|[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]|
United Kingdom House of Lords Decisions
You are here: BAILII >> Databases >> United Kingdom House of Lords Decisions >> Saif Ali & Anor v Sydney Mitchell And Company (a firm) & Ors  UKHL 6 (2 November 1978)
Cite as:  UKHL 6, [1955-95] PNLR 151,  AC 198,  3 All ER 1033,  3 WLR 849
[New search] [Context] [Printable version] [Buy ICLR report:  AC 198] [Buy ICLR report:  3 WLR 849] [Help]
Thursday 2nd November, 1978
In Rondel v. Worsley  1 AC 191, this House decided that a barrister was immune from any action for professional negligence in respect of acts or omissions during the trial of criminal proceedings against his lay client. Now in this case it is necessary to decide whether the barrister's immunity covers pre-trial acts or omissions in connection with civil proceedings brought by his lay client.
The plaintiff, Mr. Saif Ali, a passenger in a motor car driven by his friend, Mr. Akram, was injured on 26th March 1966 in a collision with another car driven by Mrs. Sugden, to whose husband the car belonged. There is no doubt that Mrs. Sugden was to blame, and possibly wholly to blame. On the instructions of solicitors the barrister settled proceedings and drafted a pleading on behalf of Mr. Ali and Mr. Akram against Mr. Sugden. This was on the basis that as Mrs. Sugden was using the car to drive their children to school, Mr. Sugden was responsible for her negligence. Though, with hindsight, it might have been wise to sue Mrs. Sugden as well as her husband, the course adopted, to sue Mr. Sugden who was the insured party, was correct in law. Indeed it appeared at first that Mr. Sugden would not deny responsibility. However, when his insurers took charge of the matter, they suggested (a) that Mrs. Sugdcn's agency for Mr. Sugden might be disputed and (b) that a case of contributory negligence might be raised against Mr. Akram. The barrister was informed of this and instructions were sent to him to consider amendment of the pleading. By this time, 24th February 1969, such was the leisurely pace of proceeding, there was little time left before the three year period of limitation from the date of the accident (on the 26th March 1969) would expire. The barrister, however, so it is said in the third party notice, orally confirmed his advice that no amendment was necessary, on what date is not stated, and later, on 1st April 1969, advised in writing. When he did this the three year period had elapsed, and the advice was that it was too late for the plaintiff to sue Mr. Akram. It was also, of course, too late to sue Mrs. Sugden. Mr. Sugden in his first defence dated 16th October 1969 denied Mrs. Sugden's agency, but by an amendment in June 1971 admitted it. Later still (apparently in June 1972) he asked leave to re-amend so as to deny the agency. This seems to have been agreed to unconditionally by the plaintiff's solicitors, and leave was given. Later, on the advice, it is said, of leading counsel, proceedings against Mr. Sugden were dropped, so the plaintiff, who started with an impregnable claim for damages, found after 5 years that he had nobody he could sue. He therefore brought proceedings against his solicitors for negligence, and the solicitors brought third party proceedings against the barrister. Later, the plaintiff also brought direct proceedings against the barrister. This appeal is concerned with the third party proceedings only. The Court of Appeal has struck them out on the ground that the barrister is immune from suit.
It is important to see what is the precise negligence alleged. According to the amended third party notice the negligence consisted of:
(i) delaying until after the expiry of the limitation period to advise whether the proceedings should be resettled in view of the non-admission by Mr. Sugden that Mrs. Sugden was driving as his agent and the possible negligence of Mr. Akram.
(ii) failing to advise until a late stage that there might be a conflict of interest between the plaintiff and Mr. Akram.
(iii) failing to advise the plaintiff that he should take proceedings against Mr. Sugden and/or Mrs. Sugden and/or Mr. Akram and advising that proceedings should be issued against Mr. Sugden only.
For the purposes of this appeal it has to be assumed that the factual basis for these allegations (as set out above) is correct, that there was some degree of negligence on the barrister's part as regards one at least of the three matters, that such negligence resulted in damage and that the solicitors are entitled to indemnity or contribution from the barrister. All of these assumptions may turn out to be incorrect if the matter goes to trial, but cannot be challenged at this stage.
The question now for this House is whether on the assumptions stated the claim by the solicitors against the barrister is so clearly unfounded that it ought to be struck out. This involves a reconsideration of Rondel v. Worsley in order to see what rule of law is to be extracted from it.
Rondel v. Worsley gave rise to a restatement of the traditional principle of barristers' immunity in the light particularly of the decision of this House in Hedley Byrne & Co. Ltd. v. Heller & Partners Ltd.  AC 465. Previously an important if not the main reason for the immunity was supposed to lie in the fact that a barrister could not sue for his fees: this reason, if valid, would of course have thrown a blanket of immunity over all barristers' actions, in or out of court, whatever their nature. This House, however, in 1967 took the inevitable view that this reason no longer applied: liability for negligence might exist in the absence of a contract for reward. Nevertheless the immunity was held to exist on grounds, essentially, of public policy; mainly upon the ground that a barrister owes a duty to the court as well as to his client and should not be inhibited, through fear of an action by his client, from performing it; partly upon the undesirability of relitigation as between barrister and client of what was litigated between the client and his opponent. This necessarily involved a removal of the total blanket immunity and a restriction of it to such cases as might fall within the area of public policy.
Rondel v. Worsley was concerned and only concerned with matters taking place in court which resulted in an outcome unfavourable to the client. But the speeches contain considered observations as to the extent of barristers' immunity for matters taking place outside court and in barristers' chambers. Since the case was not concerned with such matters, these observations have the status of obiter dicta. However, not all obiter dicta have the same weight, or lack of weight, in later cases. Of those then made in the House two things may be said. First, they were considered and deliberate observations after discussion of the same matters had taken place in the Court of Appeal and in the light of judgments in the Court of Appeal. It may be true that the counsel in the case did not present detailed arguments as to the position outside the court room—they had no interest in doing so—but I cannot agree that this invalidates or weakens judicial pronouncements. Judges are more than mere selectors between rival views— they are entitled to and do think for themselves. Secondly, it would have been impossible for their Lordships to have dealt with the extent of barristers' immunity for acts in court without relating this to their immunity for other acts. As I shall shortly show their Lordships attached the immunity to the conduct of litigation. But litigation takes some time to arrive in court for trial, so unless they were prepared to confine the immunity to that part of litigation which occurs in the court room, it was not only appropriate but necessary to deal with such acts—in relation to litigation—as occur outside the court room. A statement of principle which stopped at the door of the court would have been truncated and irrational. These factors, in my opinion, tell in favour of giving considerably more weight to their Lordships' expressions of opinion than obiter dicta normally receive. We may clarify them, but we should hesitate before disregarding them.
This leads to another point. The general principle that barristers are entitled to some immunity was established, or re-established, by unanimous decision of all their Lordships. It was argued that barristers should enjoy no greater immunity than other professional men. But that argument was rejected: barristers, it was firmly held, have a special status, just as a trial has a special character: some immunity is necessary in the public interest, even if, in some rare cases, an individual may suffer loss. Now I would accept that the existence of a duty of care, and correspondingly of liability in negligence for failure to exercise that duty, continues in the natural course of legal evolution to expand as new situations come before the courts. But I do not think that this natural process which bears upon the existence of a duty of care should lead us to sweep away after so short a time an immunity from suit on special grounds of principle, which after many centuries of existence has been restated by this House. No ground was suggested why we should reopen the decision in Rondel v. Worsley and I do not think we should do so. What is required of us is a decision on the limits of an immunity held by this House to exist—a fringe decision rather than a new pattern. I will now consider the opinions.
Lord Reid considered that there was no doubt about the position of barristers appearing in court (p. 227). There he introduced the phrase
"their work in conducting litigation" (p. 231).
He made this more explicit when he said
"the "same public duty applies when drawing pleadings or conducting subsequent stages in a case as applies to counsel's conduct during the trial"
"the same "may apply at a stage when litigation is impending" (pp. 231-2).
It is clear from this that his Lordship was not seeking with any precision to define the limit of immunity but that his thinking was in terms of litigation—more broadly than of work in court.
Lord Morris' expressions were more restrictive:
"What is said or done in "the . . . management of a case in court" (p. 247) "relating only to the limited field of the conduct and management of a case in court" (p. 248).
But his Lordship quotes from the leading case of Swinfen v. Lord Chelmsford (a case not limited to action in court) the sentence
"no action will lie against counsel for any act honestly done in the conduct or management of the cause" (5 H. & N. 890, 923),
and he expressed agreement with the judgment of Salmon L.J. in the Court of Appeal which I think it fair to say drew the line between litigation and paper work. Lord Pearce's expressions were wider: there is no distinction
"between the liability of a barrister in litigation and in his other non-litigious work as a barrister" (p. 265).
He held that a counsel in giving opinions
"not only those on which an action is to be started or not started owes a duty to the court (p. 276).
Lord Upjohn held that the immunity covers litigation
"at all events in matters pertaining to litigation" (p. 281)
and that it must "start
"before counsel enters the doors of the court to conduct the case".
"He will have had to give fearlessly to his client advice on the prospects of success; he will have settled the pleading; and on discovery and "in his advice on evidence and on many other matters he may have had to refuse to adopt his client's wishes. As a practical matter, I do no more than suggest that the immunity of counsel in relation to litigation should start at (the) letter before action" (pp. 285-6).
Finally Lord Pearson used words of exclusion. He asked
"Does the barrister's immunity extend to 'pure paper work', that is to say, drafting and advisory work unconnected with litigation?" (p. 293).
My Lords, none of these expressions are precise, in the nature of things they could not be, but they show a consensus that what the immunity covers is not only litigation in court but some things which occur at an earlier stage, broadly classified as related to conduct and management of litigation. The spectrum of the opinions is a wide one: we are now concerned to narrow it.
In considering how far a barrister's immunity extends, it is necessary to disentangle three separate strands.
The first is that of privilege. This attaches to proceedings in court and protects equally the judge, counsel, witnesses, jurors and parties. It has nothing to do with a barrister's duty to his client. It is worth noting that the courts will not allow this privilege to be outflanked by basing a claim on statements made or agreed to be made out of court if these were clearly and directly made in relation to the proceedings in court: Watson v. M'Ewan  AC 480, Marrinan v. Vibart  1 Q.B. 528.
The second is that of the defences available to barristers. Much if not most of a barrister's work involves exercise of judgment—it is in the realm of art not science. Indeed the solicitor normally goes to counsel precisely at the point where, as between possible courses, a choice can only be made on the basis of a judgment, which is fallible and may turn out to be wrong. Thus in the nature of things, an action against a barrister who acts honestly and carefully is very unlikely to succeed. But this is not an argument for giving him total immunity from proceedings.
The third is that of immunity from an action, which depends upon public policy. In fixing its boundary, account must be taken of the counter policy that a wrong ought not to be without a remedy. Furthermore, if the principle is invoked that it is against public policy to allow issues previously tried (between the client and his adversary) to be relitigated between client and barrister, it may be relevant to ask why this principle should extend to a case in which by the barrister's (assumed) fault, the case never came to trial at all. Then two considerations show that the area of immunity must be cautiously defined.
How can this be done? "Conduct and management" is the expression which has emerged and no doubt this is not a sharp definition. I think that something more precise is required if immunity in respect of acts out of court is to be properly related to the immunity for acts in court. A helpful expansion of the phrase was suggested by McCarthy P. in the New Zealand Court of Appeal: I quote his words:
"I cannot narrow the protection to what is done in Court: it must be "wider than that and include some pre-trial work. Each piece of before trial "work should, however, be tested against the one rule; that the protection "exists only where the particular work is so intimately connected with the "conduct of the cause in Court that it can fairly be said to be a preliminary "decision affecting the way that cause is to be conducted when it comes to a "hearing. The protection should not be given any wider application than is "absolutely necessary in the interests of the administration of justice, and "that is why I would not be prepared to include anything which does not "come within the test I have stated." (Rees v. Sinclair  1 NZLR 180, 187).
I do not understand this formulation as suggesting an entirely new test, i.e. a double test requiring (a) intimate connection with the conduct of the cause in court and (b) necessity in the interests of the administration of justice. The latter words state the justification for the test but the test lies in the former words. If these words involve a narrowing of the test as compared with the more general words "conduct and management" I think that this is right and for that reason I suggest that the passage, if sensibly, and not pedantically, construed, provides a sound foundation for individual decisions by the courts, whether immunity exists in any given case. I should make three observations. First, I think that the formulation takes proper account, as it should, of the fact that many trials, civil and criminal, take place only after interlocutory or pre-trial proceedings. At these proceedings decisions may often fall to be made of the same nature as decisions at the trial itself: it would be illogical and unfair if they were protected in the one case but not in the other. Second, a decision that a barrister's liability extends so far as I have suggested necessarily involves that it does not extend beyond that point. In principle, those who undertake to give skilled advice are under a duty to use reasonable care and skill. The immunity as regards litigation is an exception from this and applies only in the area to which it extends. Outside that area, the normal rule must apply. Thirdly, I would hold that the same immunity attaches to a solicitor acting as an advocate in court as attaches to a barrister. Fourthly, it is necessary to repeat that the rule of immunity is quite distinct from the question what defences may be available to a barrister when he is sued. It by no means follows that if an error takes place outside this immunity area, a liability in negligence arises.
Finally, as to the present case. The question is whether the third party claim should be allowed to go to trial, or whether it should be held that it falls within the area of immunity so as to justify striking out at this stage. In the Court of Appeal Lord Denning, M.R. and Lawton L.J. held that the acts and omissions complained of came within the general words "conduct and management of litigation". Bridge L. J. held that they came within the narrower test of Rees v. Sinclair.
My Lords, I think that the narrower test is the correct one, and I do not consider that the acts complained of come within it so as to justify striking out. An oversight, or failure to consider the consequences of not adding Mrs. Sugden a defendant before the limitation period expired, if such took place, may have been defensible, but in my opinion falls well outside the immunity area. I would allow the appeal.
The decision of this House in Hedley Byrne & Co. Ltd. v. Heller &PartnersLtd.  AC 465 cast doubt upon the facile explanation, which had been current for a hundred years, that a barrister's immunity from liability for economic loss sustained by a client in consequence of his incompetent advice or conduct, was due to his incapacity as counsel to enter into a contractual relationship with his client. In 1967 these doubts were tested in your Lordships' House in Rondel v. Worsley  1 AC 191 and the explanation, which would have covered all work undertaken as a barrister however remote from litigation it might be, was rejected as legal folk lore. If the immunity in respect of any part of his professional work was to be maintained, some other legal justification would be needed for it.
In Rondel v. Worsley the barrister, Mr. Worsley, had accepted a dock brief at the Old Bailey on behalf of Mr. Rondel, as he was bound to do. So there was no solicitor instructing counsel and no allegations of negligence save in the actual conduct of the case in court. The absence of any substance or merit in the charges of negligence made by Mr. Rondel and the fact that the "cab rank" principle had actually operated in his case lent point to the argument that unless counsel remained immune from liability for his conduct of a case in court, he would be exposed to the risk of baseless and vexatious actions for negligence on the part of disappointed clients, to whom he had no option to deny his professional services.
Rondel v. Worsley came before this House upon a summons to strike out the statement of claim as disclosing no reasonable cause of action. The only matter for decision was whether a barrister could be liable in negligence to his lay client for the way in which he had conducted the client's defence in court at his trial upon a criminal prosecution. The argument, however, ranged widely and the opinions expressed in the majority of the speeches were not confined to criminal cases nor to the actual conduct of a case in court. They referred also to civil cases and to work done out of court. Although expressed in somewhat different terms in individual speeches the highest common factor to be discerned in them is in my view accurately stated in the headnote as follows:—
"a barrister was immune from an action for negligence at the suit of a client in respect of his conduct and management of a cause in court and the preliminary work connected therewith such as the drawing of pleadings".
In the instant case the negligence alleged against the barrister has been stated by my noble and learned friend Lord Wilberforce. In substance what it amounts to is that in a very common kind of running down action he gave negligent advice as to who should be joined as defendant to his client's claim for damages and settled the pleadings in conformity with that erroneous advice. It thus falls within the words that I have italicised in the headnote to Rondel v. Worsley. The statement of law expressed in them was not necessary to the actual decision in that case.
In deciding whether propositions of law expressed in speeches of members of this House on the determination of an appeal are binding upon lower courts the question whether they form part of the ratio decidendi of the majority or are mere obiter dicta is crucial. Propositions that fall into the former class are binding: those in the latter are persuasive only. In this House, however, since the Practice Direction of 1966, all propositions of law laid down in the speeches in previous appeals are persuasive only, whether they constituted an essential logical step in the author's reasons for disposing of the appeal in the way that he proposed, and so formed part of his ratio decidendi or, though not regarded by him as necessary for that purpose, were included as a helpful guide to judges in the disposition of future cases and so were obiter dicta. Leaving aside all such invidious distinctions as depend upon the reputations which individual former members of the Appellate Committee may have gained as jurists, the persuasive value of propositions of both classes depends to a considerable extent upon the course followed in the arguments presented to the court in the particular case in which they were laid down.
In Rondel v. Worsley it was to the interest of neither party to the case to argue that any distinction was to be drawn between the liability of a barrister for negligence in that part of his work that is done in the court itself and work that he does out of court. In the dialectic between counsel this played no active part.
It was the majority of the members of this House who, in seeking a new rationale that would justify retention of a barrister's immunity from liability for negligence in doing the kind of thing that Mr. Worsley had been charged with doing negligently in a criminal trial, themselves suggested a distinction between what a barrister does in the conduct and management of litigation and his non-litigious work. In respect of the former all five members of this House were of opinion that notwithstanding that a barrister's immunity could no longer be based upon his incapacity to enter into a contract for the provision of his professional services, nevertheless the policy of the law required that the immunity should be maintained on other grounds. In respect of work not directly connected with the conduct of a case in court four members (Lord Pearce dissenting) expressed opinions that he would be liable: but they were not of one mind as to where the dividing line lay between what work attracted immunity from liability for negligence and what work did not. At the one extreme Lord Pearson limited himself to the expression of a doubt as to whether the barrister's immunity extended to "pure paper work", which he explained as drafting and advisory work unconnected with litigation. At the other extreme Lord Morris of Borth-y-Gest appears to have regarded the immunity as confined to the actual conduct of a case in court. He suggests no extension beyond this. Between these two extremes Lord Reid, in a passage clearly intended to be obiter, expressed the view that the immunity would extend to drawing pleadings or conducting subsequent stages in a case, as it applies to counsel's conduct during the trial; while Lord Upjohn, also obiter, suggested that the immunity of counsel, which he regarded as confined to the conduct of litigation in and out of court should start in a civil case at the moment when the letter before action was sent.
My Lords, in recognising a barrister's immunity from liability for negligence in the conduct of his professional work of a particular kind this House was granting to the Bar a privileged status which the common law does not accord to members of any other profession or skilled craft. Those who hold themselves out as qualified to practise other professions, although they are not liable for damage caused by what in the event turns out to have been an error of judgment on some matter upon which the opinions of reasonably informed and competent members of the profession might have differed, are nevertheless liable for damage caused by their advice, acts or omissions in the course of their professional work which no member of the profession who was reasonably well- informed and competent would have given or done or omitted to do.
This exceptional immunity of the barrister and its extension to various kinds of professional work that he does outside the court room can no longer be justified as the automatic corollary of the rule that a barrister cannot contract to render professional services. In my view, it must be justified, if at all, as a matter of policy of the law and by the special characteristics of the kind of work to which the immunity applies that distinguish it from professional work undertaken by members of other professions.
The general trend in the policy of the law as developed by your Lordships' House in recent years has been to extend to new areas of activity the notion that a man is liable for loss or damage to others resulting from his failure to take care. Hedley Byrne itself marked an important milestone in this development; and the indication of their view by the majority of this House in Rondel v. Worsley that at any rate some kinds of work done by a barrister would no longer attract immunity from liability for negligence was another, if hesitant, step along the same road. During the years that have passed since Rondel v. Worlsey was decided, the extension of liability for negligence in doing things that were not previously regarded as giving rise to any legal duty of care has gone on apace. A few examples serve to show how broad this trend has been. Architects have been held liable for negligence in valuing work for the purposes of certificates of interim payments under building contracts: Sutcliffe v. Thackrah  A.C. 727; accountants for negligence when acting as valuers for the purpose of a contract between other parties; Arenson v. Arenson  A.C. 405; building inspectors employed by local authorities for negligence in inspecting the foundations of a building in course of erection: Anns v. Merton Borough Council  2 W.L.R. 1024; Borstal officers, for negligent failure to control their charges: Dorset Yacht Co. Ltd. v. Home Office  AC 1004; and professional salvors have been held liable for negligence in carrying out salvage operations: The Tojo Maru  A.C. 242. The extension to the duty of care to trespassers to land that was made in Herrington v. British Railways Board  AC 877 illustrates the existence of a similar general trend extending beyond the limited field of professional work.
In the face of this trend it would in my view be hard to justify founding the decision of the instant appeal upon an uncritical acceptance of the highest common factor in the observations of the majority of the members of this House who spoke upon the subject in Rondel v. Worsley as defining the kind of work done by a barrister outside the courtroom door in respect of which he is immune from liability for negligence. What is needed is to identify those reasons based on public policy which were held to justify a barrister's immunity from liability for negligence for what he did in court during the trial of a criminal case and, having done so, to decide whether they suffice to justify a like immunity when advising a client, through his solicitor, as to who should be made a party to a proposed civil action and when settling pleadings in the action in conformity with that advice.
There were several reasons given in Rondel v. Worsley for distinguishing between the work done by a barrister in the conduct of a criminal trial in court and work done by members of any other profession, so as to entitle the former to an exemption from liability for negligence which no other type of professional work enjoyed.
The special characteristic of a barrister's work upon which the greatest stress is laid by their Lordships was that he does not owe a duty only to his client; he owes a duty also to the court. This is an over-riding duty which he must observe even though to do so in the particular case may appear to be contrary to the interests of his client. Furthermore a barrister has to exercise his judgment as to where the balance lies between these competing duties immediately and without opportunity for calm reflection as the trial inexorably proceeds. His ability to give his best service to the court and to his client, it is said, would be diminished if he were compelled continually to give consideration to the possible effects that the way in which he exercised that judgment might have upon his own liability to his client for negligence.
To say of a barrister that he owes a duty to the court, or to justice as an abstraction, to act in a particular way in particular circumstances may seem to be no more than a pretentious way of saying that when a barrister is taking part in litigation he must observe the rules; and this is true of all who practise any profession. The rules which may appear to conflict with the interests of the client are simple to state, although their application in border-line cases may call for a degree of sophistry not readily appreciated by the lay client, particularly one who is defendant in a criminal trial. A barrister must not wilfully mislead the court as to the law nor may he actively mislead the court as to the facts; although, consistently with the rule that the prosecution must prove its case, he may passively stand by and watch the court being misled by reason of its failure to ascertain facts that are within the barrister's knowledge. Again, although he must not abuse the privilege which the law accords to him as counsel in rendering him immune from liability for aspersions which he makes against anyone in the course of litigation, however unfounded, irrelevant or malicious they may be, questions of considerable nicety may arise as to what constitutes sufficient foundation or relevance to justify the particular aspersion which his client wants him to make.
The fact that application of the rules that a barrister must observe may in particular cases call for the exercise of finely balanced judgments upon matters about which different members of the profession might take different views, does not in my view provide sufficient reason for granting absolute immunity from liability at common law. No matter what profession it may be, the common law does not impose on those who practise it any liability for damage resulting from what in the result turn out to have been errors of judgment, unless the error was such as no reasonably well-informed and competent member of that profession could have made. So too the common law makes allowance for the difficulties in the circumstances in which professional judgments have to be made and acted upon. The salvor and the surgeon, like the barrister, may be called upon to make immediate decisions which, if in the result they turn out to have been wrong, may have disastrous consequences. Yet neither salvors nor surgeons are immune from liability for negligent conduct of a salvage or surgical operation; nor does it seem that the absence of absolute immunity from negligence has disabled members of professions other than the law from giving their best services to those to whom they are rendered.
My Lords, the argument founded upon the barrister's competing duties to court and client, upon which this House so strongly relied in Rondel v. Worsley, loses much of its cogency when the scene of the exercise of the barrister's judgment as to where the balance lies between these duties is shifted from the hurly-burly of the trial to the relative tranquillity of the barrister's chambers. The kind of judgment which a barrister has to exercise in advising a client as to who should be made defendant to a proposed action and how the claim against him should be pleaded, if made with opportunity for reflection, does not seem to me to differ in any relevant respect from the kind of judgment which has to be made in other fields of human activity, in which prognosis by professional advisers plays a part. If subsequently a barrister is sued by his own client for negligence on what he advised or did in the particular case, he has the protection that the judge before whom the action for negligence against him will be tried is well qualified, without any need of expert evidence, to make allowance for the circumstances in which the impugned decision fell to be made and to differentiate between an error that was so blatant as to amount to negligence and an exercise of judgment which, though in the event it turned out to have been mistaken, was not outside the range of possible courses of action that in the circumstances reasonably competent members of the profession might have chosen to take.
In Rondel v. Worsley some reliance was also placed upon the "cab-rank" principle as distinguishing the Bar from all other professions. A barrister is not allowed by the rules of his profession to pick and choose between clients on whose behalf he will accept instructions. If he is disengaged and a proper fee is tendered to him, he is bound to accept instructions to act on behalf of any client desirous of his services in a field of law in which he holds himself out as practising. The "cab-rank" principle was a reality in Rondel v. Worsley. Mr. Worsley was instructed directly by the lay client; he was the recipient of a dock brief. But with the virtual disappearance of the dock brief the effect of the cab-rank principle is limited to preventing a barrister from refusing from a solicitor instructions in a field of law within which he practises simply because he does not like the solicitor or the solicitor's client or the nature of a lawful claim or ground of defence of which that client wishes to avail himself. I doubt whether in reality, in the field of civil litigation at any rate, this results often in counsel having to accept work which he would not otherwise be willing to undertake. But even if there are rare cases where it does, this does not seem to me to affect the character of the decisions that the barrister has to make in carrying out instructions that he receives through the client's solicitor. True it is that he may be obliged to accept instructions on behalf of an obstinate and cantankerous client who is more likely than more rational beings to bring proceedings for negligence against his counsel if disappointed in the result of his litigation; but the existence of this risk does not, in my view, justify depriving all clients of any possibility of a remedy for negligence of counsel, however elementary and obvious the mistake he has made may be. There are other and more specific means of disposing summarily of vexatious actions.
In the light of the developments of the law of negligence which have taken place since 1967,1 could not readily find today in the reasons that I have so far discussed convincing ground for holding that a barrister ought to be completely immune from liability for negligence for what he does in court in conducting criminal or civil proceedings—let alone for anything that he does outside court in advising about litigation whether contemplated or pending, or in settling documents for use in litigation.
There are, however, two additional grounds referred to in some of the speeches in Rondel v. Worsley which can be used to supplement those reasons so far as they protect a barrister from liability in respect of the way in which he has conducted proceedings in court, including in this expression interlocutory proceedings before the master or in chambers; save to a very limited extent, however, neither of them would apply to work done out of court.
The first is that the barrister's immunity from liability for what he says and does in court is part of the general immunity from civil liability which attaches to all persons in respect of their participation in proceedings before a court of justice; judges, court officials, witnesses, parties, counsel and solicitors alike. The immunity is based on public policy, designed, as was said by Lord Morris of Borth-y-Gest (p. 251), to ensure that trials are conducted without avoidable stress and tensions of alarm and fear in those who have a part to play in them. As was pointed out by Starke J. in Cabassi v. Vila  64 C.L.R. 130, a case in the High Court of Australia, "The law protects witnesses and others, not for "their benefit but for a higher interest, namely, the advancement of public "justice". The courts have been vigilant to prevent this immunity from indirect as well as direct attack—for instance by suing witnesses for damages for giving perjured evidence or for conspiracy to give false evidence; Marrinan v. Vibart  1 Q.B. 528. In Watson v. M'Ewan  AC 480, this House held that in the case of witnesses the protection extended not only to the evidence that they give in court but to statements made by the witness to the client and to the solicitor in preparing the witness's proof for the trial; since, unless these statements were protected, the protection to which the witness would be entitled at the trial could be circumvented.
The second reason is also based upon the need to maintain the integrity of public justice. An action for negligence against a barrister for the way in which he has conducted a case in court is founded upon the supposition that his lack of skill or care has resulted in the court having reached a decision that was not merely adverse to his client as to liability or quantum of damages but was wrong in being adverse and in consequence was unjust, for otherwise no damage could be shown to have resulted from the barrister's act or omission of which complaint is made. The client cannot be heard to complain that the barrister's lack of skill or care prevented him from obtaining a wrong decision in his favour from a court of justice. So he must prove that if the action had been conducted competently by his counsel he would have succeeded instead of failed.
Under the English system of administration of justice, the appropriate method of correcting a wrong decision of a court of justice reached after a contested hearing is by appeal against the judgment to a superior court. This is not based solely on technical doctrines of resjudicata but upon principles of public policy, which also discourage collateral attack on the correctness of a subsisting judgment of a court of trial upon a contested issue by re-trial of the same issue, either directly or indirectly in a court of co-ordinate jurisdiction. Yet a re-trial of any issue decided against a barrister's client in favour of an adverse party in the action in respect of which allegations of negligent conduct by the barrister are made would be an indirect consequence of entertaining such an action.
The re-trial of the issue in the previous action, if it depended on oral evidence, would have to be undertaken de novo. This would involve calling anew after a lapse of time witnesses who had been called at the previous trial and eliciting their evidence before a different judge by questions in examination and cross- examination that were not the same as those that had been put to them at the previous trial. The circumstances in which the barrister had made decisions as to the way in which he would conduct the previous trial, and the material on which those decisions were based, could not be reproduced in the re-trial; a^d the initial question in the action for negligence: whether it has been established that the decision adverse to the client reached by the court in the previous trial was wrong, would become hopelessly entangled with the second question: whether it has been established that notwithstanding the differences in the circumstances in which the previous trial was conducted, it was the negligent act or omission of the barrister in the conduct of his client's case that caused the wrong decision by the court and not any other of those differences.
My Lords, it seems to me that to require a court of co-ordinate jurisdiction to try the question whether another court reached a wrong decision and, if so, to inquire into the causes of its doing so, is calculated to bring the administration of justice into disrepute. Parliament indeed itself stepped in to prevent a similar abuse of the system of justice by convicted criminals in bringing civil actions for libel against those who described them as having been guilty of the crimes of which they had been convicted. See Civil Evidence Act 1968, section 13. A consequence of the decision of this House in Rondel v. Worsley was to prevent its happening in actions for negligence against barristers.
A similar objection, it may be mentioned, would not apply in cases where an action has been dismissed or judgment entered without a contested hearing, and there is no possibility of restoring the action and proceeding to a trial. If the dismissal or the entry of judgment was a consequence of the negligence of the legal advisers of a party to the action, a claim in negligence against the legal advisers at fault does not involve any allegation that the order of the court which dismissed the action or entered judgment was wrong.
I find it an unsatisfactory feature of the instant appeal, which has called for a re-examination of the speeches in Rondel v. Worsley in the light of the subsequent development of the law of negligence by later decisions of this House, that your Lordships have not had the benefit of any argument from counsel in support of a more radical submission that the immunity of the advocate, whether barrister or solicitor, for liability for negligence even for what he says or does in court ought no longer to be upheld. Counsel cannot be blamed for this. The parties whom they represent are solicitors and a barrister respectively. It is not to their interest as members of either branch of the legal profession to argue that this immunity no longer exists. Nevertheless, despite this handicap, I have reached the clear conclusion that these two additional grounds of public interest which I have just discussed, when taken with those reasons upon which greater stress had been placed in most of the speeches in Rondel v. Worsley, suffice to justify your Lordships in accepting as a premise for the purpose of deciding the instant appeal that the decision of this House in Rondel v. Worsley upholding such immunity is still good law.
The two additional grounds of public policy for granting a barrister immunity for what he does in court apply with equal force to what a solicitor does when acting as advocate in those courts in which solicitors have rights of audience; but subject to what is said below neither of them applies to what a barrister does outside court in advising about litigation or settling documents for use in litigation. Without the support of those additional grounds of public interest, as I have already indicated, I can find no sufficient reason for extending the immunity to anything that a barrister does out of court; save for a limited exception analogous to the extension of a witness's protection in respect of evidence which he gives in court to statements made by him to the client and his solicitor for the purpose of preparing the witness's proof for trial. The extent of this exception was in my view well expressed by the President of the Court of Appeal of New Zealand (where the profession is a fused one) in Rees v. Sinclair (1974) 1 NZLR 180 at page 187:
"Each piece of before-trial work should,... be tested against the one rule; that the protection exists only where the particular work is so intimately connected with the conduct of the cause in Court that it can fairly be said to be a preliminary decision affecting the way that cause is to be "conducted when it comes to a hearing. The protection should not be given any wider application than is absolutely necessary in the interests of the administration of justice ...".
So for instance in the English system of a divided profession where the practice is for the barrister to advise on evidence at some stage before the trial his protection from liability for negligence in the conduct of the case at trial is not to be circumvented by charging him with negligence in having previously advised the course of conduct at the hearing that was subsequently carried out.
It would not be wise to attempt a catalogue of before-trial work which would fall within this limited extension of the immunity of an advocate from liability for the way in which he conducts a case in court.
The work which the barrister in the instant case is charged with having done negligently, viz. in advising as to who was to be a party to an action and settling pleadings in accordance with that advice, was all done out of court. In my view, it manifestly falls outside the limited extension of the immunity which I have just referred to.
It follows that in my view the third party proceedings ought not to have been struck out upon the grounds stated in the judgments in the Court of Appeal. Whatever other grounds there might have been for doing so have not been relied upon by the respondents. So, the consequence must be that the order of the Court of Appeal should be reversed, and the appeal allowed.
It should not be thought, however, that I am assenting to the proposition that the facts stated in the claim in the Third Party proceedings disclose a reasonable cause of action by the solicitors against the barrister. That is a matter that will fall to be decided at the trial—if there is one.
I would accordingly allow this appeal.
This appeal raises a point of law of great public importance, namely—what is the extent of a barrister's immunity (if any) against a claim for damages for negligence in the performance of his professional duties out of court. It is a pity that such an important point falls to be decided in a case such as the present since its facts are somewhat tenuous and unsatisfactory.
On the 26th March 1966 Mr. Saif Ali was travelling as a passenger in a van being driven by his friend Mr. Akram. The van collided with a motor car which belonged to Mr. Sugden and was being driven by his wife; she was taking their children to school. As a result of the accident (a) Mrs. Sugden was prosecuted in October 1966 for driving without due care and attention, and she pleaded guilty to that offence, and (b) Mr. Ali and Mr. Akram suffered serious injuries and were away from work for many months. Mr. Ali had an unanswerable claim for substantial damages, certainly against Mrs. Sugden, probably against Mr. Sugden and possibly against Mr. Akram.
Soon after the accident, Mr. Ali and Mr. Akram consulted a firm of solicitors. In October 1968 those solicitors laid the facts which I have recited before a barrister and instructed him to settle proceedings and to advise. The barrister promptly settled a writ and a statement of claim making Mr. Ali and Mr. Akram joint plaintiffs and Mr. Sugden the sole defendant. The solicitors issued the writ on the 14th November 1968, but did not serve the writ or the statement of claim until the 29th August 1969. in the meantime, but well before the 26th March 1969 (when the period of limitation expired) the solicitors informed the barrister, amongst other things, that the solicitor acting for Mr. Sugden's insurers apparently in the course of negotiation had stated that Mr. Akram should be joined as a defendant and that the allegation in the statement of claim that Mrs. Sugden had been driving as her husband's agent might be put in issue. The solicitors asked the barrister for his advice. The barrister then advised that there was no foundation for any challenge of the allegation that Mrs. Sugden was driving as her husband's agent; and that accordingly no amendment to the statement of claim would be worthwhile. In accordance with that advice, the statement of claim was not amended by adding either Mrs. Sugden or Mr. Akram as defendants.
The barrister's advice not to join Mrs. Sugden and Mr. Akram as defendants and his failure to do so is the negligence which is alleged against him in circumstances to which I shall presently refer.
In November 1969 the barrister advised Mr. Ali that he should be separately represented; and Mr. Ali instructed new solicitors and a new barrister to act for him.
Mr. Sugden's original defence, served on the 16th October 1969, denied that his wife had been driving as his agent at the time of the accident. On the 24th June 1971 an amended defence was served, withdrawing the denial and admitting that Mrs. Sugden had been driving as her husband's agent.
On the 21st January 1972 a consent order was made giving leave to Mr. Sugden to re-amend his defence by withdrawing the admission made in the amended defence and reverting to the stance he had taken in the original defence. The re-amended defence was duly served on the 24th January 1972. It seems strange to me that Mr. Ali's then solicitors should have consented to the order allowing the amended defence admitting Mrs. Sugden's agency to be re-amended by denying her agency. With her agency admitted, and Mrs. Sugden having pleaded guilty to driving without due care and attention, Mr. Ali must have succeeded in his action against Mr. Sugden. It was particularly important for him that he should do so as it was then years too late for Mrs. Sugden or Mr. Akram to be joined as defendants.
On the 22nd April 1974 Mr. Ali's then solicitors served notice of the discontinuance of Mr. Ali's action against Mr. Sugden. We have been told from the Bar that the notice of discontinuance was served on the advice of leading counsel. It is possible that this advice was founded on the decision of your Lordships' House in Morgans v. Launchbury  AC 127. In that case Mr. Morgans had left his wife at home for the evening and at 10.40 p.m., when the accident occurred, he was in her car on what has been described as a "pub crawl" which had then been going on for about 4 hours. This House decided that, in these circumstances, it was impossible that he was using the car on behalf of his wife or for any purpose of hers; and that accordingly his wife was not responsible for the negligent driving of her car. The crucial difference between the relevant facts in that case and those in Mr. Ali's case is surely obvious.
The present appeal concerns a simple running down action in which Mr. Ali, if properly advised, must have recovered judgment, or settled his claim, for substantial damages against Mr. Sugden, Mrs. Sugden and Mr. Akram or one or more of them. As it is, after inordinate delays, whilst he had left himself in the hands of his lawyers and followed their advice for about 8 years, he now finds himself barred in law from taking his case to court against any of the defendants whose negligence caused him damage, and accordingly he is deprived of any of the damages to which he was clearly entitled.
It would, in my opinion, be a shocking reflection on the common law if, in the melancholy circumstances I have recited, Mr. Ali has no remedy against any of his advisers who are responsible for his present situation. It may be that the solicitors, having accurately instructed counsel about the facts, cannot be held to be negligent for having acted in accordance with counsel's advice. I cannot, however, find any reason or principle or sound authority to justify counsel's immunity from being sued for damages by clients who have suffered loss as a result of counsel's negligent advice. I have no doubt that, for the reasons I shall presently explain, the common law does give Mr. Ali a remedy against his advisers—whether solicitors or counsel—whose advice negligently caused his loss.
In September 1974 Mr. Ali brought an action against his original solicitors for damages for negligently advising him that neither Mrs. Sugden nor Mr. Akram ought to be added as defendants, and for negligently failing to make either of them defendants. No claim for damages for negligence was made against the solicitor who consented to the order giving leave to Mr. Sugden to re-amend his defence and who later discontinued the action against Mr. Sugden; nor was any claim made against leading counsel who advised that the action should be discontinued.
The original solicitors entered a defence denying negligence to the claim against them and issued third party proceedings claiming an indemnity against the barrister who advised them in 1968 mainly on the ground that he had negligently advised that neither Mrs. Sugden nor Mr. Akram should be joined as defendants. Mr. Ali then joined the barrister as a defendant and claimed damages for negligence against him. The barrister applied to have the third party claim against him struck out on the grounds that as a barrister he was immune from the claim in negligence made against him in the third party proceedings. The District Registrar struck out the third party claim but Kerr J. restored it. The Court of Appeal allowed the appeal from Kerr J.'s order and ordered that the third party proceedings should be struck out on the ground that the barrister was immune from any such claim. The present appellants now appeal from that decision to your Lordships' House.
I hope that nothing in this speech will leave an impression that I hold a view, one way or another, as to whether the barrister who advised in 1968 was negligent. I have certainly formed no view on this issue; it is an issue which, if this appeal is allowed and the action is fought, will have to be decided by the judge who hears the evidence. It has, in my opinion, been rightly conceded at the Bar that, as the facts alleged in the third party claim are capable of constituting negligence, the only issue before this House is whether or not the barrister is immune from the claim made against him.
A great deal of reliance has been placed by both sides on Rondel v. Worsley  1 AC 191. It is, I think, important to remember that this authority decided only one point, albeit a point of great importance, namely that a barrister, like a judge, juryman and witness, is immune from being sued in respect of anything he says or does or omits to say or do in the course of performing his role in court. This was the first occasion when this point had come before your Lordships' House for decision but there was much long established and powerful authority to support the decision at which this House arrived: see for example Swinfen v. Lord Chelmsford  5 H & N 890; Kennedy v. Broun  13 C.B. (N.S.) 677. Each of their Lordships fully explained why public policy required the immunity which they proclaimed in respect of anything said or done in court. With respect, I entirely agree with all that was said on that point in this House; and it is unnecessary for me to repeat anything I said when Rondel v. Worsley was heard in the Court of Appeal  1 Q.B. 516-520. I would only add that, in my view, a solicitor acting as an advocate in court enjoys exactly the same immunity as a barrister.
The decision in Rondel v. Worsley is, however, almost as irrelevant to the present appeal as the question which this appeal raises was irrelevant in Rondel v. Worsley. In that case, Mr. Rondel brought an action for damages for negligence against a barrister, Mr. Worsley, who had defended him, on a dock brief, for causing bodily harm with intent to do so. Mr. Worsley applied to have the claim struck out. When this application came before the judge in chambers, the plaintiff admitted inflicting the shocking injuries in respect of which he had been convicted. His case was that he had inflicted them in self-defence. His only real complaint against his counsel was that he had not cross-examined the Crown witnesses to show that the plaintiff had inflicted those injuries with his teeth and bare hands rather than with a knife. The plaintiff also admitted to the judge in chambers that he was not alleging that had counsel cross-examined the witnesses as he had wished, he would have had any chance of being acquitted. It followed from that admission that the plaintiff had suffered no damage from the line of cross-examination which counsel had pursued. Since the claim by Mr. Rondel against his counsel was based on negligence and not on contract, and proof of damage is an essential ingredient of the tort of negligence, Mr. Rondel's claim was clearly unmaintainable. The learned judge might well have dismissed it, briefly, as disclosing no cause of action and as an abuse of the process of the court. He, however, took the view that the case involved an important point of law; and he delivered a long and learned judgment in open court reviewing all the authorities back to the Year Books and explaining why public policy demanded that counsel and also solicitors should enjoy complete immunity from an action alleging negligence in respect of their conduct of a case in court. A conclusion with which I agree. I would point out that the learned judge, in my view, quite rightly confined himself to immunity in respect of what was said or done by advocates in court: he did not express any views as to whether such immunity could extend to paper work done by counsel out of court—and for the very good reason that this issue was irrelevant to the case he was deciding.
When Rondel v. Worsley reached the Court of Appeal, that court unanimously decided that a barrister was in law immune against any claim in negligence relating to what he had said or done in court. The majority, however, considered that the court should express its view on the issue which seemed to the judge in chambers and to the minority in the Court of Appeal to be irrelevant. The majority expressed the view very clearly, that a barrister enjoyed a blanket immunity in respect of any work which he did out of court. See Lord Denning M.R. at page 506 and Dahckwerts L.J., pages 512-513.1 entirely disagreed and still disagree on that point. I need not repeat any of the reasons which I gave for dissenting, which are fully set out at pp. 521-526
I recognise that it is most unpleasant for a barrister to have to fight an allegation that he has been negligent, but such an experience is no more unpleasant for a barrister than it is for a physician or a surgeon, an architect or an accountant. I cannot understand how there can be any justification for the law affording a blanket immunity to a barrister in respect of all work done out of court when it affords none to the members of any other profession; nor do I believe that the Bar would wish to claim such an immunity.
When Rondel v. Worsley came to this House, this House faced a dilemma. The Law Lords did not agree with the majority of the Court of Appeal which had decided, obiter, that a barrister enjoyed a blanket immunity against any claim in negligence in respect of all paper work. It was indubitably plain to this House that the obiter dictum of the majority of the Court of Appeal, although not binding, would carry great weight. Indeed it was extremely doubtful that any judge of first instance or any division of the Court of Appeal would depart from that obiter dictum unless this House disagreed with it. Accordingly, this House had no real choice but to deal with it. And this they did. By a majority of 4 to 1 the Law Lords rejected the proposition that the Bar enjoyed the blanket immunity proclaimed by the majority of the Court of Appeal in respect of all paper work done by a barrister. They all considered that some paper work should be covered by immunity, but they differed from each other as to where the line ought to be drawn.
In the present case, the question for decision is—on what side of the line does the advice fall which was given by the barrister against whom the third party proceedings have been brought?
It must be remembered that although all the four differing opinions in Rondel v. Worsley carry great weight and command profound respect, each of them is obiter. It should also be remembered that, during the period of almost 11 years since Rondel v. Worsley was decided, there has been a strong tendency for your Lordships' House to extend the ambit of the duty of care in negligence cases and to cut down the immunity enjoyed by professional men from being sued in actions for negligence. I shall refer only to the latter class of case.
It used to be thought that an architect employed by a building owner to supervise the erection of a building and to certify the sums due to the contractor, enjoyed an immunity against being sued for negligence however negligent he may have been in issuing his certificates; this immunity strangely enough was based on the fact that he owed a duty to the building owner and to the contractor to act impartially: see Chambers v. Goldthorpe  1 K.B. 624. That authority was however overruled and the immunity of architects was abolished by the unanimous decision of your Lordships' House in Sutcliffe v. Thackrah  A.C. 727.
Similarly, it was formerly held that if an accountant was called in to assess the value of shares knowing that the price which A. might pay B., or B. might accept for them, was to be the price at which the accountant valued them, then however negligently he may have conducted his valuation, he was immune from being sued either by A. or B. in respect of his negligence: Finnegan v. Allen  K.B. 425. That supposed immunity accorded to accountants was swept away by the unanimous decision of your Lordships' House in Arenson v. Arenson  A.C. 405.
There are several excellent reasons to which I have already referred and each of which is cogently set out by this House in Rondel v. Worsley explaining why public policy demands that a barrister, in common with a judge, juryman, or witness, shall be immune against being sued in respect of anything he does or says in courts. I entirely agree with that immunity for the reasons I gave when Rondel v. Worsley was before the Court of Appeal, and I shall not repeat any of them. I cannot, however, understand how any aspect of public policy could possibly confer immunity on a barrister in a case such as the present should he negligently fail to join the correct persons or to advise that they should be joined as defendants; or for that matter should he negligently advise that the action must be discontinued. It seems plain to me that there could be no possibility of a conflict between his duty to advise his client with reasonable care and skill and his duty to the public and to the courts. I do not see how public policy can come into this picture. This is certainly not a case where it could possibly be regarded as oppressive to join Mrs. Sugden as a defendant. After all, she had pleaded guilty to driving without due care and attention at the material time. Nor do I understand how any aspect of public policy could have required counsel to advise that the action against Mr. Sugden should be discontinued. Once it is clear that the circumstances are such that no question of public policy is involved, the prospects of immunity for a barrister against being sued for negligently advising his client vanish into thin air, together with the ghosts of all the excuses for such immunity which were thought to exist in the past.
The theory that because the barrister had no contractual relationship with his client he could not be liable for negligent advice causing financial loss, vanished with Hedley Byrne & Co. Ltd. v. Heller & Partners Ltd.  AC 465 which overruled Candler v. Crane Christmas  1 All E.R. 426.
The other theories supporting a barrister's supposed blanket immunity from liability for negligence in respect of any paper work cannot survive in the realistic atmosphere of the late 20th century. These were based on the fact that (a) a barrister cannot sue his client for his fees and (b) he is obliged to accept briefs relating to a field of law in which he normally practises, providing he is offered a proper fee.
Although a barrister cannot sue for his fees, he can demand that his fees be paid before he appears in court. If the barrister does not demand his fees in advance and the lay client does not pay them after the barrister's services have been rendered, the solicitor can sue the lay client for the barrister's fees. It is true that if the solicitor recovers the fees from the lay client and does not pay them over to the barrister, the barrister cannot sue the solicitor; but he can report him to the Law Society, and this as every lawyer knows would be likely to cost the solicitor far more than the fees he retains. There is no reason to suppose that the Bar incurs more bad debts than any other profession.
The rule that a barrister must accept a brief in the circumstances which I have described was made to ensure that every accused person or litigant could be represented in court by counsel. I do not, however, know of any firm rule which obliges counsel to accept instructions to advise or to draft pleadings.
Unless what seems to me to be an untenable proposition is accepted, namely that public policy always requires that a barrister should be immune from liability for his neglect or incompetence in respect of all paper work, he is rightly in no better position than any other professional man who is sued for negligence. The normal rule applied by the law is that if anyone holding himself out as possessing reasonable competence in his avocation undertakes to advise or to settle a document, he owes a duty to advise or settle the document with reasonable competence and care. This duty is owed to anyone he should foresee may suffer loss if the duty is breached.
If in breach of that duty, he fails to exercise reasonable competence or care and as a result the person to whom the duty was owed suffers damage he is liable to compensate that person for the damage he has suffered. The law requires the damage to be borne by the person whose breach of duty has caused it, rather than by the innocent person who has suffered it.
I am far from saying that if the advice or document turns out to be wrong, it necessarily follows that he who gave or drew it is liable for the loss caused by its imperfection. The barrister is under no duty to be right; he is only under a duty to exercise reasonable care and competence. Lawyers are often faced with finely balanced problems. Diametrically opposite views may and not infrequently are taken by barristers and indeed by judges, each of whom has exercised reasonable, and sometimes far more than reasonable, care and competence. The fact that one of them turns out to be wrong certainly does not mean that he has been negligent. In my opinion, however, it can only be in the rarest of cases that the law confers any immunity upon a barrister against a claim for negligence in respect of any work he has done out of court; and this case is certainly not amongst them.
I ought to add that when Rondel v. Worsley came to the Court of Appeal, I felt bound for the reasons I have given to deal with points which I considered to be wholly irrelevant to anything we had to decide. I may have put the case too high if I used words which might give the impression that counsel's immunity always extended to the drafting of pleadings and to advising on evidence. I should have said that the immunity might sometimes extend to drafting pleadings and advising on evidence. If in an advice on evidence counsel states that he will not call Y. as a witness whom he believes his client wishes to call solely to prejudice his opponent, counsel is immune on grounds of public policy from being sued in negligence by his client for advising that Y. must not be called or for refusing to call him. In such a case the advice would be so closely connected with the conduct of the case in court that it should be covered by the same immunity. It would be absurd if counsel who is immune from an action in negligence for refusing in court to call a witness could be sued in negligence for advising out of court that the witness should not be called. If he could be sued for giving such advice it would make a travesty of the general immunity from suit for anything said or done in court and it is well settled that any device to circumvent this immunity cannot succeed, see e.g. Marrinan v. Vibart  1 Q.B. 234, 528.
I think that the passage from the judgment of McCarthy P. in Rees v. Sinclair  1 N.Z.L.R. at page 187, cited in the Court of Appeal, relates to facts similar to those I have just postulated. I think that McCarthy P. would be astonished to hear his judgment cited in support of the decision made by the Court of Appeal in the circumstances of the present case. The facts in Rees v. Sinclair do not emerge from the report of that case in the New Zealand Court of Appeal. The report, however, at first instance,  1 N.Z.L.R. 236, throws light on those facts. It appears that the plaintiff living apart from his wife under a deed of separation had paid her maintenance since 1962 and had entered into an agreement in 1964 to pay her future maintenance; he also abandoned his divorce petition. The defendant who was a solicitor and barrister acted for the plaintiff in proceedings in which the plaintiff claimed a variation of the agreed maintenance and his wife asked for an order for permanent maintenance. The plaintiff wished to support his case by alleging wrongful conduct by his wife prior to 1962. The defendant apparently considered that there was no justification for making these allegations and no evidence to support them. He accordingly advised that they should not be made and he refused to put them forward on the plaintiff's behalf. The defendant acting in his capacity as counsel, or indeed as solicitor, clearly owed a duty to the court on grounds of public policy not to put those allegations forward, taking the view that he did about them. The plaintiff sued him for negligence and the court held rightly that the defendant was immune from such proceedings. There is all the difference in the world between Rees v. Sinclair and the present case. I respectfully agree with McCarthy P. when he says:
"The protection exists only where the particular (paper) work is so intimately connected with the conduct of the cause in Court that it can fairly be said to be a preliminary decision affecting the way that cause is to be conducted when it comes to a hearing."
The "intimate connection" to which McCarthy P. referred, undoubtedly existed in Rees v. Sinclair and in the case I have just postulated. In my opinion there is no such connection between the advice given in the present case and conduct of the case in court. The advice given made it impossible for the plaintiff's unanswerable case to be heard in court. It was not even remotely connected with counsel's duty to the court nor with public policy.
My Lords, for the reasons I have stated I would allow the appeal.
Lord Russell of Killowen
In arriving at a conclusion in this appeal I have striven, I hope successfully, to avoid being influenced by the fact that I find it difficult to see where the barrister erred in selecting as defendant the owner of the car, whose wife was driving it (negligently) for the purpose of taking their children to school. However unsatisfactory it may be, we are obliged to assume for the purposes of this appeal that the barrister was negligent in the respects summarized by my noble and learned friend on the Woolsack. Should, on that assumption, the third party notice be struck out as disclosing no cause of action against the barrister recognised by the law?
The history relating to claims for negligence by professional people, and barristers in particular, has been very fully canvassed in your Lordships' speeches, as has also the tendency of recent years for decisions of the courts to stress responsibility in all fields of those guilty of errors of commission or omission towards others to whom harm has been caused by such errors.
For myself, my Lords, I accept without qualification the decision of this House in Rondel v. Worsley. The negligence there alleged was the method of conduct of a criminal defence in court. I cannot find that there is any sound distinction between that and the conduct of civil proceedings in court.
It is to be observed that nowhere, so far as I am aware, has it been suggested that a barrister in respect of his conduct of a case in court should have a special immunity from suit for negligence because what he does in court is (so to speak) in the heat of battle, when he must make a decision in the obscurity of dust and smoke. Nor do I consider that such immunity is rightly connected with a quite different (as it seems to me) immunity of judges, witnesses, and jurors.
What then is the justification for immunity from claims for negligence that has so far been established ? It lies in my opinion in the public duty that rests upon the bar in particular to participate in and contribute to the orderly proper and expeditious trial of causes in our courts. In the exercise of such a duty it appears to me that it is highly undesirable and against public policy that a barrister in deciding in court upon a particular course of action, or attitude, should be in a position of having to wonder whether a different course of action, or attitude, would not be more likely to avoid a claim that he was negligent. I take a simple illustration. It is desirable in a civil case to prove a fact. Five people say that they can testify to it. Counsel decides to call two, saving the time of the court. Those two turn out to be not good witnesses, or are not in the end accepted to be such by the judge. Is the barrister to be subjected to a claim in negligence by his disappointed client for his attempt to facilitate the administration of justice ? I consistently stress the word claim. He may or may not have been negligent in not calling the other three witnesses. It is his vulnerability to a claim that may interfere with his duty.
The example which I have given is of something done (or rather not done) in the course of the trial of a civil action inside the court. I cannot accept that there is a universally acceptable distinction between what is done (or omitted) in the court in the presence of the judge and what is done (or omitted) outside court or before trial. Preliminary hearings before a master may be regarded as upon the same basis as the trial. But what of actions (or omissions) of counsel in the course of steps taken upon which the trial (if any) is to be based? I do not accept the suggestion that the calm atmosphere of chambers in the Temple (or even Lincoln's Inn) marks a distinction: I have already remarked that the heat, dust and smoke of the battle has never been urged as a special ground of immunity from suit for negligence. In my opinion if a decision by counsel is arrived at—albeit negligently—before trial which might well have been arrived at at trial, or before a master, it should not be open to a claim for negligence.
In particular, as here, a decision as to the proper party to be joined as defendant.
I can find no justifiable line to be drawn at the door of the court, so that a claim in negligence will lie against a barrister for what he does or omits negligently short of the threshold though not if his negligent omission or commission is over the threshold. His immunity from claims of negligence should (granted that it is to exist at all) extend to areas which affect or may affect the course of conduct of litigation, in which areas are to be found the public duty and obligation of the barrister to participate in the administration of justice. And this should be so even if the result of the alleged negligence is that litigation does not in fact come about. A decision which shapes, or may shape, the course of a trial should be within the umbrella (or blanket) of freedom from claims whether it is arrived at before trial or during it. This must include advice on settlement: advice on evidence: advice on parties: to list only examples. A barrister is offered an opportunity in the course of a trial to add a party: he misunderstands the case and allegedly negligently declines the opportunity: as I understand Rondel v. Worsley he is immune from the claim. Is there any reason for not holding him also immune from a claim for not originally adding that party ? I think not.
My Lords, there may be much to be said for denying immunity from claims for negligence by a barrister in the conduct of civil litigation in court. But while that immunity stands, as I think it does as involved in the decision of this House in Rondel v. Worsley, I see no escape from the extension to pre-trial alleged negligence so strongly supported (obiter) in that case.
I find myself in agreement with the views expressed in his speech (which I have had the advantage of seeing in draft) of my noble and learned friend Lord Keith of Kinkel and, finding myself in concurrence with them, I would dismiss this appeal.
Lord Keith of Kinkel
The facts of this case and the circumstances under which it comes before your Lordships' House have been narrated in the speech of my noble and learned friend Lord Wilberforce, whose account I gratefully adopt.
According to the headnote in the report of Rondel v. Worsley  1 AC 191 this House held in that case that a barrister was immune from an action for negligence at the suit of a client in respect of his conduct and management of a case in court and the preliminary work connected therewith such as the drawing of pleadings. A majority of the House expressed the view that the immunity does not extend to matters unconnected with cases in court.
Rondel v. Worsley (supra) was on its facts concerned only with alleged negligence in connection with a barrister's work in court in the course of a criminal trial, so all that was said in their Lordships' speeches about other aspects of a barrister's work may strictly be described as obiter dicta. But their Lordships were clearly concerned to review, in the light of the authorities, the law as to the scope of a barrister's immunity from suits for negligence in connection with all aspects of his work, and what they said about those activities which are carried on elsewhere than in the actual courtroom in the course of a trial was carefully considered. Accordingly the views so expressed should not, in my opinion, lightly be departed from.
In the present appeal it was not argued that no aspect of a barrister's work attracted immunity, but it was contended that the immunity was strictly limited to work done actually in court in the conduct of a cause.
It is apparent that in the opinion of at least four of their Lordships in Rondel v. Worsley (supra) the scope of the immunity is not so limited. The position of Lord Morris of Borth-y-Gest in this respect is not entirely clear. The relevant passages from the speeches have been quoted by my noble and learned friend Lord Wilberforce, and I need not repeat them.
The principle upon which the decision in Rondel v. Worsley (supra) proceeded was that of public interest. The principal aspect was the consideration that the proper administration of justice had better prospects of being achieved if barristers, in their conduct of litigation, were not inhibited in any decision they might make by fear of being sued for negligence by a disappointed client. Such fear, so it was considered, might detract from the degree of attention given by the barrister to his duty to the court and to the administration of justice generally in situations where that duty conflicted with the personal interests of his client. In my view that consideration is valid and has a sound basis. Further, I have no doubt that its application extends further than the actual conduct of a case in court. A barrister's duty to the court and the due administration of justice has to be kept firmly in view when he directs his mind to whether an action should be brought and against what parties, to whether an action should be settled or abandoned and to advising on evidence and on the discovery of documents. It is true that decisions on such matters normally are taken in situations offering more opportunity for reflection than is present in face of the court in the course of a trial. But that might well mean that the decision is less instinctively correct in the light of the barrister's duty to the court and more likely to be influenced by thoughts of the action which the client, in the absence of an immunity, might take.
I am therefore of opinion that the grounds of this aspect of public interest extend beyond the actual conduct of a case in court and are applicable to all stages of a barrister's work in connection with litigation, whether pending or only in contemplation.
The second aspect of public interest considered relevant in Rondel v. Worsley (supra) was the undesirability of re-litigating between barrister and client issues which had ex hypothesi been decided adversely to the client in previous litigation. Here again I regard as convincing the reasons which led their Lordships to rely upon this ground for affirming the existence of immunity. It was argued that this ground could not apply where as a result of a barrister's negligence the client's claim had never been litigated at all. The answer to the argument is, in my view, that the purpose of the indemnity is to exclude in the public interest the bringing of actions of a certain category. The matter must rest upon principle, and it is not critical that the relevant factors of public interest may apply less strongly upon the facts to certain cases in the category covered by the principle than they do to others.
In Rondel v. Worsley (supra) there also entered to some extent into the grounds of decision stated by certain of their Lordships the rule that a barrister may not decline to act for a particular client in a field in which he normally practises, if he is disengaged and is tendered a reasonable fee. I do not myself regard this aspect as having much importance from the point of view of public interest, as I do not consider that if barristers were liable to be sued for negligence this would, having regard to the standards of the profession, be likely to result in any person being deprived of competent representation in any litigation.
In Rees v. Sinclair  1 N.Z.L.R. 180 the New Zealand Court of Appeal accepted that a limit should be placed upon a barrister's immunity from actions for negligence in respect of work done in connection with litigation. The nature of the limitation was thus expressed by McCarthy P. at p.187:
"But I cannot narrow the protection to what is done in Court: it must be wider than that and include some pre-trial work. Each piece of before-trial work should, however, be tested against the one rule; that the protection exists only where the particular work is so intimately connected with the conduct of the cause in Court that it can fairly be said to be a preliminary decision affecting the way that cause is to be conducted when it comes to a hearing. The protection should not be given any wider application than is absolutely necessary in the interests of the administration of justice, and that is why I would not be prepared to include anything that does not come within the test I have stated."
In my opinion the restriction thus expressed is not consistent with the principal ground of the decision in Rondel v. Worsley (supra), namely that immunity should apply to all situations where there is the possibility of conflict between the barrister's duty to the court and to the proper administration of justice and the personal interests of his client. That possibility was considered by the majority at least in Rondel v. Worsley (supra) to exist in relation to all aspects of a barrister's work in connection with litigation, and I respectfully agree with them. I am unable to perceive any distinction valid for the purposes of the relevant principle between work in connection with litigation which affects the way the case is conducted when it comes to a hearing and that which does not. Further, I consider that it would be undesirable, and of marginal utility to disappointed litigants, to open the door to inquiry in individual cases whether or not the distinction is established. To do so would go some length towards defeating the purpose of the immunity, which is to enable barristers to apply their minds to litigation work undistracted by consideration of whether or not they might be sued for negligence. If the considerations of public interest held valid in Rondel v. Worsley (supra) are to be accepted as being so, as in my opinion they should be, then they should in my view be applied so as to achieve their purpose as effectively as possible. Further, I think there is merit in the maintenance of a rule which is relatively simple and easy to apply. That immunity extends to all of a barrister's work in connection with litigation is such a rule. The suggested restriction of the immunity would presumably exclude from its scope all cases relating to contemplated litigation which did not actually reach the stage of a hearing in court, and all litigation settled, compromised or abandoned. In other cases the suggested restriction would, in my opinion, prove difficult to apply in practice and would almost inevitably require inquiry into the facts. It would seldom, if ever, be possible to decide the issue of immunity upon an application for striking out. So the objective of relieving the barrister of any apprehension of contentious litigation regarding the conduct of his cases would not be achieved.
In the present case the negligence alleged against the third party took place in connection with his conduct of litigation. I am of opinion that the immunity therefore applies to it.
My Lords, for these reasons I would dismiss the appeal.