogatories in defamation cases where the issue was malice, and the plaintiff desired to discover their sources (cf. Hope v. Brash [1897] 2 Q.B. 188 ; Hennessy v. Wright [1888] 21 Q.B.D., Plymouth Mutual Co-operative and Industrial Society Ltd. v. Traders' Publishing Association Ltd. [1906] 1 K.B. 403). This practice, robustly developed by the judges of the Queen's Bench Division (in contrast with the contemporary Chancery Division practice even after 1873) can only have been based on public policy. It has been stressed that these cases relate to discovery and not to questions to witnesses at the trial. This may well be so, at least at present, but certainly they illustrate the use of the court of a discretion, and its sensitiveness to public policy where discretion exists. Until the introduction of the new rules it is within my recollection that interrogatories and discovery on the lines dis- allowed in the newspaper cases were frequently allowed in other defamation cases where malice was in issue, although it was pointed out in argument that the newspaper principle was, at least once, applied rather strangely, to M.Ps. in Adam v. Fisher (1914) 30 T.L.R. 288.

In all this argument, however, two facts stand out unmistakeably as true beyond dispute. The first is that the welfare of children, particularly of young children at risk of maltreatment by adults, has been, from the earliest days a concern of the Crown as parens patriae, an object of legal charities, and in latter years the subject of a whole series of Acts of Parliament, of which the Act of 1969 is only an example, and that not the latest. The second is that the information given by informants to the police or to the Director of Public Prosecutions, and now, since Reg. v. Lewes Justices (supra) to the Gaming Board is protected from disclosure in exactly the manner demanded by the appellants. The question, and it is I believe the only question necessary to be decided in this appeal, is whether an extension of this established principle to information relating to possible child abuse supplied to the appellants is a legitimate extension of a known category of exception or not. For this purpose it is necessary to consider the position of the appellants in relation to the enforcement provisions of the Children and Young Persons Act 1969.

For the purposes of this enquiry the provisions of the Act can be shortly summarised. A child at risk in certain ways can be brought before a juvenile court in care proceedings. Such proceedings can only be brought by three classes of person, each of whom has a locus standi (see section 1(1) of the Act). These classes of person are(i) a constable

(ii) the relevant local authority and

(iii) an authorised person, that is a person authorised by the Secretary of State for Home Affairs in this behalf. Other persons may of course bring prosecutions for breaches of the criminal law, and under section 28 can bring proceedings to enable a child in imminent danger to be brought before the court, detained, and taken to a place of safety. But care proceed- ings can only be brought before the juvenile court by one of the three classes of person. It is common ground that the only " person " authorised by the Secretary of State within the meaning of section 1 of the Act is the appellant

Society, a body founded in 1889, and incorporated since 1895 by Royal Charter. To that extent the Society has been charged with the performance of a public responsibility by the Home Secretary under the direct authority of an Act of Parliament.

It is true that, by virtue of section 2 of the Act, whereas the local authority is under a residual duty to take care proceedings where no-one else does, the other two classes of person have only the capacity to do so. Though the contrary was argued, and the argument may have found favour with the majority in the Court of Appeal, I do not believe that this can make any difference to the degree of protection afforded to a member of the public who desires to give information regarding child abuse. Such potential informants can hardly be expected to understand such niceties. In point of fact the evidence is that they more frequently prefer the appellants to the other two classes of potential initiators of care proceedings as recipients of their confidences.

It is also true that statistically only a small proportion of the Society's cases in fact end up in court. This again appears to me to be irrelevant to the point under consideration. The function of the Society, or, for that matter, the police or the local authority, appears to me to receive the information and to sift it. In a majority of cases no court proceedings result, even where the information proves reliable. A wise word to the parent, an indication to the child that he should return home, may well prove to be enough. But the Society must make up its mind as to the appropriate course to take only after receiving and processing the information. No doubt the same is true of the police and the local authority. The undisputed evidence for the appellants is that Society which solicits information from the public under the pledge of secrecy would suffer a " drastic reduction " in the flow of informa- tion available to it to the danger of many children were it known that the name of the informant was subject to disclosure in the event of subsequent proceedings against the Society.

Of the three classes with locus standi to initiate care proceedings, it is common ground that information given to the police is protected to the extent demanded by the Society. This is clear from many cases including Marks v. Beyfus

[1890] 25 QBD 494

C.A. (which applied the principle to the Director of Public Prosecutions), and many of the recent cases in your Lordships' House. The rule relating to the immunity accorded to police informants is in truth much older, so old and so well-established, in fact, that it was not and could not be challenged in the instant case before your Lordships. Once, however, it is accepted that information given to the police in the instant case would have been protected, it becomes, in my judgment, manifestly absurd that it should not be accorded equally to the same information if given by the same informant to the local authority (who would have been under a duty to act on it) or to the appellant Society, to whom according to the undisputed evidence, ordinary informants more readily resort.

The last point seems to have been realised, at least to some extent, by Sir John Pennycuick. See for instance the passage in his judgment at [1976] 3 W.L.R. at p. 144E. But I cannot see the sense of allowing the immunity where care proceedings actually result, but not in cases where the society or the local authority, after sifting the information, and assessing the credentials of the informants, decide in the event upon an alternative course. It is not for the informant to predict what course the recipient of the information may take, nor does his (or her) right to anonymity depend upon the outcome. The public interest is that the parties with locus standi to bring care proceedings should receive information under a cloak of confi- dentiality. It may well be that neither the police, the local authority, nor the society can give an absolute guarantee. The informant may in some cases have to give evidence under subpoena. In other cases their identity may come to light in other ways. But the police, the local authority and the society stand on the same footing. The public interest is identical in relation to each. The guarantee of confidentiality has the same and not different

values in relation to each. It follows that the Society is entitled to succeed upon the appeal.

The Master of the Rolls, in his dissenting judgment, places his own reasoning on the pledge of confidentiality given by the Society, and seeks to found the immunity upon this pledge. I do not think that confidentiality by itself gives any ground for immunity (cf. e.g. per Lord Cross in Alfred Crompton Amusement Machines Ltd. v. Customs and Excise Commissioners (No. 2) [19741 A.C. 405 at p. 433). Confidentiality is not a separate head of immunity. There are, however, cases when confidentiality is itself a public interest and one of these is where information is given to an authority charged with the enforcement and administration of the law by the initiation of court proceedings. This is one of those cases, whether the recipient of the information be the police, the local authority or the N.S.P.C.C. Whether there be other cases, and what these may be must fall to be decided in the future. The categories of public interest are not closed, and must alter from time to time whether by restriction of extension as social conditions and social legislation develop.

The result is that this appeal must be allowed, and an order made in the terms formulated by Croom-Johnson, J.

Lord Simon of Glaisdale

My Lords,

In many unsettled societies antisocial behaviour is restrained, if at all, by fear of reprisal. But in such societies self-interest (or its idealisation in the concept of honour) is thought to require counter-reprisal; and the vendetta ensues as a social institution. In many such societies the earliest intervention of central government in the establishment of public order is by control of the vendetta through limitation of the range and class of kindred entitled to visit reprisal or be its victim, and also of modes of reprisal. But as society gradually becomes more settled and central govern- ment more effective, society is able to insist that antisocial conduct shall be controlled, not at all by individual or clan violence, but by adjudication in independent courts of law according to an objective and enduring code of law. Even though sometimes delegated, the punishment of antisocial conduct becomes exclusively the prerogative of society itself—in its poli- tically organised aspect, the State. Moreover, civil wrongdoing, in primitive societies barely differentiated from criminal conduct, must also be brought to the bar of independent judgment according to a code. The foregoing, even if not true of the evolution of every society, marks the early evolution of our own.

Thus it is clear that the administration of justice is a fundamental public interest. But it is also clear that it is not an exclusive public interest. It is an aspect (a crucially important one) of a broader public interest in the maintenance of social peace and order. The same conclusions can be as readily reached analytically as historically.

As the judicial function becomes differentiated so also do the executive functions of central and local government. Central and local government, indeed, come to take on functions additional to the maintenance of internal order and of security against external enemies; and are recognised to be so performing functions which are also in the public interest. For the most part the various organs of the constitution so evolved can proceed on their complementary functions without collision or friction. But occasionally priorities have to be established. Conway v. Rimmer [19681 A.C.910 exemplifies the legal recognition of a situation where two public interests are in competition, and establishes how the competition may be resolved.

The conflict in Conway v . Rimmer arose because the public interest in the administration of justice prima facie requires that the objective and enduring

code of law established by society should be applied to all facts relevant to the dispute before the court, whereas another public interest was said to call for the withholding of such evidence. (Lord Reid in Conway v. Rimmer at p.940.) But even the public interest in the administration of justice allows some qualification of a rule of universal admissibility of all relevant evidence. Experience has shown that some evidence may appear to have more weight than it really has. For example, past experience showed that evidence of facts which are retold at second or further hand was generally less reliable than of those told at first hand ; so hearsay evidence, however relevant, is still rarely admitted in a criminal trial. Then again, evidence of propensity to commit crime in general, or the particular sort of crime in question, is only exceptionally admitted; because, although undoubtedly relevant, its relevance is, in general, liable to be grossly outweighed by its prejudice to the accused—in other words, experience has shown that, if admitted, it is apt to be given more weight than is its due so that a fair trial of the accused may be prejudiced.

Nearer the heart of the argument in the instant appeal lies another class of relevant evidence which the public interest in the administration of justice itself demands should be withheld from the courts. This is evidence excluded by legal professional privilege. Our national experience found that justice is more likely to ensue from adversary than from inquisitorial procedures—Inquisition and Star Chamber were decisive, and knowledge of recent totalitarian methods has merely rammed the lesson home. To promote justice the adversary procedure involves advocacy of contrary contentions by representatives with special gifts and training. In the words of Dr. Johnson :

" As it rarely happens that a man is fit to plead his own cause, lawyers " are a class of the community who, by study and experience, have " acquired the art and power of arranging evidence, and of applying to " the points at issue what the law has settled. A lawyer is to do for " his client all that his client might fairly do for himself if he could."

This process would be undermined if the trained advisers were compelled to divulge weaknesses of their cases arising from what they had been told by their clients. Indeed, the adversary system, involving professional assistance, could hardly begin to work effectively unless the client could be sure that his confidences would be respected. And a legal representative with only partial knowledge of his case would be like a champion going into battle unconscious of a gap in his armour. But it is only the rare case which has to be fought out in court. Many potential disputes, civil especially, are obviated or settled on advice in the light of the likely outcome if they had to be fought out in court. This is very much in the interest of society ; since a lawsuit, though a preferable way of settling a dispute to actual or threatened violence, is wasteful of human and material resources. Thus similar considerations apply whenever a citizen seeks professional guidance from a legal adviser—whether with a view to undertaking or avoiding litigation, whether in arranging his affairs in or out of court.

My Lords,

I have just dealt with a class of evidence which may be excluded by the balance of the public interest in the administration of justice partly

in furtherance of the forensic process itself, partly in its obviation. The next stage is a class of evidence which may be excluded by public interest in the obviation of litigation alone. These are "without prejudice" communications. Since litigation is wasteful and disruptive, society benefits if disputes can be settled out of court through negotiation between the parties. This inevitably involves concessions by one side or the other or by both. But a party would be deterred from making a concession if it could in the event of breakdown in negotiation, be used against himself in court' is more important to society that parties should be able without such embarrassment to negotiate towards a settlement of their dispute than that

any admission made in the course of negotiation should, in the event of breakdown, be adduced in court so that judgment might be pronounced in the light of all relevant evidence. The law therefore, in response to society's needs, says that the desirability of the objective code being applied to all relevant evidence is outweighed by the desirability that parties should be able to negotiate freely towards a settlement.

Then the law proceeds to recognise that the public interest in the administration of justice is one facet only of a larger public interest—namely, the maintenance of the Queen's Peace. Another facet is effective policing. But the police can function effectively only if they receive a flow of intelligence about planned crime or its perpetrators. Such intelligence will not be forthcoming unless informants are assured that their identity will not be divulged. (See Lord Reid in Conway v. Rimmer at pp.953G-954A.) The law therefore recognises here another class of relevant evidence which may —indeed, must—be withheld from forensic investigation—namely, sources of police information: Rex v. Hardy 24 State Tr.199, 508; Hennessy v. Wright (1888) 21 Q.B.D.509, 519; Marks v. Beyfus (1890) 25 Q.B.D.494.

Here, however, the law adds a rider. The public interest that no innocent man should be convicted of crime is so powerful that it outweighs the general public interest that sources of police information should not be divulged ; so that, exceptionally, such evidence must be forthcoming when required to establish innocence in a criminal trial: see citations at [1973] A.C. p.408A. It would appear that the balance of public interest has been struck, both in the general rule and in its rider, in such a way as to conduce to the general advantage of society, with the public interest in the administration of justice as potent but not exclusive.

I cannot leave this particular class of relevant evidence withheld from the court without noting, in view of an argument for the respondent, that the rule can operate to the advantage of the untruthful or malicious or revengeful or self-interested or even demented police informant, as much as of one who brings information from a high-minded sense of civic duty. Experience seems to have shown that though the resulting immunity from disclosure can be abused, the balance of public advantage lies in generally respecting it.

Then, to take a further step still from the public interest in the administration of justice, the law recognises other relevant public interests which may not always even be immediately complementary. For example, national security. If a society is disrupted or overturned by internal or external enemies, the administration of justice will itself be among the casualties. Silent enim leges inter arma. So the law says that, important as it is to the administration of justice that all relevant evidence should be adduced to the court, such evidence must be withheld if, on the balance of public interest, the peril of its adduction to national security outweighs its benefit to the forensic process—as to which, as regards national security in its strictest sense, a ministerial certificate will almost always be regarded as conclusive: see Lord Parker of Waddington in The Zamora [1916] 2 A.C.77, 107.

My Lords,

without attempting to be exhaustive I have tried to show that there is a continuum of relevant evidence which may be excluded from forensic scrutiny. This extends from that excluded in the interest of the forensic process itself as an instrument of justice (e.g., evidence of propensity to commit crime), through that excluded for such and also for cognate interests (e.g., legal professional privilege), through again that excluded in order to facilitate the avoidance of forensic contestation (e.g., " without prejudice" communications), to evidence excluded because its adduction might imperil the security of that civil society which the administration of justice itself also subserves (e.g., sources of police information or state secrets). These various classes of excluded relevant evidence may for ease of exposition be presented under different colours. But in reality they constitute a spectrum, refractions of the single light of a public interest which may outshine that of the desirability that all relevant evidence should be

adduced to a court of law. I have emphasised this because it bears on both the argument for the respondent and the main argument for the appellants.

The argument for the respondent was that there are two separate categories of excluded evidence, which go respectively by the names of " Privilege and "Public Policy"; and that each contains sub-categories which are established and unextendable, either through the recognition of new sub- categories or by analogy with those which have been already recognised. They are not to be extended because, first, the public interest with which your Lordships are paramountly concerned is in the administration of justice, which demands that all relevant evidence should be adduced to the court, and, secondly, it is nowadays for Parliament and not the courts to give legal recognition to new heads of public policy (see Fender v. St. John-Mildmay fl938] A.C.I, 10-12, 23, 28). It is true that one of the established sub-categories under the head of " Public Policy " is " Evidence excluded by considerations of State interest"; but, "State", it was argued, here refers only to the Crown and the departments of central government. This last was indeed the main ground of decision of the majority of the Court of Appeal; and I shall return to it later.

With regard to the argument based on categorisation, counsel for the respondent relied on the high authority of Phipson and Cross on Evidence, both of which deal in separate chapters with evidence excluded by respectively " Privilege" and " Public Policy". But this, I venture to suggest, is merely for convenience of exposition. I hope that I have shown, first, that the classes of excluded evidence extend beyond what is dealt with in these chapters and, secondly, that the principle underlying the exclusion from forensic scrutiny of any relevant evidence is an overriding public interest which demands its exclusion. But when it comes to the point of forensic impact the potentiality of exclusion may enure solely for the benefit of one of the parties or a witness before the court. Such, for example, is the actual impact of legal professional privilege, notwithstanding that it is required generally for the rule of law, forensically its particular incidence is exclusively for the benefit of the client. Quilibet potest renunciare juri pro se introducto. The client can therefrom waive the exclusion ; so that it is properly and conveniently called a " privilege " and grouped with other similar evidence. By contrast, the exclusion of evidence because its adduction would imperil the security of the State enures to the advantage of citizens generally, so that its exclusion cannot be waived by any party or witness before the court: Marks v. Beyfus at p.500. Nor, of course, in such case can secondary evidence be given of the excluded matter. There is no harm in categorising this sort of non-waivable exclusion under the heading of " Public Policy ": provided that it is recognised, first, that the exclusion of any relevant evidence is to be justified on grounds of public policy (which may be the public interest in the efficiency and fairness of the forensic process itself); secondly, that dealing with it in this way merely signifies methodologically that it constitutes the residuum of classes of excluded evidence which cannot be dealt with appropriately under other headings (e.g., hearsay, " the best evidence rule", non-compellability or privilege); and, thirdly, that the label " Public Policy " here does not mean that the courts must necessarily wait upon Parliament or must necessarily refrain from the normal common-law process of applying an established rule to circumstances analogous to those in which the rule was established (in the absence of contra-indication on grounds of justice or convenience): see Parke B., giving the advice of the judges to your Lordships' House, in Mirehouse v. Rennell (1833) 1 Cl. & F. 527 546, cited with approval in Shaw v. D.P.P. [1962] A.C.220, 261, 289, 291, 302-3.)

The overriding rule is the general one that courts of law must recognise their limitations for decision-making—that there are many matters in which the decision is more appropriately made by the collective wisdom of Parlia- ment on the advice of an executive (itself collective in a. system of Cabinet government) briefed by officials who have investigated over a wide field the repercussions of the decision. Such, for example, are those decisions which

may affect the national balance of payments or the public safety, in contra- distinction to decisions where the court can feel reasonably confident that there are unlikely to be unforeseen repercussions requiring extra-forensic action—for example, where the subject-matter is " lawyers' law " (see, e.g., D.P.P. v. Shannon [1975] A.C.717). '

The question in the instant case is peculiarly within the competence of the judge in his control of the cursus curiae; the development of the law has generally been judicial; parliamentary intervention is likely to be at best occasional and delayed; and the Law Commissioners do not appear to have taken cognisance of the problem.

Nevertheless, invocation of " public policy " does impose even more than normal self-restraint on a court. Of course, every rule of law is a legal manifestation of public policy. But your Lordships are, as counsel for the respondent rightly emphasised, instantly concerned with "public policy" in a narrower sense—namely, consideration of social interests beyond the purely legal which call for the modification of a normal legal rule (here the rule that all relevant evidence should be adduced to the court): cf. Lord Wright in Fender v. St. John-Mildmay at p.28. In such circumstances the correct approach to the function of the courts is, in my view, expressed by Lord Thankerton in Fender v. St. John-Mildmay at p. 23 :

" Their duty is to expound, and not to expand, such policy. That " does not mean that they are precluded from applying an existing " principle of public policy to a new set of circumstances, where such " circumstances are clearly within the scope of the policy."

This suggests, in my judgment, that a narrow rather than a wide ground of decision would be desirable in the instant case, if the former is available; that your Lordships should primarily look to see whether the law has recognised an existing head of public policy which covers the case; and that, if so, your Lordships should if possible vindicate such policy by means already adapted by the law to vindicate some analogous head of public policy.

I turn, though, first to the argument that your Lordships are here con- cerned with what was formerly called " Crown Privilege " and that it was " Privilege " not " Crown " which was misleading ; so that the public interest which may exclude relevant evidence is only that of the Crown or the public service of the State—" State " here being restricted to the organs or departments of central government. There seems to me to be a number of objections to these arguments. First, so to restrict the principle on which relevant evidence may be excluded runs counter to its rationale as I have ventured to submit it to your Lordships, according to which both " Crown " and " Privilege " must be misnomers. Secondly " the State " cannot on any sensible political theory be restricted to the Crown and the departments of central government (which are, indeed, part of the Crown in constitutional law). The State is the whole organisation of the body politic for supreme civil rule and government—the whole political organisation which is the basis of civil government. As such it certainly extends to local—and, as I think, also statutory—bodies in so far as they are exercising autonomous rule. Thirdly, there is a recurrent transfer of functions between central, local and statutory authorities. For example, near the heart of the issue before your Lordships, the Crown as parens patriae had traditionally a general jurisdiction over children: a residue is now exercised in the High Court, but the bulk has been devolved by statute on local authorities. Recently, posts and telegraphs have ceased to be the concern of a department of central government responsible to Parliament and have become the function of a virtually autonomous statutory body. They have become a nationalised industry under a largely independent Board—in no real political or legal sense the Crown or a department of central govern- ment! Some nationalised industries are carried on by Regional Boards: others have been, but no longer are. The delimitation proposed on behalf

of the respondent would thus work out capriciously and mutably. Fourthly, a number of the cases are against the respondent's contentions. In Adam v. Fisher (1914) 30 T.L.R.288 it was held, on a summons relating to interro- gatories in a libel suit, that an M.P.'s sources of information need not be disclosed. In Conway v. Rimmer the appeal concerned reports on a pro- bationary police constable in a county constabulary. In In re D [1970] 1 W.L.R.599 the material protected from disclosure was case records compiled in pursuance of statutory regulation relating to a child boarded out with foster-parents by a local authority. In Reg. v. Lewes Justices it was infor- mation supplied to the Gaming Board of Great Britain. Adam v. Fisher may be explained away as relating solely to discovery of documents, which still retains an element of its discretionary origin (see the explanations in Attorney-General v. Clough [1963] 1 Q.B. 773, 790, and in Attorney-General v. Mulholland, Attorney-General v. Foster [1963] 2 Q.B.477, 490, where Lord Denning M.R. described Adam v. Fisher as expressing a rule of prac- tice not law). But the other cases simply cannot be fitted into the respon- dent's attempted limitation of this category of excluded evidence.

Before I leave the authorities I venture to note that there is a line of cases which defies the respondent's argument that there is a closed number of unextendable categories of relevant evidence which may be withheld from forensic scrutiny. These are the cases relating to marriage conciliation (see e.g., McTaggart v. McTaggart [1949] P.94; Mole v. Mole [1951] P.21; Theodoropoulas v. Theodoropoulas [1964] P.311), where we can watch a recent development of the law. Incidentally, too, they provide a neat example of the way that " Privilege " in this branch of the law is based on public interest just as much as those categories of excluded evidence collected under the heading of " Public Policy ". With increasingly facile divorce and a vast rise in the number of broken marriages, with their concomitant penury and demoralisation, it came to be realised, in the words of Bucknill L.J. in Mole v. Mole at p.23.

" in matrimonial disputes the State is also an interested party ; it is more " interested in reconciliation than in divorce."

This was the public interest which led to an application by analogy of the privilege of " without prejudice " communications to cover communications made in the course of matrimonial conciliation (see McTaggart v. McTaggart, Cohen L.J. at p.96, Denning L.J, at p.97; Denning L.J. in Mole v. Mole at p. 24; Theodoropoulas v. Theodoropoulas at p.314)—so indubitably an extension of the law that the textbooks treat it as a separate category of relevant evidence which may be withheld from the court. It cannot be classified, like traditional " without prejudice" communications (see 16th Report of Law Reform Committee, Cmnd. 3472 of 1967, para. 18), as a " privilege in aid of litigation ".

There have been three attempts to impose a comprehensive and coherent pattern on this branch of the law: I have great sympathy with the object, though I feel bound to express reservations in the case of each. They are that of Lord Denning M.R. in the instant case, that of the Law Reform Committee in their 16th Report, and that of counsel for the appellants in his main argument. The solution of the learned Master of the Rolls was to suggest confidentiality of a communication (or in the relationship of the parties) as the criterion for exclusion. The Law Reform Committee found a common factor in:

" a wide discretion [in the court] to permit a witness, whether a party to " the proceedings or not, to refuse to disclose information where dis- " closure would be a breach of some ethical or social value and non- " disclosure would be unlikely to result in serious injustice in the " particular case in which it is claimed."

Counsel for the appellants, while relying on much in the Law Reform Committee's Report, put his case with slightly different emphasis. He argued that in each case (save those governed by an existing rule against disclosure) the court will weigh any public interest in the withholding of information against the public interest that all relevant evidence should be adduced to court; and if the former is preponderant the evidence will be excluded.

I do not think that the confidentiality of the communication provides in itself a satisfactory basis for testing whether relevant evidence should be withheld. First, it does not sufficiently reflect the true basis on which any evidence is excluded—namely, the public interest. Even Wigmore (Evidence, 1st ed. 1905, vol. IV, sec. 2285; ib . 3rd ed. 1940), who stipulates for a principle of confidentiality as a condition of testimonial privilege (and I emphasise that he is dealing only with privilege) states (sec. 2286 ; ib. 3rd ed.):

" In general, then, the mere fact that a communication was made in " express confidence, or in the implied confidence of a confidential " relation, does not create a privilege. This [3rd ed., Common Law] " rule is not questioned today." (his italics)

In the words of Wigmore, for the privilege to attach, the relationship between the parties to the communication:

" must be one which in the opinion of the community ought to be " sedulously fostered." (his italics)

Secondly, a juridical basis of confidentiality does not explain why, in relation to certain classes of excluded evidence, there can be no waiver of the immunity. Thirdly, certain evidence is excluded, not because it is confidential (even in the sense of being secret), but because it relates to affairs of state. For example, it was on that ground and not irrelevance that Cobbett was precluded in his trial for seditious libel from asking a witness whether it would not be wise to follow his (Cobbett's) advice as to how to deal with current civil disturbance (R. v. William Cobbett (1831) 2 State Tr. 789, 877). Fourthly, the law would operate erratically and capriciously according to whether or not a particular communication was made confidentially: Delane, the great 19th century editor of The Times, always refused to receive information under the seal of secrecy, because sooner rather than later he would get the same information from a source he could use. Fifthly, it is undesirable that exclusion should be conferred by confidentiality irrespective of the public interest: after all, an attempt to bribe is generally made confidentially (cf. Lewis v. James (1887) 3 T.L.R. 527; In re Hooley, Rucker's Case (1898) 79 L.T. 306; McGuinness v. Attorney-General of Victoria (1940) 63 C.L.R. 73). Sixthly, confidentiality was in fact the original and far-reaching ground of exclusion. A man of honour would not betray a confidence, and the judges as men of honour themselves would not require him to. Thus originally legal professional privilege was that of the legal adviser, not the client. (For the foregoing see Wigmore, secs. 2286, 2290). But, with the decline in the ethos engendering the rule, the law moved decisively away from it. The turning point was The Duchess of Kingston's Case (1776) 20 Howell St.Tr.353, 386-391, where both the Duchess's surgeon and a personal friend, Lord Harrington, were compelled to give evidence in breach of confidence. Seventhly, there is massive authority in addition to Wigmore and the Duchess of Kingston's Case against confi- dentality by itself conferring exclusion: Sir G. Jessel M.R. in Wheeler v. le Marchant (1881) 17 ChD 675 , 681 ; Lord Parker of Waddington C.J. in Attorney-General v. Clough [1963] 1 Q.B.773, 787; Lord Denning, M.R., in Attorney-General v. Mulholland, Attorney-General v. Foster [1963] 2 Q.B. 477, 489, Donovan L.J. and Danckwerts L.J. agreeing; Lord Salmon in Reg. v. Lewes Justices at pp.411H-412A; Lord Cross of Chelsea in Alfred Crompton Ltd. v. Customs and Excise Commissioners [1974] A.C.405, 433H ; O'Brennan v. Tully (1935) 69 Ir. L.T. 115 (cited with approval in Attorney- General v. Mulholland at p. 491); McGuiness v. Attorney-General of Victoria (1940) 63 C.L.R. 73, which contains a judgment of characteristic authority by Dixon J. dealing with the plea of confidentiality, (cited with approval in Attorney-General v. Clough at pp.790-791 and in Attorney-General v. Mulholland at p.491). See also Bray on Discovery (1885 ed. p.303). 1 think the true rule is expressed in Wigmore and in the passage referred to in the speech of Lord Cross of Chelsea:

"' Confidentiality ' is not a separate head of privilege, but it may " be a very material consideration to bear in mind when privilege is " claimed on the ground of public interest."

(It is only right to say that counsel for the appellants did not rely on con- fidentiality pur sang as a criterion of exclusion, but rather on the way it was put by Lord Cross.) For the reasons I have given I do not myself think that confidentiality in itself establishes any public interest in the exclusion of relevant evidence: but rather that it may indirectly be significant where a public interest extrinsically established (e.g., provision of professional legal advice or effective policing) can only be vindicated if its communications have immunity from forensic investigation.

I naturally feel the same temerity in approaching the Report of the powerful Law Reform Committee as I do in approaching the judgment of the learned Master of the Rolls. But since counsel for the appellants relied greatly on the Report for his wide general proposition, I feel bound to express my reservations. I would start by pointing out that the Committee was concerned only with civil proceedings, and within them only with " privilege " from disclosure. Even though the rules of criminal evidence may differ in some respects from civil, any wide judicial discretion to admit or reject evidence should, I think, at least be tested against what would be acceptable in a criminal trial. Secondly, I do not think that dough's case or Mulholland's really supports the existence of such a wide discretionary power as the Committee considered to vest in the court (except for the judgment of Donovan L.J. at p.492 of the latter case). Thirdly, the massive authority I referred to in the preceding paragraph of this speech must at least be weighed in the other scale. See also Marks v. Beyfus (at pp.498 Lord Esher M.R., 500 Bowen L.J.—not a matter of discretion, but a rule of law.) Fourthly, I think that the true position is that the judge may not only rule as a matter of law or practice on the admissibility of evidence, but can also exercise a considerable moral authority on the course of a trial. For example, in the situations envisaged the judge is likely to say to counsel:

" You see that the witness feels that he ought not in conscience to answer that question. Do you really press it in the circumstances?"

Such moral pressure will vary according to the circumstances—on the one hand, the relevance of the evidence ; on the other, the nature of the ethical or professional inhibition. Often indeed such a witness will merely require a little gentle guidance from the judge to overcome his reluctance. I have never myself known this procedure to fail to resolve the situations acceptably. But it is far fom the exercise of a formal discretion. And if it comes to the forensic crunch, as it did in many of the cases I have referred to (to which can be added the Parnell Inquiry Commission —103rd day, see footnote to Wigmore para. 2286 ; also another passage cited in Mulholland's case at pp.490-491), it must be law not discretion which is in command. It may be that the members of the Law Reform Committee considered that a consistent use of moral suasion had resulted in a rule of practice emerging, cf. Povey v. Povey [1972] Fam. 40, 48-49 (although I am not convinced myself that it has). Lastly, many of the practical objections voiced by my noble and learned friend, Lord Hailsham of St. Marylebone, to the main and wider proposition advanced on behalf of the appellants seem to me to apply equally to the proposition of the Law Reform Committee. But it may be that some of the relationships will need re-examination as matters of practice or law ; and it is to be borne in mind that it has been found expedient in some jurisdictions to modify the common law rule of disclosure by giving statutory immunity to, e.g., doctors or priests.

My Lords,

I have dwelt on this matter because, as I said, counsel for the appellants relied considerably on the Report for his wide proposition—a general discretion in the court to weigh conflicting public interests in the adduction or exclusion of evidence. He also, of course, relied on Conway v. Rimmer, where conflicting public interests were indeed weighed. But your Lordships' House was really there concerned with the validity of claims by the Crown (based on Duncan v. Cammell Laird & Co. Ltd. [19421 A.C.624) that the executive could procure the exclusion of evidence by a conclusive ministerial certificate that the evidence belonged to a class the disclosure of any part of which would be detrimental to the public interest. Your Lordships' House overruled Duncan v. Cammell Laird & Co. Ltd. in this

respect and further laid down that if in doubt the court could itself look at a document in the light of any ministerial certificate in order to ascertain whether its forensic publication could really affect the public interest adversely. I do not think that Conway v. Rimmer provides any real foundation for the appellants' wide proposition.

That proposition does, on the other hand, reflect the general principles underlying this branch of the law, as I endeavoured to state them near the outset of this speech. Nevertheless, your Lordships are here concerned with public policy, with all the circumspection which such concern enjoins.

The first question on such a circumspect approach is not so much to canvass general principle as to ascertain whether the law has recognised an existing head of public policy which is relevant to this case. Of that there can be no doubt. The need of continuity in society; the legal application to children of the traditional role of the Crown as parens patriae; its exercise in the Court of Chancery in such a way as to make the welfare of a child the first and paramount consideration in matters of custody and guardianship (In re Thain

[1926] Ch. 676

); a vast code of legislation starting with the Prevention of Cruelty to Children Act 1889 and culminating in the Children Act 1975; In re D decided in this very branch of the law—all this attests beyond question a public interest in the protection of children from neglect or ill-usage.

The patria potestas in respect of children in need of help has been largely devolved on local authorities. But the appellants, not only by royal charter but also by statutory recognition, have an important part to play. Apart from the police and the local authority, they are the only persons authorised to take care proceedings in respect of a child or young person (Children and Young Persons Act 1969, section 1 ; S.I. 1970/1500). They have, of course, other important functions for the protection of children from neglect or ill-usage; my noble and learned friends who have preceded me have set them out.

Before passing to the next question I must deal with an argument on behalf of the respondent which arises at this point. Counsel emphasised that the appellants have legal and other powers and functions, but no legal duties in this field. Only the local authority has a duty to take care poceedings and, for example, provide places of safety. The law, it was argued, will only exclude sources of information from disclosure in court if the information is given to someone who has a duty to act. No authority was cited in support of this assertion ; and, with all respect, I cannot agree with it. First, it is the performance of the junction of safeguarding children who may be in peril which is the concern of society ; enjoining a legal duty is merely a way of ensuring that the junction is performed. Secondly, the police too have only a function (not a duty); but it is accepted that police sources of information about children who may be in peril cannot be as regards care proceedings investigated in court.

This brings me to the penultimate question. Is protection of their sources of information necessary for the proper performance of their functions by the appellants? As to this there is uncontradicted and entirely plausible evidence. The answer is Yes. This satisfies Wigmore's second test: the element of confidentiality is " essential to the full and satisfactory maintenance of the relation between " the appellants and their informants. And the answers to this and the preceding question together meet Wigmore's third criterion: the relation is " one which in the opinion of the community ought to sedulously fostered."

The final question, my Lords, is whether the appellants' sources of information can be withheld from forensic investigation by extending on strict analogy an established rule of law. I have already cited long-standing and approved authority to the effect that sources of police information are not subject to forensic investigation. This is because liability to general disclosure would cause those sources of information to dry up. so that police

protection of the community would be impaired. Exactly the same argument applies in the instant case, if for " police " you read " N.S.P.C.C." and tot "community" you read "that part of the community which consists of children who may be in peril". There can be no material distinction between police and/or local authorities on the one hand and the appellants on the other as regards protection of children. It follows that, on the strictest analogical approach and as a matter of legal rule, the appellants are bound to refuse to disclose their sources of information.

I would therefore allow the appeal.

I would, however, add three observations. First, if the correct view were that the judge has a discretion to weigh conflicting public interests, either as indicated by the Law Reform Committee or as urged on behalf of the appellants in their main and wide proposition, I think that it was exercised by Croom-Johnson J. in his careful judgment. Secondly, I agree with my noble and learned friend, Lord Hailsham of St. Marylebone, that your Lordships are here concerned with an aspect of the law of evidence, not discovery. Thirdly, I agree with my noble and learned friend on the Woolsack as to the inappropriateness of treating judgments as if they were instruments to be interpreted by application of the maxim " expressio unius . . . ".

Lord Kilbrandon

My Lords,

I have had the advantage of reading in draft the speech of my noble and learned friend Lord Hailsham of St. Marylebone. I am so entirely in agreement with his reasoning and conclusion that it would not be helpful if I were to attempt to state my own. I would allow this appeal on what has been termed the narrow ground.

Lord Edmund-Davies

My Lords,

It is a truism that, while irrelevant facts are inadmissible in legal proceedings in this country, not all inadmissible facts are irrelevant. To be received in evidence, facts must be both relevant and admissible, and under our law relevant facts may nevertheless be inadmissible. It is a serious step to exclude evidence relevant to an issue, for it is in the public interest that the search for truth should, in general, be unfettered. Accordingly, any hindrance to its seeker needs to be justified by a convincing demonstration that an even higher public interest requires that only part of the truth should be told.

The most unfortunate litigation giving rise to this appeal has been conducted on the basis that the plaintiff has a good cause of action against the defendant Society if she can establish the allegations she makes. It is also common ground that the identity of the person who made an unfounded complaint to the Society regarding the welfare of her child is relevant to that cause of action. But it is established in our law that the mere fact that information is imparted in confidence does not, of itself, entitle the recipient to refuse disclosure of the identity of the informer (Attorney-General v. Clough [1963] 1 Q.B. 773, per Lord Parker CJ. at 787; A. Crompton Ltd. v. Customs & Excise [1974] A.C. 405, per Lord Cross of Chelsea at 432H.) This is so even although the information takes the form of allegations of blameworthy conduct, and it led Professor Hanbury to observe that: "Few situations " in life are more calculated to arouse resentment in a person than to be

" told that he has been traduced, but cannot be confronted with his traducer. " It is submitted that, ideally, nothing but the very pressing demands of public " security, where the vital interests of the community are unquestionably " involved, can require that private individuals should be expected " to acquiesce in their vulnerability by an invisible foe. Yet this is precisely " the situation which may become common, if the principle of anonymity " of informers is to be treated as universal. . . . Does our law issue a " tacit invitation to any spiteful or eccentric busybody, such as Mr. Frankland, " in The Hound of the Baskervilles, to ransack the Statute-book, draw up " for his own guidance a list of offences, and watch for breaches of them, " sheltered by the assurance that his identity will never be revealed to his " victims? May not the common informer, now mercifully deprived of the " power of personal action, continue his career through this, as an alternative " form of activity? " (1952 68 L.Q.R., at p. 181.)

There are further ways in which injustice may result from the exclusion of relevant matter. Sir Rupert Cross has illustrated this by his comments on Hennessy v. Wright (188) 21 Q.B.D. 509 (see his " Evidence ", 4th Ed. 265). And, since a just conclusion is one arrived at after considering all relevant evidence, something far short of it would have been attained in Garner v. Garner (1920) 36 TLR. 196, where a wife petitioned for dissolution on the ground of her husband's cruelty in communicating syphilis to her, had the husband's doctor not been compelled by McCardie, J., to disclose that he had in fact treated him for that disease. In Ellis v. Home Office [19531 2 All E.R. 149 Devlin J., when upholding a Home Office claim against disclosure of documents on the ground that it would be against the public interest to do so, expressed the view that, whether or not justice had been done, it certainly would not appear to have been done.

Many other cases to the like effect could be cited. Accordingly, it would be unthinkable to vest the judiciary with a power to exclude in its discretion evidence relevant to the issues in civil proceedings merely because one side wants it kept out and the judge thinks that its disclosure is likely to prove embarrassing. In other words, the exclusion of relevant evidence always calls for clear justification. The importance of the present appeal lies in the clash between the widely differing submissions of the parties in civil actions as to the manner in which the courts should direct themselves where relevance is said to be at variance with admissibility.

I should serve no useful purpose by again reciting the facts or by embarking upon yet another review of the decided cases. And I am more than content to adopt the observations of my noble and learned friend, Lord Hailsham of St. Marylebone, regarding what were called the ' broad' and the " narrower " submissions advanced on behalf of the appellants and his reasons for rejecting the former. But, while I am also in agreement with his rejection of " the more restricted and even, occasionally, pedantic view of the authorities advanced on behalf of the respondent", I have unfortunate misgivings about his treatment of the 16th Report of the Law Reform Committee (1967 Cmnd. 3471). In the first paragraph its distinguished signatories referred to " the common law, whose policy, pragmatic as always, has been to limit to a minimum the categories of privileges which a person has an absolute right to claim, but to accord to the judge a wide discretion to permit a witness, whether a party to the proceedings or not, to refuse to disclose information where disclosure would be a breach of some ethical or social value and non-disclosure would be unlikely to result in serious injustice in the particular case in which it is claimed."

The only decisions cited in support of that " wide discretion " are Attorney- General v. Clough (ante) and Attorney-General v. Mulholland [1963] 2 Q.B.477. It is to be noted that in the latter Lord Denning, M.R., began (at p.489) by stressing that :

" The only profession that I know which is given a privilege from " disclosing information to a court of law is the legal profession, and " then it is not the privilege of the lawyer but of his client. Take the

" clergyman, the banker or the medical man. None of these is entitled " to refuse to answer when directed to by a judge."

Having made that matter perfectly clear, it is then only that the Master of the Rolls continued:

" Let me not be mistaken. The judge will respect the confidences " which each member of these honourable professions receives in the " course of it, and will not direct him to answer unless not only it is " relevant but also it is a proper and, indeed, necessary question in the " course of justice to be put and answered. A judge is the person " entrusted, on behalf of the community, to weigh these conflicting " interests—to weigh on the one hand the respect due to confidence in " the profession and on the other hand the ultimate interest of the " community in justice being done . . .If the judge determines that " the journalist must answer, then no privilege will avail him to refuse."

During the hearing of this appeal it was suggested that the Report demonstrated that it is no longer right to say that " the only profession . . . " which is given a privilege from disclosing information to a court of law is " the legal profession " and that a judge is, for example, nowadays always entitled to direct a doctor not to answer a question concerning his patient's health on the simple ground that disclosure was opposed by his patient. I know of no decided cases supporting such a view or of any text-book writers who indicate that such is the law. On the contrary, the writers are unanimous that only in the case of lawyers and their clients is the court so empowered; see Phipson (llth Ed., 1970, para.587), Cross (4th Ed., 1974, p.255); and Halsbury's Laws of England, 4th Ed., Vol.17 (1976) p.166. The Criminal Law Revision Committee similarly stated in its llth Report (1972), to which Sir Rupert Cross was a signatory, that "We have no doubt that the only kind of professional privilege which English law allows is that of legal adviser and client " (para.272). This is in accordance with the statement in para.41 of the 16th Report itself that:

" At Common law there exists no absolute privilege of non-disclosure " of communications made pursuant to a confidential relationship other " than that of client and lawyer which we have classified as a privilege " in aid of litigation."

No reported case supports the proposition advanced in para. 51 of that report that a judge is entitled to direct a doctor not to disclose information regarding his patient's health, save that where

" for instance, there was an issue as to whether the patient's mental " state was caused by the alleged cruelty of her husband or by anxiety " resulting from her own adultery, it may well be that the balance of " public interest will tilt in favour of her psychiatrist's being required " to answer a question about any admission of adultery she may have " made to him. But in the absence of some issue of this kind, we think " it likely that, if the psychiatrist had persisted in his refusal, the " Commissioner's exercise of his discretion in insisting on an answer " would have been reversed by the Court of Appeal."

I believe the law to be quite otherwise. The report of Nuttall v. Nuttall [1964] 108 Sol.J. 105 (to which para. 51 refers) is far too exiguous to entitle one to criticise the conduct of the Divorce Commissioner who there directed a doctor to answer questions about his patient's health. To assert, without more, that his direction " would [probably] have been reversed by the Court " of Appeal" is to put the doctor/patient relationship in the same class as that of lawyer/client, to which under our law it does not belong. If the law is to be altered (as I have long thought it should), that can be done only by a decision of this House, in a suitable case raising the issue, or by the legislature.

In the result, I believe that the law applicable to all civil actions like the present one may be thus stated:

(I) In civil proceedings a judge has no discretion, simply because what is contemplated is the disclosure of information which had passed between persons in a confidential relationship (other than that of lawyer and client), to direct a party to that relationship that he need not disclose that information even though its disclosure is (a) relevant to, and (b) necessary for the attain- ment of justice in the particular case. If (a) and (b) are established, the doctor or the priest must be directed to answer if, despite the strong dissuasion of the judge, the advocate persists in seeking disclosure. This is also true of all other confidential relationships in the absence of a special statutory provision, such as the Civil Evidence Act 1968, regarding communications between patent agents and their clients.

(II) But where(i) a confidential relationship exists (other than that of lawyer and client) and

(ii) disclosure would be in breach of some ethical cr social value involving the public interest, the court has a discretion to uphold a refusal to disclose relevant evidence provided it considers that, on balance, the public interest would be better served by excluding such evidence.

In conducting the necessary balancing operation between competing aspects of public interest, the presence (or absence) of involvement of the central government in the matter of disclosure is not conclusive either way, though in practice it may affect the cogency of the argument against dis- closure. It is true that in Blackpool Corporation v. Locker [1948] 1 K.B.349 the Court of Appeal dismissed a local authority's claim to exclude their interdepartmental communications in the public interest, Scott L.J. saying (at p. 380):

" No such privilege has yet, as I know, been conceded by the " courts to any local government officer when his employing authority is in " litigation "

. But it is worthy of note that he went on to observe that, although " Public interest is, from the point of view of English justice, a " regrettable and somewhat dangerous form of privilege, though at times " unavoidable, . . . no such ground was put forward in the plaintiff's " affidavit ". (The italics are mine.) We therefore cannot be sure how that case would otherwise have been decided, but we do know from Conway v. Rimmer 11968) A.C.910 and In re D [1970] 1 W.L.R.599 that an organ of central government does not now necessarily have to be involved before a claim for non-disclosure can succeed. In my judgment, Scarman L.J. therefore went too far in asserting in the Court of Appeal in the present case [1976] 3 W.L.R. at 139B) that, " State interest alone can justify the withholding of relevant " documents ". So to assert is, in the wise words of one commentator, " to " place too high a value on the arbitrary factor of the status of the possessor " of the information. It also assumes that organisations can be classified into " those which have the status of ' a central organ of government' . . . and " those who do not. Such a classification is surely impracticable." (Jacob, 1976 Public Law, 138).

The sole touchstone is the public interest, and not whether the party from whom disclosure is sought was acting under a " duty " —as opposed to merely exercising " powers ". A party who acted under some duty may find it easier to establish that public interest was involved than one merely exercising powers, but that is another matter.

(V) The mere fact that relevant information was communicated in confidence does not necessarily mean that it need not be disclosed. But where the subject-matter is clearly of public interest, the additional fact (if such it be) that to break the seal of confidentiality would endanger that interest will in most (if not all) cases probably lead to the conclusion that disclosure should be withheld. And it is difficult to conceive of any judicial discretion to exclude relevant and necessary evidence save in respect of confidential information communicated in a confidential relationship.

(VI) The disclosure of all evidence relevant to the trial of an issue being at all times a matter of considerable public interest, the question to be determined is whether it is clearly demonstrated that in the particular case the public interest would nevertheless be better served by excluding evidence

despite its relevance. If, on balance, the matter is left in doubt, disclosure should be ordered.

Croom-Johnson J. appears to have directed himself on the foregoing lines in the course of his admirable judgment in this case. He concluded that, " there is no doubt at all that the public interest in protecting the defendants' " sources of information overrides the public interest that Mrs. D. should " obtain the information she is seeking in order to obtain legal redress ". In my judgment, it cannot properly be said that the learned trial judge was disentitled to come to that conclusion. On the contrary, I think he was clearly right. I would therefore be for restoring his decision and allowing this appeal.

(306983) Dd 896296 140 1/77 St.S.

Cited in 51 later judgments

and 41 more of the most senior

Showing the 50 most senior of 51.