Giannarelli v Wraith

The first three appellants claim that the first respondent, who is a barrister, represented them in the committal proceedings, that the second respondent, who is a Queen's Counsel, represented them at the trial, and that the third respondent, who is also a Queen's Counsel, represented the second and third appellants in the Court of Criminal Appeal. The fourth appellant (i.e., Shulkes) was the instructing solicitor in the committal proceedings and at the trial.

The first three appellants sued the respondents and the fourth appellant in the Supreme Court of Victoria for damages for negligence, alleging a negligent failure (a) to advise that s. 6DD would render the evidence given in the Royal Commission inadmissible and thus defeat the Crown case; and (b) to object on that ground to the tender of that evidence. The fourth appellant issued a third party notice against the first and second respondents. However, the first three appellants withdrew their claim against him so that he is a nominal party to the present proceedings.

The question whether the respondents could be liable in negligence in the circumstances just outlined was argued as a preliminary question of law. At first instance Marks J. held that the respondents were not immune from liability in negligence by reason of s. 10(2) of the Legal Profession Practice Act 1958 Vict. ("the 1958 Act"). The Full Court (Young C.J., Crockett and Fullagar JJ.) came to a different conclusion and allowed the respondents' appeal. The present appeal is brought from the order of the Full Court.

The appellants' case is that s. 10(2) of the 1958 Act imposes liability on the respondents for negligence and that, in the alternative, the respondents are subject to a common law duty of care. It is convenient to consider, first, the argument that the respondents are subject to a common law duty of care. That consideration may throw some light on the interpretation of the statute.

As Tindal C.J. observed as long ago as 1838 in Lanphier v. Phipos [23] : "Every person who enters into a learned profession undertakes to bring to the exercise of it a reasonable degree of care and skill." On the other hand, the common law has for a very long time recognized that the barrister is not subject to such a general duty of care. The immunity of the barrister from liability in negligence to his client, at least in respect of court work, is supported by powerful authority, ancient and modern, in England, Scotland and Ireland: see Rondel v. Worsley [24] ; Saif Ali v. Sydney Mitchell & Co. [25] . The fact that advocacy in these jurisdictions is primarily the function of small independent bars distinguishes the situation there from that in the American States and consequently American authority (see, e.g., Woodruff v. Tomlin [26] ) is not particularly relevant.

1. (1838) 8 Car. & P. 475, at p. 479 [173 E.R. 581, at p. 583]. 2. [1969] 1 A.C. 191, at pp. 240-244, 258-263, 277-279, 288-289. 3. [1980] A.C. 198. 4. (1980) 616 F. 2d 924, at p. 930.

Various explanations for the barrister's immunity have been advanced. Historically it has been linked to the barrister's inability to sue the client for his professional fees: see In re Le Brasseur and Oakley [27] ; Robertson v. MacDonogh [28] . However, in Rondel v. Worsley the House of Lords squarely rejected the suggestion that the barrister's inability to sue for his fees could support his immunity in negligence. The reason given for that conclusion is compelling. The negligent performance of a service, even if it be undertaken without consideration, gives rise to liability in negligence, if the person for whose benefit the service is performed relies upon that service.

1. [1896] 2 Ch. 487, at p. 494. 2. (1880) 6 L.R. Ir. 433, at p. 438.

So the barrister's immunity, if it is to be sustained, must rest on considerations of public policy. Of the various public policy factors which have been put forward to justify the immunity, only two warrant serious examination. The first relates to the peculiar nature of the barrister's responsibility when he appears for his client in litigation. The second arises from the adverse consequences for the administration of justice which would flow from the re-litigation in collateral proceedings for negligence of issues determined in the principal proceedings.

The peculiar feature of counsel's responsibility is that he owes a duty to the court as well as to his client. His duty to his client is subject to his overriding duty to the court. In the performance of that overriding duty there is a strong element of public interest. So, in Swinfen v. Lord Chelmsford [29] Pollock C.B., after speaking of the discharge of counsel's duty as one in which the court and the public, as well as the client, had an interest said:

The conduct and control of the cause are necessarily left to counsel A counsel is not subject to an action for calling or not calling a particular witness, or for putting or omitting to put a particular question, or for honestly taking a view of the case which may turn out to be quite erroneous. If he were so liable, counsel would perform their duties under the peril of an action by every disappointed and angry client. In the result the Court of Exchequer concluded [30] "that no action will lie against counsel for any act honestly done in the conduct or management of the cause". 1. (1860) 5 H. & N. 890, at p. 921 [157 E.R. 1436, at p. 1449]. 2. (1860) 5 H. & N., at p. 923 [157 E.R., at p. 1450].

The performance by counsel of his paramount duty to the court will require him to act in a variety of ways to the possible disadvantage of his client. Counsel must not mislead the court, cast unjustifiable aspersions on any party or witness or withhold documents and authorities which detract from his client's case. And, if he notes an irregularity in the conduct of a criminal trial, he must take the point so that it can be remedied, instead of keeping the point up his sleeve and using it as a ground for appeal.

It is not that a barrister's duty to the court creates such a conflict with his duty to his client that the dividing line between the two is unclear. The duty to the court is paramount and must be performed, even if the client gives instructions to the contrary. Rather it is that a barrister's duty to the court epitomizes the fact that the course of litigation depends on the exercise by counsel of an independent discretion or judgment in the conduct and management of a case in which he has an eye, not only to his client's success, but also to the speedy and efficient administration of justice. In selecting and limiting the number of witnesses to be called, in deciding what questions will be asked in cross-examination, what topics will be covered in address and what points of law will be raised, counsel exercises an independent judgment so that the time of the court is not taken up unnecessarily, notwithstanding that the client may wish to chase every rabbit down its burrow. The administration of justice in our adversarial system depends in very large measure on the faithful exercise by barristers of this independent judgment in the conduct and management of the case. In such an adversarial system the mode of presentation of each party's case rests with counsel. The judge is in no position to rule in advance on what witnesses will be called, what evidence should be led, what questions should be asked in cross-examination. Decisions on matters such as these, which necessarily influence the course of a trial and its duration, are made by counsel, not by the judge. This is why our system of justice as administered by the courts has proceeded on the footing that, in general, the litigant will be represented by a lawyer who, not being a mere agent for the litigant, exercises an independent judgment in the interests of the court.

There is a real risk that, if counsel were exposed to liability in negligence, the existence of that potential liability would influence the exercise of his independent judgment by making him more mindful of the need to avoid any possibility of liability to his client. In some situations, in order to avoid that possibility, counsel would pursue matters which he would not otherwise pursue if the exposure to liability in negligence did not exist. To expect that counsel's conduct of a case would not be influenced by his exposure to such a potential liability would be little more than a pious hope. Inevitably some counsel would be more inclined to act as mere agents of their clients to the detriment of the interests of the court and of the administration of justice generally. Insurance might alleviate but would not eliminate the problem. Counsel would naturally be concerned to avoid allegations of negligence.

It follows that the exposure of counsel to liability in negligence for breach of a common law duty of care would create a real risk of adverse consequences for the efficient administration of justice. Litigation would tend to become more lengthy, more complex and more costly.

To deny the litigant a cause of action for negligence, even if it be limited to in-court negligence, on the part of his counsel is a serious step. It is to sanction a continuing exception in favour of counsel, as against his client, from the ever-expanding tort of negligence. But the exception which the law creates is not to benefit counsel but to protect the administration of justice. And the exception in favour of counsel is in conformity with the privilege which the law has always conferred in the interests of public policy on those engaged in the administration of justice, whether as judge, juror, witness, party counsel or solicitor, in respect of what they say in court: Cabassi v. Vila [31] ; Munster v. Lamb [32] .

1. (1940) 64 C.L.R. 130, at p. 141. 2. (1883) 11 Q.B.D. 588.

The foundation for that principle is the perception that great mischief would result if those engaged in the administration of justice were not at liberty to speak freely. The immunity is not confined to actions for defamation. As McTiernan J. noted in Cabassi [33] with reference to the rule in its application to witnesses:

It is a rule of law that no civil action lies at the suit of any person for any statement made by a witness in the course of giving evidence in a judicial proceeding. The rule, which is founded on public policy, is not confined to actions for defamation but applies to any form of action. 1. (1940) 64 C.L.R., at p. 144.

The considerations which dictate the need to protect freedom of speech in court likewise dictate the need to protect the advocate's freedom of judgment with respect to what is said and done in court. Just as the principle protects the judge and the juror in relation to what they decide, so it protects the advocate. The advocate is as essential a participant in our system of justice as are the judge, the jury and the witness and his freedom of judgment must be protected: see the discussion by Brett M.R. in Munster [34] . The need for that protection arises from "the fear that if the rule were otherwise, numerous actions would be brought against persons who were merely discharging their duty", to repeat the words of Fry L.J. in Munster [35] .

1. (1883) 11 Q.B.D., at pp. 603-604. 2. (1883) 11 Q.B.D., at p. 607.

The second aspect of public policy that calls for attention is the impact on the administration of justice of allowing court decisions to become the subject of collateral attack by means of actions against counsel for in-court negligence. Exposure of counsel to liability for such negligence would unquestionably encourage litigation by unsuccessful litigants anxious to demonstrate that, but for the negligence of counsel, they would have obtained a more favourable outcome in the initial litigation. That would be the central issue for decision in secondary litigation of this kind. If the plaintiff were to succeed, the resolution of this issue by a different court and on materials which might well differ from those presented in the initial litigation, due to lapse of time or other reasons, would undermine the status of the initial decision. Yet an appeal against that decision might not succeed with the result that it would stand, though its status would be tarnished by the outcome of the collateral proceedings. The impact of a successful challenge to a criminal conviction resulting in a sentence of imprisonment would be all the greater. It would be destructive of public confidence in the administration of justice. And for this very reason there would be a strong incentive on the part of a disappointed litigant to sue counsel for negligence as an indirect means of calling in question the decision in the initial litigation.

True it is that exposure of counsel to liability for in-court negligence would not result in liability for a mere error of judgment. But the dividing line between a non-negligent error of judgment and a negligent error of judgment in particular factual situations is by no means easy to draw. For example, many tactical decisions about the way in which a case is to be conducted might well be characterized as mere errors of judgment. On the other hand, some tactical decisions about the conduct of a case may be so flawed that they go beyond an error of judgment. It would be a mistake to attach too much importance to this concept as affording a substantial brake on counsel's liability. It would not eliminate the prospect of actions being brought against counsel for negligence in the conduct and management of the initial litigation with a view to challenging by collateral means the decision in that litigation.

On the issue of liability for in-court negligence I would draw no distinction between barristers and solicitors. The reasons for holding that a barrister is not under a duty of care apply with equal force to the solicitor who acts as an advocate. In this respect I would agree with the views expressed by Lord Wilberforce, Lord Diplock and Lord Salmon in Saif Ali [36] , to the effect that the same immunity attaches to a solicitor acting as an advocate in court as attaches to a barrister. It is the function performed, not the label attached, which gives rise to the limited immunity. And as an historical matter, the nineteenth century English decisions provide no acceptable support for the proposition that the solicitor acting as an advocate was liable for in-court negligence. Apart from Stokes v. Trumper [37] , there is no case which is capable of supporting this proposition. If the proposition were valid, it would be well documented by authority.

1. [1980] A.C., at pp. 215, 224, 227. 2. (1855) 2 K. & J. 232 [69 E.R. 766].

However, the grounds for denying liability for in-court negligence have no application to work done out of court which is unconnected with work done in court: Saif Ali. The public policy considerations underlying immunity from in-court negligence have no relevance to a barrister's liability for negligent advice in relation to out-of-court matters, in accordance with the principles expounded in such cases as San Sebastian Pty. Ltd. v. The Minister [38] ; Hawkins v. Clayton [39] ; and Hedley Byrne & Co. Ltd. v. Heller & Partners Ltd. [40] . The problem is: where does one draw the dividing line? Is the immunity to end at the courtroom door so that the protection does not extend to preparatory activities such as the drawing and settling of pleadings and the giving of advice on evidence? To limit the immunity in this way would be to confine it to conduct and management of the case in the courtroom, thereby protecting the advocate in respect of his tactical handling of the proceedings. However, it would be artificial in the extreme to draw the line at the courtroom door. Preparation of a case out of court cannot be divorced from presentation in court. The two are inextricably interwoven so that the immunity must extend to work done out of court which leads to a decision affecting the conduct of the case in court. But to take the immunity any further would entail a risk of taking the protection beyond the boundaries of the public policy considerations which sustain the immunity. I would agree with McCarthy P. in Rees v. Sinclair [41] where his Honour said:

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. This persuasive statement of the limits of the immunity was indorsed by four members of the House of Lords in Saif Ali [42] . The statement is all the more important in that it acknowledges the existence and the limits of the immunity in a country where the legal profession is fused. 1. (1986) 162 C.L.R. 340. 2. (1988) 164 C.L.R. 539. 3. [1964] A.C. 465. 4. [1974] 1 N.Z.L.R. 180, at p. 187. 5. [1980] A.C., at pp. 215, 224, 232, 236.

Once this view is accepted, the immunity necessarily extends to a decision about the way in which the proceedings for perjury under s. 314 were to be conducted. Just how such a decision came to be made we do not know as the issue has arisen as a preliminary question of law. It is possible that the critical operation of s. 6DD was overlooked. But it is more likely that counsel thought that this Court's decision in Reg. v. Winneke; Ex parte Gallagher [43] , holding that the Royal Commissions Act operated in concert with the Victorian Act, indicated that the provisions of s. 6DD of the Royal Commissions Act did not affect the admissibility in perjury proceedings of evidence given to the Commissioner to the extent that he was acting under Victorian as well as Commonwealth authority. That was the submission advanced by the Crown in Giannarelli v. The Queen [44] which was rejected by this Court. Be that as it may, s. 6DD was a matter of defence to the charge. The failure to raise it as a defence, like the failure to raise it as a ground of objection to the reception of the evidence, was an incident of the conduct and management of the case in court. Accordingly, the negligence complained of falls within the common law immunity.

1. (1982) 152 C.L.R. 211. 2. (1983) 154 C.L.R. 212.

(1) Every barrister shall be entitled to maintain an action for and recover from the solicitor or client respectively by whom he has been employed his fees costs and charges for any professional work done by him.

(2) Every barrister shall be liable for negligence as a barrister to the client on whose behalf he has been employed to the same extent as a solicitor was on the twenty-third day of November One thousand eight hundred and ninety-one liable to his client for negligence as a solicitor.

Twenty-third November 1891 was the date on which the Legal Profession Practice Act 1891 Vict. ("the 1891 Act"), the ancestor of the 1958 Act, came into force. The provisions of s. 5 of that Act were to all intents and purposes the same as those of the present section, except in so far as that part of s. 5 which corresponded with s. 10(2) of the 1958 Act referred to "the same extent as a solicitor is now liable to his client for negligence as a solicitor". The enactment of the 1891 Act was the culmination of popular agitation to place the two separate branches of the legal profession in Victoria on the same footing, no doubt with a view to amalgamation. That fusion did not happen. However, the point to be made is that the object of the 1891 Act was to place the two branches of the profession on the same footing. Section 5 should be construed accordingly rather than in the light of any supposed notion that it was intended to subject barristers to common law liability for in-court negligence because solicitors were already subject to that liability. As we have seen, the nineteenth century cases did not establish that a solicitor was liable in negligence to a client in respect of work undertaken by the solicitor as an advocate. The provisions of s. 5 and the later provisions of s. 10(2) are to be interpreted against that background. I have had the advantage of reading the reasons for judgment prepared by Wilson J. and I agree with what he has to say about the interpretation of the statutory provisions. Accordingly, I am of the opinion that s. 10(2) of the 1958 Act does not subject a barrister to a common law duty of care in negligence.

The sole question in these appeals is whether a barrister in Victoria can be held liable for negligence in respect of his conduct of a case in court.

The question arises in this way. In October 1981 each of the appellants in the first appeal ("the Giannarelli appellants" or, when referred to collectively, "the plaintiffs") gave evidence before the jointly conducted Commonwealth and Victorian Royal Commission into the Federated Ship Painters' and Dockers' Union. In consequence of that evidence, they were each charged and convicted of perjury contrary to s. 314 of the Crimes Act 1958 Vict.. The first-named Giannarelli appellant was released on a good behaviour bond; the second and the third-named Giannarelli appellants were sentenced to imprisonment. An appeal by the second and third-named Giannarelli appellants to the Court of Criminal Appeal of Victoria failed. However, an application to this Court for special leave to appeal was successful. The application was granted, the appeal was allowed and the convictions of the second and third-named Giannarelli appellants were quashed: see Giannarelli v. The Queen [45] . The argument upon which the appeal succeeded was that by reason of s. 6DD of the Royal Commissions Act 1902 Cth the evidence given before the Royal Commission was not admissible in the criminal proceedings taken against the plaintiffs. In the present proceedings the plaintiffs allege that the point was raised for the first time in the application to this Court. Prior to that time, they had been legally represented by a number of persons, including the three respondents and the appellant in the second appeal ("the appellant Shulkes"). The respondents in the first appeal are the same respondents as in the second appeal. The first-named respondent, a barrister, appeared on behalf of the plaintiffs at the committal proceedings. The second-named respondent, one of Her Majesty's counsel in the State of Victoria, was leading counsel at the trial. The third-named respondent, also one of Her Majesty's counsel, appeared as senior counsel for the second and third-named Giannarelli appellants before the Court of Criminal Appeal. The appellant Shulkes was the instructing solicitor at the committal proceedings and the trial.

After their convictions were quashed by this Court, the plaintiffs instituted civil proceedings in the Supreme Court of Victoria against, inter alia, the respondents and the appellant Shulkes, claiming that they had each been negligent in failing to advise the plaintiffs about the provisions of s. 6DD and in failing to make a submission to the courts that that section rendered the evidence on which they had been convicted inadmissible. The appellant Shulkes issued a third-party notice against the first and second-named respondents and, notwithstanding that the plaintiffs withdrew their claims against him, he remained a party to the proceedings without objection.

Subsequent to the criminal trial and the appeal to the Court of Criminal Appeal (but prior to the ruling of this Court), a submission based on s. 6DD had been made in another trial arising out of substantially the same circumstances but the learned trial judge on that occasion had ruled that the evidence was admissible: Reg. v. Guiseppe Giannarelli [46] .

1. Unreported; Supreme Court of Victoria; 22 June 1983.

Following the institution of the plaintiffs' action, questions of law, the essence of which are stated summarily at the beginning of this judgment, were isolated for determination prior to trial. Marks J., founding himself primarily on s. 10(2) of the Legal Profession Practice Act 1958 Vict. ("the 1958 Act"), held that the respondents were subject to liability for all negligence that may be proved against them; they had no immunity from suit. However, the Full Court (Young C.J., Crockett and Fullagar JJ.), in a unanimous judgment, came to a different conclusion and upheld the respondents' appeal. The appeals to this Court were heard together.

My consideration of the problem is greatly assisted by the detailed and comprehensive judgments, both of Marks J. and of the Full Court [47] .

The appellants argued that s. 10(2) of the 1958 Act imposes liability on the respondents for any negligence, including "in-court" negligence; in the alternative, the respondent barristers are so liable under the common law of this country, which admits no immunity from suit and thus differs from the course taken by the House of Lords in Rondel v. Worsley [48] and Saif Ali v. Sydney Mitchell & Co. [49] .

Section 10(2) of the Legal Profession Practice Act 1958 Vict. ("the 1958 Act").

It is necessary to trace the legislative history of s. 10(2). That history begins with the enactment of the Legal Profession Practice Act 1891 Vict. ("the 1891 Act"). Prior to the commencement of that Act, the legal profession in Victoria was structured substantially in the same way as in England. It was divided into two classes, barristers and solicitors. The former enjoyed an exclusive right of audience before the superior courts. The latter class comprised attorneys, solicitors and proctors. They were generally referred to as solicitors: see Baker, The Legal Profession and the Common Law (1986), Chs 8 and 9, for a discussion of the position in England.

From about 1880 onwards, the amalgamation of the legal profession in Victoria became the subject of public debate and this debate eventually resulted in the passage of the 1891 Act. It is necessary to note a number of the provisions of this Act. Section 2 defined a "barrister" to mean "a barrister of the Supreme Court" and a "solicitor" to mean "an attorney solicitor and proctor of the said court". Other provisions were as follows:

3. Every person who has heretofore been admitted as a barrister shall by virtue of this Act be and is hereby admitted as a solicitor as from the date of the passing of this Act, and in addition to his right to practise as a barrister shall be entitled to practise as a solicitor and to all the rights powers and privileges of a solicitor.

4. Every person who has heretofore been admitted as a solicitor shall by virtue of this Act be and is hereby admitted as a barrister as from the date of the passing of this Act, and in addition to his right to practise as a solicitor shall be entitled to practise as a barrister and to all the rights powers and privileges of a barrister.

5. Every barrister shall in future be entitled to maintain an action for and recover from the solicitor or client respectively by whom he has been employed his fees costs and charges for any professional work done by him. And every barrister shall in future be liable for negligence as a barrister to the client on whose behalf he has been employed to the same extent as a solicitor is now liable to his client for negligence as a solicitor.

7. After the passing of this Act no barrister attorney solicitor or proctor shall be in any manner whatsoever an officer of the Supreme Court. Provided always that nothing herein contained shall prevent the Supreme Court from having and exercising the same summary jurisdiction over every barrister and solicitor as the Court can now exercise in the case of solicitors.

Section 8(1) provided that a barrister and solicitor be deprived of the whole or any portion of a fee for work involving attendance in court if he failed to give substantial attendance; however this did not affect any liability which the barrister or solicitor may have incurred for negligence: s. 8(2). Section 10 provided: After the passing of this Act no person shall be admitted to practise as a barrister or a solicitor solely, but every person admitted by the Supreme Court shall be admitted both as a barrister and solicitor.

The 1891 Act was consolidated in 1915: Legal Profession Practice Act 1915 Vict.. Section 5 was re-enacted as s. 10 save that there was substituted for the words "is now liable" in the second sentence the words "was on the twenty-third day of November One thousand eight hundred and ninety-one liable", that being the day on which the original Act received the Royal Assent. The section was again re-enacted in the consolidation in 1928 when the words "in future" were omitted: Legal Profession Practice Act 1928 Vict., s. 10. Upon the subsequent consolidation of the legislation in the 1958 Act, which is currently in force, the section took the following form:

(1) Every barrister shall be entitled to maintain an action for and recover from the solicitor or client respectively by whom he has been employed his fees costs and charges for any professional work done by him.

(2) Every barrister shall be liable for negligence as a barrister to the client on whose behalf he has been employed to the same extent as a solicitor was on the twenty-third day of November One thousand eight hundred and ninety-one liable to his client for negligence as a solicitor.

The other provisions of the 1891 Act to which I have referred are also to be found in substantially the same form in the 1958 Act.

It is plain that the purpose of the 1891 Act was to amalgamate the two branches of the legal profession in Victoria thereby placing them on the same footing. This process involved, inter alia, the two steps which are dealt with in s. 5. Henceforth, barristers, like solicitors, were to be able to sue for their fees. Conversely, they were to become liable to be sued, to some degree at least, for negligence. There seems no reason to doubt that the historical immunity to suit enjoyed by barristers was related to their inability to sue for their fees. Remuneration for the services of a barrister was by way of an honorarium, a voluntary payment. The absence of any contract served to distinguish barristers from solicitors and to explain, in part, the distinction in liability: see Fell v. Brown [50] ; Swinfen v. Lord Chelmsford [51] ; Kennedy v. Broun [52] . But a barrister's protection from suit was not explicable solely in terms of his inability to sue for his fees. It was, even as early as the decisions in Swinfen and Batchelor v. Pattison [53] , explicable also by reference to considerations of public policy: see Rondel [54] . It is also to be noted that, notwithstanding the objective of the 1891 Act, it was not fully achieved. A separate Bar continued in existence and the profession, although fused in law, remained in practice a divided one: see Sawer, Division of a Fused Legal Profession; The Australian Experience, University of Toronto Law Journal, vol. 16 (1966) 245; Hayes v. Jones [55] . With this background in mind, I turn to the construction of s. 10(2) of the 1958 Act.

1. (1791) Peake 131 [ 170 E.R. 104]. 2. (1860) 5 H. & N. 890 [157 E.R. 1436]. 3. (1863) 13 C.B. (N.S.) 677 [143 E.R. 268]. 4. (1876) 3 R. (Ct. of Sess.) 914. 5. [1969] 1 A.C., at pp. 241-243, 258-263, 289-293. 6. [1926] V.L.R. 459, at p. 460.

It is common ground between the parties that the words "Every barrister" in s. 10(2) should be taken to mean "every barrister and solicitor acting as a barrister". A submission that s. 10(2) was merely a re-enactment of a transitional provision, and therefore concerned only a person who was a barrister prior to 1891, was raised before Marks J. but, rightly in my opinion, proved unsustainable.

The sub-section provides that a barrister is to be liable for negligence as a barrister to his client to the same extent as a solicitor was on 23 November 1891 liable to his client. The initial impression created by such a provision is that the intention was not to expose a barrister to negligence in respect of all the work undertaken by him on behalf of a client in his capacity as a barrister. If such was the intention it would have been sufficient to conclude the sub-section with the word "employed". The words "to the same extent as a solicitor was on [23 November 1891] liable to his client for negligence as a solicitor" cannot be ignored. They are words of limitation. They evoke the question: "To what extent was a solicitor liable in 1891 for his work as a solicitor?" It will be noted that neither the 1891 Act nor the subsequent consolidations purport to deal in any way with the liability of a solicitor for negligence. The provisions in question simply take that existing liability as some kind of a benchmark for determining the liability to which after 1891 a barrister is to be exposed as a barrister.

It was argued for the appellants that the sub-section renders barristers liable for negligence in respect of the work ordinarily done by barristers in the same way as solicitors have always been liable for negligence in respect of the work ordinarily done by solicitors. The ordinary work of barristers was advocacy; the ordinary work of solicitors did not involve advocacy. But to construe the provision in this way assumes that there exists a neat dichotomy in function between barristers and solicitors, as if the work of barristers is fully comprehended by their role as advocates, while advocacy has no relevant place in the work of solicitors. This is not so. Barristers engage in important areas of legal work as barristers which are not directly related to advocacy, including, e.g., the opinion work of counsel and advice on pleadings and evidence. Conversely, the word "solicitors" in the 1958 Act is defined (in s. 3) to mean an attorney, solicitor and proctor of the Supreme Court. The work of solicitors in 1891 included advocacy, albeit in the inferior courts. Until the commencement of the 1891 Act only barristers enjoyed the right of audience in the superior courts of Victoria, but in other courts they shared that right with solicitors. The extension of the right of audience in the Supreme Court to a solicitor was a consequence of the amalgamation effected by the Act. But it does not follow that a solicitor is necessarily practising as a barrister when exercising that right of audience. Whether or not he or she is practising as a barrister or a solicitor at any particular time is determined by the practitioner-client relationship. I understand the feature which has always distinguished a barrister from a solicitor - and still does - is that he acts through a solicitor and has no dealings directly with the lay client.

There are two other obstacles in the way of acceptance of the appellants' submission. First, as I have indicated, it is tantamount to saying that a barrister is liable for negligence in respect of all work done by him in that capacity and therefore pays no regard to the phrase commencing "to the same extent as ". Secondly, it would lead to the result that if one were to find that prior to 1891 solicitors were not liable for negligence as advocates before inferior courts, the 1891 Act, while extending their right of audience to advocacy in the superior courts, would leave them immune from suit while barristers would be liable. Such a result would be ludicrous and could only be avoided by treating a solicitor qua advocate as a "barrister" within the meaning of s. 10(2), with the result that the pre-1891 immunity of solicitors was destroyed. As I have said, the section does not purport to affect the liability of solicitors for negligence. Moreover, to include a solicitor qua advocate within the definition of "barrister" is not only inconsistent with the continuing role of solicitors in advocacy but it sits uneasily with what is sought to be overcome by s. 10(1), namely, the absence of a contract between barrister and client.

I reject the notion that in s. 10(2) the phrases "barrister" and "as a barrister" on the one hand, and "solicitor" and "as a solicitor" on the other hand, refer to different and mutually exclusive professional tasks - advocacy and non-advocacy - which are carried out by legal practitioners.

It was also argued that the phrase "to the same extent" which appears in s. 10(2) relates to the standard of care and not to the area of liability; in other words, s. 10(2) means that as solicitors are not liable for mere errors of judgment (or were not liable except for what in earlier times was called "gross negligence" or "crassa negligentia": see Godefroy v. Dalton [56] ; Purves v. Landell [57] ), neither are barristers. The Full Court rejected this argument and noted that "to exclude liability for errors of judgment would have been to exclude from negligence something that was not negligence". For present purposes I am prepared to assume that the phrase "to the same extent" is inclusive of the standard of care, but I cannot accept that it could have the effect of imposing liability on barristers in respect of something for which in 1891 solicitors could not be held liable. In my view, the phrase must denote a comparison of the area of liability of solicitors and barristers.

1. (1830) 6 Bing. 460 [130 E.R. 1357]. 2. (1845) 12 Cl. & Fin. 91 [8 E.R. 1332].

The sub-section is plainly a "fixed-time" provision: cf. Bennion, Statutory Interpretation (1984), p. 368. To determine whether it exposes a barrister to liability for in-court negligence it is necessary to ask whether, on 23 November 1891, a solicitor was liable to his client for in-court negligence in respect of his handling of a case as an advocate.

Having regard to the limited right of audience enjoyed by solicitors both in England and in Victoria prior to 1891, it is not surprising that relevant case law on this question is sparse. Counsel for the appellants cited a number of cases in support of a submission that prior to that time solicitors were liable to their clients for negligence in relation to their activities as advocates before those tribunals in which they had a right of audience. The cases referred to included Stokes v. Trumper [58] ; Montriou v. Jefferys [59] ; Hart v. Frame [60] ; Swannell v. Ellis [61] ; Smith v. Grant and Leslie [62] ; Manley v. Palache [63] ; Fergusson v. Lewis [64] ; Clarke v. Couchman [65] . The Full Court subjected those cases, as well as others, to detailed examination. I agree with the analysis made by their Honours and their conclusion that there is little, if any, support for the view that in 1891 solicitors were liable for in-court negligence. I do not find it necessary to cover the same ground in these reasons. Their Honours recognized the force of the argument that while there may be no actual examples of a solicitor being held liable for in-court negligence, nevertheless the cases supply no evidence of a recognition of the existence of any immunity from suit for solicitors whether based on reasons of public policy or otherwise. Their Honours explained, in my view correctly, the absence of such recognition on the different histories surrounding the two branches of the profession, the closer supervision exercised by the courts over solicitors than with respect to barristers and the absence of any right of audience for solicitors in the superior courts.

1. (1855) 2 K. & J. 232 [69 E.R. 766]. 2. (1825) 2 Car. & P. 113 [172 E.R. 51]. 3. (1839) 6 Cl. & Fin. 193 [7 E.R. 670]. 4. (1823) 1 Bing. 347 [130 E.R. 140]. 5. (1858) 20 Dunlop 1077. 6. (1895) 73 L.T. 98. 7. (1879) 14 L.J. 700. 8. (1885) 20 L.J. 318.

It is true that nineteenth century writers made no suggestion that solicitors, when acting as advocates, were immune from suit for in-court negligence: see, e.g., Pulling, A Summary of the Law and Practice Relating to Attorneys, General and Special, 3rd ed. (1862), pp. 424-425; Smith, A Treatise on the Law of Negligence, 2nd ed. (1884), pp. 198-203; Cordery, Law Relating to Solicitors of the Supreme Court of Judicature, 2nd ed. (1888), pp. 111-115; Beven, Principles of the Law of Negligence (1889), pp. 792-796; White, A Treatise on the Constitution and Government of Solicitors, Their Rights and Duties (1894), p. 175; and see also Halsbury's Laws of England, 1st ed. (1914), vol. 26, par. 1253. But none of these writers can be said to have focused precisely on the question here at issue, namely, the liability of a solicitor for in-court negligence whilst acting as an advocate, and it would be wrong to read any of the texts as positively propounding that solicitors were so liable. In any event, even assuming that it is the belief of Parliament as to what the law in 1891 was, rather than what it actually was, that bears on the proper interpretation of s. 10(2) (this being the view expressed by Marks J.), there is no warrant for according to Parliament the same perception of the law as that expressed by the writers. Indeed, it is more likely that the Parliament was not concerned to identify the extent of the liability for negligence of solicitors. As I have said, whatever the liability of solicitors was, s. 5 of the 1891 Act simply adopted it as the benchmark by which equality of treatment of barristers and solicitors in their amalgamated state was to be achieved. That concern for equality is also reflected in the first limb of s. 5 (now s. 10(1)) and in the provisions surrounding s. 5 in the 1891 Act: see ss. 3, 4, 7 and 10.

It bears repeating that, even in the nineteenth century, the seeds of the barrister's immunity from suit had been sown not only in the absence of a contract between barrister and client and the related inability of the barrister to sue the client to recover fees, but also in public policy grounds. Those policy grounds were extant in 1891. They were recognized in cases shortly before 1891, such as Swinfen and Batchelor , and there is no logical reason why, in 1891, they were not relevant to the in-court activities of a solicitor acting as an advocate.

In considering the question of the extent to which solicitors were in 1891 liable for negligence, assistance is also to be gained from the decision of the Court of Appeal in Munster v. Lamb [66] . Mr. Lamb was a solicitor who appeared as advocate for a defendant in a court of petty sessions. In the course of the case he spoke defamatory words of the prosecutor, who afterwards sued him for defamation. It was held that the action would not lie and it was immaterial whether the words were spoken maliciously and without any justification or excuse. Brett M.R., speaking of the absolute privilege accorded to judges and witnesses for words uttered in the course of a judicial proceeding, continued [67] :

If upon the grounds of public policy and free administration of the law the privilege be extended to judges and witnesses, although they speak maliciously and without reasonable or probable cause, is it not for the benefit of the administration of the law that counsel also should have an entirely free mind? Of the three classes - judge, witness, and counsel - it seems to me that a counsel has a special need to have his mind clear from all anxiety. A counsel's position is one of the utmost difficulty. He is not to speak of that which he knows; he is not called upon to consider, whether the facts with which he is dealing are true or false. What he has to do, is to argue as best he can, without degrading himself, in order to maintain the proposition which will carry with it either the protection or the remedy which he desires for his client. If amidst the difficulties of his position he were to be called upon during the heat of his argument to consider whether what he says is true or false, whether what he says is relevant or irrelevant, he would have his mind so embarrassed that he could not do the duty which he is called upon to perform. For, more than a judge, infinitely more than a witness, he wants protection on the ground of benefit to the public. It seems to me that much of this reasoning points to a principle of the common law which must be invoked in support of immunity for the advocate from an action for in-court negligence: see also Cabassi v. Vila [68] . 1. (1883) 11 Q.B.D. 588. 2. (1883) 11 Q.B.D., at pp. 603-604. 3. (1940) 64 C.L.R. 130.

In Rondel, a majority of the members of the House of Lords, speaking of the common law of England and Scotland, held that a solicitor acting as an advocate has the same immunity from an action for negligence as a barrister [69] . The same conclusion was reached in Saif Ali [70] . It was argued for the appellants that the views expressed on this topic in each of these two cases were obiter, perfunctory and not historically convincing; alternatively, it was said that, even if correct, the cases stated new law for the United Kingdom. However, I find the reasoning persuasive, and see no difficulty in accepting the conclusions of their Lordships in Rondel and in Saif Ali as strong evidence of the law of England as it was in 1891 with respect to the immunity of solicitors from liability for in-court negligence as advocates. Nor is there any reason to doubt that the common law of Victoria with respect to a solicitor's in-court immunity in 1891 was the same as the common law of England at that time.

1. [1969] 1 A.C., at pp. 232, 243-244, 265-267, 284-285. 2. [1980] A.C., at pp. 215, 224, 227.

I am therefore of the opinion that the proper construction of s. 10(2) does not lead to the conclusion that, by virtue of that provision, barristers in Victoria are liable for in-court negligence. They are made liable as barristers "to the same extent" as solicitors were liable in Victoria in 1891. Since solicitors were not then liable for in-court negligence in their work as advocates, so today are barristers immune from liability to that extent.

It was argued that this construction of s. 10(2) would be inconsistent with the intent of s. 12(2) of the 1958 Act. That sub-section mirrors s. 8(2) of the 1891 Act (referred to earlier in these reasons) and relevantly provides that a decision by a taxing officer as to whether a barrister has given substantial attendance to a matter "in court" shall not affect any liability which such barrister "may have incurred for negligence". There is no reason to doubt that a solicitor is liable for negligently failing to give his attendance in court. It follows, therefore, that by virtue of s. 10(2) a barrister may be similarly liable. That being so, the two sub-sections sit together comfortably.

It may be conceded that the fixed-time character of s. 10(2) of the 1958 Act is capable of giving rise to some difficulty. The goal of equality sought by the 1891 Act was open to be undermined if there was any contraction or expansion of the liability of a solicitor to his client occurring after 1891. In that event, a solicitor's liability in relation to certain work would be determined by the changed state of the law whereas the liability of a barrister as a barrister in relation to the same type of work would be governed by the law as it was in 1891. This possibility followed from the word "now", read in the context of the words "in future", which appeared in the second sentence of s. 5 of the 1891 Act. The same freezing effect follows from the reference in s. 10 of the 1958 Act to 23 November 1891. But in my opinion the possibility is of minimal significance. Not only is a change in the liability of a solicitor likely to be a rare occurrence (of which, however, the recent decision of this Court in Hawkins v. Clayton [71] may provide an example), but it is likely that the scope of any disparity would be confined to work performed by a barrister as a barrister otherwise than in respect of work intimately connected with the conduct of a cause in court (in the sense described by McCarthy P. in Rees v. Sinclair [72] ). It is to be remembered that any work which a practitioner who normally acts as a barrister undertakes directly for a lay client will be performed by that practitioner in the capacity of a solicitor and as such will be governed by the then current law relating to the liability for negligence of a solicitor.

1. (1988) 164 C.L.R. 539. 2. [1974] 1 N.Z.L.R. 180, at p. 187.

I should say that, in interpreting the Victorian legislation, I have been unable to find assistance in the Parliamentary Debates and other extrinsic material put forward on behalf of the appellants.

There would seem to be much to be said for the view that the entire question of the liability of a Victorian barrister for in-court negligence (indeed for out-of-court negligence also) is resolved by the proper construction of s. 10(2) of the 1958 Act regardless of any extension of common law liability for negligence that may have occurred since 1891. However, it was argued for the appellants as a second ground of appeal that, independently of the provisions of s. 10(2), the common law of Australia as presently discerned imposes a liability in tort upon a barrister who is negligent in the conduct of a case in court.

Even if it be assumed that s. 10 of the 1958 Act does not exhaustively cover the field of a barrister's liability for negligence, I do not think that the appellants can derive any comfort from the common law. In 1891 the common law recognized that advocates, whether solicitors or barristers, were immune from liability for in-court negligence. Notwithstanding the recognition that an action in professional negligence could lie in tort as well as contract, the important principle underlying the immunity remains compelling. The public interest in the due administration of justice, as enunciated by Brett M.R. in Munster v. Lamb in 1883, remains a valid unifying theme for the various issues of public policy canvassed by the House of Lords in Rondel. Five distinct grounds of public policy were advanced in Rondel in support of immunity: the concern that if counsel could be sued for negligence, they would be tempted to prefer the interests of their clients and would be deflected from observing their duty to the court; the adverse effect that the fear of litigation may have on the barrister's efficient conduct of the court proceedings; the "cab-rank" principle, whereby a barrister is not free within his field of practice to choose whether or not to act for a person who desires his services and can pay his fee; the special character of the judicial process wherein judges, jurors and witnesses are immune from civil action; lastly, the threat to the public interest centred in the finality of litigation.

Not all these considerations are of equal weight. There is no reason to suppose that counsel would be deterred by the possibility of a negligence claim from discharging a clear duty to the court in preference to observing the wishes of the client. Counsel could never be in breach of duty to the client by fulfilling the paramount duty.

Nevertheless, counsel's duty to the court is often easier to state than to apply in specific situations. For example, in a particular case, what will constitute sufficient evidence so as to justify counsel attacking the character of an opposing witness? There will be cases where it will not be easy to determine when the interests and instructions of the client collide with counsel's duty to the court. It is in those cases that counsel's judgment may be consciously or unconsciously impaired, leading him to favour his client's interests over his paramount duty. Yet the court "has and must continue to have implicit trust in counsel": Rondel [73] . True it is that counsel who errs on the side of his duty to the court is unlikely to be found liable in negligence even if he has misinterpreted the position and the reality is that there was no relevant duty to the court operating to prevent him from doing something which would be in his client's best interests. Counsel is not negligent merely because of a mere error of judgment. But it is the threat of litigation, not the likelihood of defeating such litigation, which is material. The expectation that an action in negligence brought against him would fail does not counter the instinctive motivation of counsel to err on the side of caution by bending to the client's interests and avoiding the possibility of troublesome litigation. For the same reason, counsel may be led, contrary to his view of what the justice of the case requires, to extend his examination of witnesses and submissions on the law, thereby unnecessarily prolonging the proceedings. The administration of justice would be at risk.

Next, there is the "cab-rank" principle; however, important though this principle is in itself, I would not accord it significant weight in considering the question of immunity: see Saif Ali [74] .

As I have said, the public policy reflected in the immunity from suit for defamation, which is conferred upon participants in the legal process, extends in an analogous way to protect counsel from liability for in-court negligence. It would be odd, to say the least, if counsel was immune from liability for malicious in-court slander yet liable for the negligent conduct of a case in court.

In addition to the foregoing matters, there are the many difficulties associated with relitigation, which would be a common feature of trials of actions against counsel. These difficulties provide a powerful argument for counsel's immunity. The situation is not to be compared with a case where an appeal is allowed, a decision set aside and a re-trial ordered. Such a course of events merely portrays the normal course of appellate review. It is altogether different where a disappointed litigant institutes a civil proceeding in a court of co-ordinate jurisdiction with a view to proving that the original decision was wrong by reason of counsel's negligence. If the negligence action succeeds, then the original decision, notwithstanding that it may have been affirmed on appeal, is necessarily tarnished by the later inconsistent decision. Yet nothing can correct the record or interfere with the original judgment. Furthermore, the result will have come about without the successful party to the original action being a party to the negligence action, which will fall to be determined in his or her absence. These situations clearly have a capacity to bring the administration of justice into disrepute: see Rondel, particularly at pp. 249-251.

The problem would be exacerbated in the case of a client who claimed to have been wrongly convicted by a jury of a criminal offence by reason of counsel's negligence in conducting the trial. The issue of causation in the negligence action where questions of fact for a jury were involved could be a mind-boggling exercise, piling "speculation upon speculation": Rondel [75] . The issue would have to be joined without any evidence from those most closely connected with it, the judge and members of the jury. Public policy would not permit them to be called as witnesses. There are other fundamental problems. Suppose a person is convicted, the jury being satisfied of guilt beyond a reasonable doubt. All avenues of appeal are pursued without success. The convicted person then institutes an action for negligence against the defence counsel, assuming the onus of proof on the balance of probabilities. If the action succeeds but the conviction remains, public confidence in the integrity of the law must be seriously and adversely affected. Even if a way is found to have the conviction set aside notwithstanding the earlier unsuccessful appeals, the end result is that the civil action is converted into a de facto avenue of appeal outside the carefully constructed statutory framework of criminal appeals. It may be noted that there is a long-standing policy of the common law against allowing the re-trial of criminal cases by collateral means: see the discussion by Harding, "Recent Cases, Rondel v. Worsley ", Western Australian Law Review, vol. 8 (1968) 242, at pp. 248-249.

The common law principle of immunity from civil action for in-court negligence thus derives support from the fundamental principle favouring finality of litigation. This principle has been found necessary to conserve public confidence in the administration of justice. It is this confidence which would be most at risk if the appellants' case were to succeed.

It was argued for the appellants that there is nothing novel to our courts about relitigation. In criminal law, a conviction may be set aside and a retrial ordered where justice miscarries because an accused is represented by incompetent counsel: Re Knowles [76] . Litigants may sue subpoenaed witnesses who fail to attend to give evidence and recover costs thrown away through the non-attendance (but it appears that damages at large are not recoverable: see, generally, Roberts v. J. & F. Stone Lighting & Radio Ltd. [77] ). Solicitors have been held liable for failing to prepare properly for a criminal trial which resulted in the client's conviction: Hatch v. Lewis [78] . It was also argued that since the strict rules of issue estoppel operate to exclude relitigation, where those rules do not apply there should be no barrier to relitigation. And the rule in Hollington v. F. Hewthorn & Co. [79] , where it has not been abolished by statute, may enable the factual issues leading to a criminal conviction to be relitigated in civil proceedings. It seems to me, however, that none of the above situations bear the same concatenation of policy factors as inhere in the present problem.

1. [1984] V.R. 751. 2. (1945) 172 L.T. 240. 3. (1861) 2 F. & F. 467 [175 E.R. 1145]. 4. [1943] 1 K.B. 587.

Counsel for the appellants asserted the existence of countervailing considerations of public policy. It was argued that every wrong should attract a remedy, lest a sense of injustice erode the fabric of the law. Another consideration advanced was the likely perception in the community, best expressed at first instance in Rondel [80] , that "barristers, with the connivance of the judges, [have] built for themselves an ivory tower and have lived in it ever since at the expense of their clients"; to allow such a perception to continue would lead to an erosion of public confidence in the law and thereby strike at the integrity of the legal system itself, ironically one of the very things which is said to justify the immunity of counsel.

It must be acknowledged that the law ought not readily grant privileges or immunities. Favouritism and inequality of treatment under the law are capable of breeding contempt for the law, particularly when it is perceived that those who are favoured are themselves lawyers. There is undoubtedly force in the view of Burke, Privileges and Immunities in American Law, South Dakota Law Review, vol. 31 (1985) 1, at p. 39 :

Privileges and immunities are per se conceptual contradictions of equal justice under the law. Narrowly and precisely defined and confined, a few of them may serve a useful purpose in contemporary society. However, a heavy burden of justification should be placed on any who would claim privilege or immunity, and any doubt should be resolved in favour of denial of the claim.

I accept that ordinarily there is a heavy burden of justification on those who claim an immunity or privilege. The point has particular relevance in the context of a case such as the present, since many of the arguments in support of immunity for counsel are predicated upon what might happen if immunity was foregone. There is simply no hard empirical evidence to indicate, e.g., whether or not counsel would become less fearless and independent, or more prolix, in the presentation and management of their cases: see Veljanovski and Whelan, Professional Negligence and the Quality of Legal Services - An Economic Perspective, Modern Law Review, vol. 46 (1983) 700, at pp. 711-718 .

In Rondel [81] , Lord Reid saw the issue before him as being "whether the abolition of the rule [of immunity] would probably be attended by such disadvantage to the public interest as to make its retention clearly justifiable" (emphasis added). And later his Lordship said [82] :

So the position appears to me to be this: if the present rule were changed there would at least be a grave risk of consequences much against the public interest. And what is to be the advantage? Privileges have been granted in other contexts on the basis of a perceived real risk (unsupported by empirical evidence) that if they were not granted, great detriment to the administration of justice would follow. For example, Fry L.J. in Munster v. Lamb said that the rule which accords court participants an absolute privilege for words uttered in the course of a judicial proceeding exists because of "the fear that if the rule were otherwise, numerous actions would be brought against persons who were merely discharging their duty" [83] (emphasis added). 1. [1969] 1 A.C., at p. 228. 2. [1969] 1 A.C., at p. 230. 3. (1883) 11 Q.B.D., at p. 607.

The administration of justice is of transcendent importance to the public interest. If, as I think it does, the imposition of liability for in-court negligence raises a real risk of grave damage to that public interest, or of serious loss of public confidence in the administration of justice, then the countervailing considerations identified by counsel for the appellants cannot be allowed to stand in the way of recognizing an immunity from suit. It must always be remembered that the immunity is granted not for the benefit of counsel but in the interests of the administration of justice.

The fact that the profession in Victoria is in law a fused profession does not warrant that it be treated differently to a strictly divided profession, as in England. The critical factor in this problem is not the manner in which the legal profession is structured. It is immaterial for present purposes whether a lawyer is admitted to practise as a barrister and solicitor, with freedom to engage in the whole range of activities associated with the profession. The critical factors are the functions he is performing at the material time and the impact which non-recognition of immunity might have upon the administration of justice. It is the function of advocacy that attracts the immunity, and, accordingly, it matters little whether the advocate is admitted to practise as a solicitor or as a barrister or as both. I draw support in this regard from the decision in Rees v. Sinclair, where the Court of Appeal of New Zealand held the law enunciated in Rondel to be relevant and appropriate to New Zealand notwithstanding the fusion of the profession in that country. A similar course was taken by Bray C.J. in South Australia: see Feldman v. A Practitioner [84] .

On the other hand, in both the United States and Canada, where there are also fused professions, the course of the law has been different. Counsel for the appellants referred the Court to various writings pertaining to the experience in the United States. It does not appear that the absence of a broad immunity of general application has given rise to undue concern, perhaps because actions in respect of in-court negligence are relatively infrequent. Most cases brought against attorneys for alleged negligence in the conduct of litigation have been unsuccessful, mainly because the disappointed clients face difficulty in proving that the damages they claim resulted from the negligence, or because they have failed to appreciate that an error in judgment does not constitute negligence: see generally the annotations at 45 A.L.R. (2d) 5 and 53 A.L.R. (3d) 731; "Attorney Malpractice", Columbia Law Review, vol. 63 (1963) 1292, at pp. 1307-1308; Kaus and Mallen, "The Misguiding Hand of Counsel - Reflections on "Criminal Malpractice" ", U.C.L.A. Law Review, vol. 21 (1974) 1191, at pp. 1192-1193; Mallen and Levit, Legal Malpractice, 2nd ed. (1981), p. 675. Since, however, the conduct of litigation and the role of the courts and of lawyers in the administration of justice in the United States is not comparable to the practice in Australia, I am unable to draw any assistance from the United States' experience.

The Canadian situation arguably has greater relevance to this country. It was suggested in Leslie v. Ball [85] that an attorney might be sued in negligence for the manner in which he conducted a case in court. The question does not appear to have been litigated again until the relatively recent case of Demarco v. Ungaro [86] . There, Krever J., sitting as the Ontario High Court of Justice, held against any immunity. His Lordship examined in detail the reasoning of their Lordships in Rondel and concluded that the law in Ontario had taken a different course. He relied, inter alia, on the fact that the legal profession was fused and that Leslie v. Ball had stood for more than a century without any intervening occasion requiring its reconsideration. His Lordship observed that negligence actions against lawyers respecting their conduct of court cases had not attained serious proportions. The decision of Krever J. appears to have received general acceptance from commentators in Canada (see, e.g., Hutchinson, Comments, Canadian Bar Review, vol. 57 (1979) 326, at pp. 346-355 ; Linden, Canadian Tort Law, 3rd ed. (1982), pp. 136-137; but see Traviss, "A Barrister's Liability to Civil Suit in Ontario: A Case Comment on Demarco v. Ungaro and Barycky", Law Society of Upper Canada Gazette, vol. 13 (1979) 262) and has been followed in other Ontario courts: see Wechsel v. Stutz [87] ; Karpenko v. Paroian, Courey, Cohen and Houston [88] ; Pelky v. Hudson Bay Insurance Co. [89] .

1. (1863) 22 U.C.Q.B. 512, at p. 519. 2. (1979) 95 D.L.R. (3d) 385. 3. (1980) 15 C.C.L.T. 132. 4. (1980) 117 D.L.R. (3d) 383. 5. (1981) 35 O.R. (2d) 97.

The reasoning of Krever J. is diametrically opposed to Rondel. Rondel gave prominence to the grave consequences that would be likely to flow from failing to recognize a rule of immunity. Krever J., on the other hand, did not perceive the consequences to be so grave and placed reliance on the absence of any empirical evidence supporting the case for immunity. However, for the reasons already given, I cannot accept that the absence of empirical evidence is decisive. Overall, I find the reasoning in Rondel more in tune with what the public interest in the due administration of justice required in Victoria in 1891 and with what it still requires in Australia today.

The purpose of court proceedings is to do justice according to law. That is the foundation of a civilized society. According to our mode of administering justice, parties with inconsistent interests are cast in the role of adversaries and the court or judge is appointed to be an impartial arbiter between them. Counsel (whether barrister or solicitor) may appear to represent the adversaries, but counsel's duty is to assist the court in the doing of justice according to law. A client - and perhaps the public - may sometimes think that the primary duty of counsel in adversary proceedings is to secure a judgment in favour of the client. Not so. The true position was stated by Lord Eldon in Ex parte Lloyd (5 November 1822, reported as a note in Ex parte Elsee [90] ):

He lends his exertions to all, himself to none. The result of the cause is to him a matter of indifference. It is for the court to decide. It is for him to argue. He is, however he may be represented by those who understand not his true situation, merely an officer assisting in the administration of justice, and acting under the impression, that truth is best discovered by powerful statements on both sides of the question. By a paradox which is obvious to any who have experience in our courts, the client is best served by a counsel who is manifestly independent. In representing a client, counsel is expected not only to exercise due skill and diligence but also to do to the best of counsel's ability whatever may legitimately be done in the client's interests, for that is the way in which counsel assists in doing justice according to law: cf. Tombling v. Universal Bulb Co. Ltd. [91] . The privileges of counsel are accorded to that end; they are not accorded to protect counsel. If a duty owed to the client were seen to be separate from the primary duty of assisting to do justice according to law, the two duties might not be wholly compatible. No duty to a client which stands apart from the primary duty can be allowed to impair performance of the primary duty. Counsel who take part in proceedings in court (as well as witnesses and judges) must be able to perform their primary duty free from the chilling threat of civil suit by the parties to the litigation. A similar immunity, granted for similar reasons, attaches to members of Parliament taking part in the proceedings of Parliament. Participants in the proceedings of the courts or in the proceedings of Parliament must not be diverted from the performance of their primary duties by the threat of private litigation. This broad consideration underlies a number of more particular considerations of public policy which favour the rule of counsel's immunity for "in court" conduct, but I am relieved from the need to canvass the particular considerations in this case. They have been canvassed by the judgment of the Chief Justice with which I agree on this aspect of the case and by the speeches in Rondel v. Worsley [92] and Saif Ali v. Sydney Mitchell & Co. [93] . Therefore I would hold the common law to be this: neither a barrister nor a solicitor may be sued by a client in respect of any act done or omission made in the conduct of the client's case in court or in the making of preliminary decisions affecting the way in which the case is to be conducted when it comes to a hearing. 1. (1830) Mont. 69, at p. 70n., at p. 72. 2. [1951] 2 T.L.R. 289, at p. 297. 3. [1969] 1 A.C. 191. 4. [1980] A.C. 198.

If the immunity of counsel were abrogated, the assistance which the courts obtain from the advocacy of an independent profession would be imperilled. But there is a corollary. If counsel generally were to fail to adhere to the standards of advocacy which the courts expect and on which they rely, there would be no justification for the immunity. That has not happened. Hopefully it never will. Unless such a general failure were to occur, it is better to maintain the immunity and to rely on the publicity of court proceedings, judicial supervision, appeals, peer pressure and disciplinary procedures to prevent neglect in the performance of counsel's duty and to avoid any injustice which might result therefrom in an individual case.

I would add, obiter, an observation about the duty of counsel to accept any brief which is offered to him or her at a reasonable fee provided it is in a field in which the counsel ordinarily practises and the counsel is not otherwise committed: the "cab-rank" rule. This rule is of ancient origin. Lord Macmillan in Law & Other Things (1937), p. 179, refers to a rule of the Court of Session in 1532 which provided: "No advocate without very good cause shall refuse to act for any person tendering a reasonable fee under pain of deprivation of his office of advocate." A similar rule may be found in the law of mediaeval France (Forsyth, Hortensius - An Historical Essay on the Office and Duties of an Advocate (1879), pp. 196-197) and Erskine eloquently embraced the rule in his celebrated defence of Tom Paine: see the speech by Sir Hartley Shawcross Q.C. reproduced in Boulton, A Guide to Conduct and Etiquette at the Bar of England and Wales, 6th ed. (1975), p. 22. Whatever the origin of the rule, its observance is essential to the availability of justice according to law. It is difficult enough to ensure that justice according to law is generally available; it is unacceptable that the privileges of legal representation should be available only according to the predilections of counsel or only on the payment of extravagant fees. If access to legal representation before the courts were dependent on counsel's predilections as to the acceptability of the cause or the munificence of the client, it would be difficult to bring unpopular causes to court and the profession would become the puppet of the powerful. If the cab-rank rule be in decline - and I do not know that it is - it would be the duty of the leaders of the Bar and of the professional associations to ensure its restoration in full vigour.

The proposition that barristers and solicitors are immune from suit in respect of the conduct of a case in court is a proposition of the common law. It resolves the common law ground of these appeals in favour of the respondents. But the other, and principal, ground is that a barrister's immunity was abolished in Victoria by s. 5 of the Legal Profession Practice Act 1891 Vict., which was substantially re-enacted in successive consolidations: in s. 10 of the Legal Profession Practice Act 1915, then in s. 10 of the Legal Profession Practice Act 1928 and now in s. 10 of the Legal Profession Practice Act 1958 Vict.. The 1891 Act contained at least three other sections which bear upon the construction of s. 5, namely, ss. 3, 4 and 10. The relevant sections read as follows:

3. Every person who has heretofore been admitted as a barrister shall by virtue of this Act be and is hereby admitted as a solicitor as from the date of the passing of this Act, and in addition to his right to practise as a barrister shall be entitled to practise as a solicitor and to all the rights powers and privileges of a solicitor.

4. Every person who has heretofore been admitted as a solicitor shall by virtue of this Act be and is hereby admitted as a barrister as from the date of the passing of this Act, and in addition to his right to practise as a solicitor shall be entitled to practise as a barrister and to all the rights powers and privileges of a barrister.

5. Every barrister shall in future be entitled to maintain an action for and recover from the solicitor or client respectively by whom he has been employed his fees costs and charges for any professional work done by him. And every barrister shall in future be liable for negligence as a barrister to the client on whose behalf he has been employed to the same extent as a solicitor is now liable to his client for negligence as a solicitor.

10. After the passing of this Act no person shall be admitted to practise as a barrister or a solicitor solely, but every person admitted by the Supreme Court shall be admitted both as a barrister and solicitor.

Section 3 confers on a barrister "heretofore admitted" statutory admission and the right to practise as a solicitor; s. 4 confers on a solicitor "heretofore admitted" statutory admission and the right to practise as a barrister; s. 10 requires that after the passing of the Act, persons be admitted as barristers and solicitors. From 1891 onwards there were three categories of legal practitioners: those who had been admitted as barristers, those who had been admitted as solicitors, and those admitted after 1891 with the joint status of barrister and solicitor. Those admitted after 1891 had no need to rely on the authority to practise conferred by s. 3 or s. 4. Construing s. 5 in this context, the "barrister" who "shall in future be liable for negligence" appears to be the barrister who "has heretofore been admitted" and who, by force of s. 3, has become admitted and entitled "to practise as a solicitor". Section 5 imposes on that barrister liability "in future" for negligence while it gives him a right "in future" to sue for his fees. A person admitted after 1891 with the joint status of barrister and solicitor does not seem to fall within the provisions of s. 5. A joint status could be acquired on admission only by practitioners admitted "in future" (s. 10) and such a practitioner would not need the right to sue conferred by s. 5 and would be liable for negligence as a solicitor without having that liability imposed upon the practitioner by s. 5. Section 5 was directed to barristers already admitted, not to those who were admitted after the 1891 Act came into force. I should have thought that, on its true construction, s. 5 imposed liability on barristers "heretofore admitted" in order to ensure that, if the barrister practised as a solicitor pursuant to the authority conferred by s. 3, the barrister should be liable for negligence in the performance of the functions of a solicitor to the same extent as a solicitor was liable in the performance of those functions. The status of barrister was not to exempt him from that liability. It is noteworthy that no liability is imposed on a solicitor "heretofore admitted" in respect of the functions he might perform if he were to practise as a barrister pursuant to the authority conferred by s. 4.

The relevant provisions of the 1891 Act have been preserved in the later consolidations. The three categories of practitioners have been preserved by the Acts of 1915, 1928 and 1958 (s. 4 in each Act), although the categories of "barrister" and "solicitor" are expressed to relate respectively to a barrister admitted prior to 23 November 1891 (the date on which the 1891 Act was assented to) and a solicitor admitted prior to that date. In each of the consolidating Acts, a provision (s. 5) appears requiring the admission of practitioners with the joint status of barrister and solicitor. Section 5 of the 1891 Act is preserved as s. 10 in each of the consolidating Acts, except for amendments appropriate to the dates on which the later Acts came into effect. The words "in future" no longer appear and, in place of the temporal adverb "now", the phrase "on the twenty-third day of November One thousand eight hundred and ninety-one" appears.

It may be surprising that the 1958 Act should have retained provisions relating to persons who had been admitted as either a barrister or solicitor sixty-seven years earlier: perhaps it is a tribute to the longevity of Victorian practitioners and the caution of the draftsman. At all events, s. 10 of the 1958 Act appears in virtually the same context as that in which s. 5 of the 1891 Act appeared. As the construction of the earlier Act is relevant to the construction of the later (Woolworths Ltd. v. Crotty [94] ) s. 10 of the 1958 Act should be construed as imposing on barristers admitted prior to 23 November 1891 a liability corresponding in extent with a solicitor's liability at that date. It was, and is, a transitional provision the effect of which must surely now be exhausted. This construction of s. 10 makes it irrelevant to the questions arising in this case where the defendants are all alleged to be practitioners with the joint status of barristers and solicitors. (It is probably safe to assume that each of these gentlemen was admitted after 23 November 1891.)

However, in the Full Court it was common ground that every barrister in s. 10(2) of the 1958 Act should betaken to mean "every barrister and solicitor acting as a barrister". I do not think that that was the meaning of "every barrister" in the 1891 Act; when it came into force, it could not have applied to practitioners who were thereafter to be admitted with the joint status. The agreed construction raises considerable difficulties. If "acting as a barrister" in the agreed formulation means performing the functions which, prior to the 1891 Act, barristers alone performed, how is it possible to define the extent of a solicitor's liability in respect of the performance of those functions on 23 November 1891? And if, to meet this point, the "same extent" is taken to connote no more than a standard of care then expected of solicitors in the performance of solicitors' functions, why is the standard of care statutorily defined when a barrister and solicitor is acting as a barrister, and defined by the common law when the same practitioner is acting as a solicitor? I am unable to accept the construction of s. 10(2) agreed in the Full Court. In my opinion, s. 10(2) is not expressed to have any application to practitioners admitted as barristers and solicitors after 23 November 1891 and has never applied to such practitioners.

If, contrary to this view, one adopts the agreed construction, the liability in negligence of a barrister and solicitor who is acting as a barrister must be determined in accordance with s. 10(2) of the 1958 Act. Then it is necessary to ascertain the extent to which "a solicitor was on the twenty-third day of November One thousand eight hundred and ninety-one liable to his client for negligence as a solicitor", the modern barrister and solicitor being liable "to the same extent". What was the relevant liability of a solicitor on 23 November 1891? To answer this question, it is necessary to assume that, on that date, a solicitor might have performed the functions which the defendant practitioners were performing when they were acting as barristers. If, making that assumption, a solicitor in 1891 might have been liable in negligence to a client whose case he was conducting, a modern barrister and solicitor might be held liable in negligence under the statute to a client whose case he or she is conducting. It is submitted that s. 10(2) does produce that result because, it is said, a solicitor was liable for "in court" negligence in 1891. As earlier stated, that is not the common law. Has the law changed since 1891?

If the relevant law of 1891 were statutory or affected by statute, reference to the statute book of 1891 would be required. But it seems to have been accepted in argument, and rightly, that the law governing the extent of a solicitor's liability is entirely common law. A court is not ordinarily concerned to apply to the resolution of a current case a proposition of common law plucked from a moment in history, though the court will often refer to the history of the common law in ascertaining a principle for contemporary application. In declaring and applying the common law to a current case, a court is bound by earlier decisions of courts above it in the hierarchy, for those decisions state what that court is bound to take the common law to be. But when the court is not so bound, it may undertake its own inquiry into the common law and it may depart from earlier decisions. The doctrine of stare decisis requires no greater adherence to precedent, though curial policy may lead a court to adhere to earlier authority which is merely persuasive.

The declaratory theory of judicial precedent treats earlier decisions which are not binding merely as evidence of the common law from which a court may depart if it is satisfied that the decision is not a correct statement of contemporary common law. Thus, in Australian Agricultural Co. v. Federated Engine-Drivers and Firemen's Association of Australasia [95] , Isaacs J. said:

A prior decision does not constitute the law, but is only a judicial declaration as to what the law is. The declaration, unless that of a superior tribunal, may be wrong, in the opinion of those whose present function is to interpret and enforce the law. That was the view of Blackstone (Blackstone's Commentaries, (1765), vol. I, pp. 69-71) and of Jessel M.R.: Osborne to Rowlett [96] . 1. (1913) 17 C.L.R. 261, at p. 275. 2. (1880) 13 Ch. D. 774, at p. 784.

In the view of a court sitting at the present time, earlier decisions which are not binding upon it do not necessarily represent the common law of the earlier time, though they record the perception of the common law which was then current. But if a court, when free to do so, perceives the common law to be different from what it was earlier perceived to be, effect is given to the later perception as truly representing the common law: State Government Insurance Commission v. Trigwell [97] .

It may be objected, of course, that the declaratory theory of judicial precedent embraces the fiction that the common law is an immutable body of doctrine existing from time immemorial, to be perceived by a judicial wisdom which proves to be ever more penetrating with the passing of the years. The reality is different: the common law is judge-made law, albeit made under the restraints of traditional judicial method. Neither the declaratory theory nor the positivist theory of judicial legislation is wholly satisfactory: see Simpson, The Common Law and Legal Theory in Simpson (ed.), Oxford Essays in Jurisprudence (Second Series) (1973), p. 77; Stone, Precedent and Law: Dynamics of Common Law Growth (1985), pp. 185 ff. and Wesley-Smith, Theories of Adjudication and the Status of Stare Decisis in Goldstein (ed.), Precedent in Law (1987), p. 73. Whatever may be said of the theories, a court which must apply a common law principle of an earlier time is faced with a peculiar difficulty. It is one thing for a judge to look to precedents as an aid to the ascertainment of a contemporary rule; it is another to decide, as a judge of an earlier time would have decided, a rule which accords with the jurisprudence of that earlier time. Judges are accustomed to performing the former function; they lack the requisite professional experience and cast of mind to perform the latter function with confidence. The judges of today are aliens to the society of a hundred years ago and, unless the relevant earlier period has produced an applicable precedent of binding authority, a judge could hardly be translated back to the earlier time and, ignoring the experience of the intervening years, declare common law principles different from what the judge now considers the common law to be. Take the present case. Could a judge in 1988 confidently declare the extent to which a solicitor would have been held liable for "in court" negligence prior to 23 November 1891? Notions derived from the speeches in Donoghue v. Stevenson [98] , not to mention those derived from later cases, would have to be put aside and precedents prior to 1891 analyzed from the viewpoint of a judge of that time. It may be that, despite the flaws in the declaratory theory, the uncertainty inherent in endeavouring to apply to present cases the principles current at an earlier time has led some judges to assert the unchanging continuity of the common law. This theory accounts for what Barwick C.J. said in Atlas Tiles Ltd. v. Briers [99] , in reference to the authority of this Court:

There being no authority which binds or course of acceptable decision which compels, the common law is what the Court, informed as to the past course of decision, decides what in truth it is and, of course, always has been. (Emphasis added.) And in Dugan v. Mirror Newspapers Ltd. [1] his Honour said: The Court can, of course, decide what the common law always has been: and, if earlier judicial decision is not to that effect, overrule or depart from such a decision: and the Court can, as it were, extend the principles of the common law to cover situations not previously encountered, or not as yet the subject of binding precedent. There is much to commend an approach which takes a contemporary declaration of common law principle prima facie to be a correct statement of what the common law has always been. Such an approach not only avoids uncertainty in historical research; it is also conducive to an orderly development of the common law, for the notion of continuity of the common law gives much weight to past declarations of its content, while permitting a new departure when required by further reflection or changes in social conditions. 1. [1932] A.C. 562. 2. (1978) 144 C.L.R. 202, at p. 208. 3. (1978) 142 C.L.R. 583, at p. 586.

But the approach has not hardened into a rule. If statute directs the application of a common law principle as of a specified date, it is the duty of the court to ascertain the principle as of that date. In strict theory, I should think such a principle must represent either the settled law of the specified date or the principle which, it is thought, the then ultimate court of appeal would have held the principle to be. But in ascertaining the principle of the earlier time, a presumption that that principle accords with the contemporary common law may be made if there is nothing which clearly displaces it. The presumption is a starting point and some earlier authority or historical fact must appear in order to displace it: e.g., a pivotal case which demonstrates a modern departure or a modern doctrine on which the present common law is based. The authorities cited in this case as bearing on the liability of a solicitor for "in court" negligence prior to 23 November 1891 are, in my opinion, inconclusive. That is not surprising. None of them was addressing that principle which, it is said, s. 10(2) of the 1958 Act requires this Court to ascertain for the purposes of this case.

On the other hand, it is now seen that at common law there is no liability in a solicitor for anything done or omitted in the conduct of a client's case in court or in the making of preliminary decisions affecting the way in which the case is to be conducted when it comes to court. (I leave aside liability for failure to carry out preliminary decisions, whether made by the solicitor or by counsel briefed by the solicitor, when that failure impairs the conduct of the case in court in the way intended.) I would therefore take the common law as this Court declares it today to be the common law of 1891.

I would therefore reject the appellants' case so far as it rests on s. 10(2) of the 1958 Act. The principal reason for this conclusion is that s. 10(2) is a vestigial provision applying only to barristers admitted prior to 23 November 1891 and probably only with respect to their practices as solicitors pursuant to s. 3 of the 1891 Act or its later equivalents. A secondary reason is that, in any event, I take the common law of 1891 to be that neither a barrister nor a solicitor would have been liable to a client in respect of any act or omission in the conduct of the client's case in court. The present law is the same.

At the time when the Legal Profession Practice Act 1891 Vict. ("the Act") was enacted, an action by a client against a professional person, such as a solicitor, for failure to exercise due skill and care was seen as lying in contract alone: see Bean v. Wade [2] . That being so, a consequence of the absence of contractual relations between a barrister and his client was that a barrister was seen as immune from such actions: see In re Le Brasseur and Oakley [3] . The Act "amalgamated the two branches of the legal profession", "altered the relation of counsel to his client" and made it "a contractual one, with legal responsibilities on both sides": per Hodges J., Levy v. Union Bank of Australia Ltd. [4] . In particular, the legislative intent disclosed by s. 5 (and confirmed by s. 8) of the Act was that, thereafter, the legal practitioner (in the amalgamated profession) acting as a barrister should be under a liability for negligence to the client corresponding with the then "clearly recognized" (see Cordery, Law Relating to Solicitors of the Supreme Court of Judicature, 2nd ed. (1888), p. 112-115) contractual liability for negligence of such a legal practitioner acting as a solicitor. It is true that, in this country at least, the ordinary liability of a professional man to his client for negligence has been, in recent times, seen as lying in tort rather than in contract (see Hawkins v. Clayton [5] ). It would however, be inconsistent with the legislative intent to be discerned in the Act (and, in view of the contractual liability established by the Act, pointless) to isolate the legal practitioner in Victoria, on perceived grounds of public policy, from the reach of the modern development of the tort of negligence.

1. (1885) 2 T.L.R. 157, at pp. 158-159. 2. [1896] 2 Ch. 487, at p. 494. 3. (1896) 21 V.L.R. 683, at p. 685. 4. (1987) 164 C.L.R. 539.

In the result, it is strictly unnecessary that I form any concluded view on the important general question which is addressed in the judgments of the members of the Court who constitute the majority in the present case. That question is whether, under the common law of this country, a barrister or other legal practitioner is, in the absence of applicable statutory provision and regardless of fees charged and received, immune from any liability to his or her client for "in court" negligence. I would, however, indicate my dissent from the conclusion reached by the majority of the Court in relation to that question.

There is no decided case which requires this Court to treat a barrister or other legal practitioner acting professionally in court for a client as beyond the reach of the modern common law of negligence. Nor, in my view, is any convincing justification of such an immunity to be found in general principle; plainly enough, the traditional view that the relationship between a barrister and his client is non-contractual does not provide one. If the recognition of such an immunity can be justified, it must be by reference to largely pragmatic considerations of public policy: cf. Rondel v. Worsley [6] ; Saif Ali v. Sydney Mitchell & Co. [7] . In that regard however, I do not consider that the considerations of public policy which are expounded in Rondel v. Worsley [8] and in the majority judgments in the present case outweigh or even balance the injustice and consequent public detriment involved in depriving a person, who is caught up in litigation and engages the professional services of a legal practitioner, of all redress under the common law for "in court" negligence, however gross and callous in its nature or devastating in its consequences.

1. [1969] 1 A.C. 191, at pp. 244, 280-281, 289. 2. [1980] A.C. 198, at pp. 212, 219, 230-231, 233, 235. 3. [1969] 1 A.C., at pp. 227-231, 247-254, 267-276, 281-284. Dawson J.

The question in these appeals is whether in the State of Victoria a barrister is immune from liability for negligence in the conduct of his client's case in court.

It has long been held that such immunity does exist in the United Kingdom, although the basis for it was not clearly identified as being public policy rather than absence of contract until the decision of the House of Lords in Rondel v. Worsley [9] . Subsequently, in Saif Ali v. Sydney Mitchell & Co. [10] , the House of Lords limited the immunity to the actual conduct of a cause in court and to work so intimately connected with the conduct of the cause that it could fairly be said to be a preliminary decision affecting the way in which the cause was to be conducted when it came to a hearing.

It is said that these two English decisions do not represent the law in Victoria, first, because of s. 10 of the Legal Profession Practice Act 1958 Vict. ("the Legal Profession Practice Act ") and, secondly, because the common law is, in any event, different in this country and does not accord immunity from liability for negligent conduct to barristers any more than it does to persons in other professions.

Section 10 of the Legal Profession Practice Act reads:

(1) Every barrister shall be entitled to maintain an action for and recover from the solicitor or client respectively by whom he has been employed his fees costs and charges for any professional work done by him.

(2) Every barrister shall be liable for negligence as a barrister to the client on whose behalf he has been employed to the same extent as a solicitor was on the twenty-third day of November One thousand eight hundred and ninety-one liable to his client for negligence as a solicitor.

The reference to 23 November 1891 is a reference to the date upon which the Legal Profession Practice Act 1891 Vict., which enacted the precursor of s. 10, received Royal Assent. The date was substituted for the words "in future", which appeared in the 1891 enactment.

At the time of separation of Victoria from New South Wales in 1851, the practice of the law was divided between barristers and attorneys in the latter colony. The division was inherited by Victoria, but it was not long before there was agitation for the two branches to be amalgamated. The debate continued and grew stronger in the 1880s and private member's bills were unsuccessfully introduced more than once into Parliament in attempts to abolish the division. It was not until 1891 that the proponents of amalgamation were successful in securing the passage of the original Legal Profession Practice Act.

The pivotal sections of the 1891 Act provided for barristers to be thereby admitted as solicitors (s. 3) and vice versa (s. 4). Each was thereafter to be entitled to practise as the other and to have all the rights, powers and privileges of the other. After the Act came into force every person admitted to practise was to be admitted both as a barrister and solicitor: s. 10. These provisions made the scheme of things plain enough. The Act aimed to make members of the profession both barristers and solicitors, leaving them free to practise as either or both. No doubt it was anticipated by some and hoped by others that the Act would result in the amalgamation in practice of the two branches. For this purpose it was necessary to remove the barriers against the performance of the functions of each but not to eliminate the differences between those functions. In the result, because the Act did not preclude practitioners from practising exclusively in one branch or the other, there was practically no change. The de facto division continued. In the beginning this was due largely to the determined and well-organized efforts of the barristers. Later there was general acceptance of the division and it may now be said that there is a measure of legislative recognition of a separate Bar: see, e.g. Legal Profession Practice Act, Pt IIA.

There is no reason to suppose that the 1891 Act was intended to go further than its wording suggests. In particular, the provision equating the liability for negligence of barristers with that of solicitors paid no attention to the extent of the liability of either at the time the Act came into force. That was not necessary. What was important was that both should in the future be treated equally. The practical way of achieving this was to say that barristers would be liable for negligence to the same extent as solicitors, without spelling out the limits on either side. This is what the Act did. It is obvious that at the time it was thought, correctly, that the liability of solicitors extended beyond that of barristers, but it would be wrong in my view to suppose that any significance was, or needed to be, attached to the precise differences. The object was equal treatment and for this purpose it mattered not whether solicitors were or were not liable for negligence when acting as advocates in court so long as practitioners were thenceforth placed upon an equal footing. It is apparent that when the Act was passed it was thought that the liability of solicitors rested upon the contractual relationship between them and their clients. No such relationship existed between barristers and their clients. No attempt was made, however, in s. 5 of the 1891 Act to alter that relationship other than to provide that barristers should in the future be able to sue for their fees, something which had previously been denied to them. If the result was to establish a contractual relationship between barristers and their clients (see In re Melbourne Parking Station Ltd. [11] ) then it was achieved indirectly. In other words, the section went no further than to bestow upon barristers an advantage enjoyed by solicitors just as in the matter of liability for negligence it went no further than to impose upon barristers the same liability for negligence as that imposed upon solicitors.

Since the question in this case is whether barristers are immune in Victoria from liability for negligence in the conduct of their clients' cases in court and under the Act that depends upon the liability of solicitors in 1891, the first inquiry must be whether solicitors were then liable for negligence when they conducted their clients' cases in court. If they were, then there can be no immunity on the part of barristers.

Until the decision in Rondel v. Worsley there can be little doubt that the liability of solicitors for negligence was thought to be based upon the contract which existed between them and their clients by reason of their retainer. Before 1891 solicitors in Victoria had a right of audience in Magistrates' Courts, County Courts and Courts of Mines. Similarly their counterparts in England were entitled to appear before inferior tribunals. There was, therefore, the possibility of negligence on the part of solicitors in the conduct of proceedings in court, and, indeed, there are cases in which solicitors were before 1891 held liable for negligence which was at least related to the function of an advocate. Reliance was placed upon those cases: Stokes v. Trumper [12] ; Montriou v. Jefferys [13] ; Hart v. Frame [14] ; Swannell v. Ellis [15] ; Smith v. Grant and Leslie [16] ; Purves v. Landell [17] ; Manley v. Palache [18] ; Fergusson v. Lewis [19] ; Clarke v. Couchman [20] ; Murray v. Reilly [21] . An examination of these decisions reveals, however, that with the possible exception of Stokes v. Trumper , none of them is concerned with negligence in actual advocacy before a court. They are for that reason of little assistance, but in any event the liability which they recognize is clearly based upon the view prevailing at the time that solicitors or their equivalent were liable for negligence because of their contract with their clients. This appeared to be sufficient to differentiate them from barristers and no need seems to have arisen to distinguish the activities of solicitors out of court from their activities in court. Immunity on the part of barristers was not then seen to be based as much upon public policy as upon the absence of contract. Their situation was described by Pollock C.B. delivering the judgment of the Court of Exchequer Chamber in Swinfen v. Lord Chelmsford [22] . After referring to Fell v. Brown [23] , a case in which the plaintiff was non-suited in an action against a barrister for unskilfully and negligently settling and signing a bill in equity, he said [24] :

We are all of opinion that an advocate at the English bar, accepting a brief in the usual way, undertakes a duty, but does not enter into any contract or promise, express, or implied. Cases may, indeed, occur, where, on an express promise (if he made one), he would be liable in assumpsit; but we think a barrister is to be considered, not as making a contract with his client, but as taking upon himself an office or duty, in the proper discharge of which not merely the client, but the Court in which the duty is to be performed, and the public at large, have an interest. See also Kennedy v. Broun [25] ; Perring v. Rebutter [26] . Whilst there is more than a hint of public policy in the passage which I have just cited, it was the absence of contract or the absence of any right to sue for his fees which was generally regarded as the basis of a barrister's immunity from liability for professional negligence. 1. (1855) 2 K. & J. 232 [69 E.R. 766]. 2. (1825) 2 Car. & P. 113 [172 E.R. 51]. 3. (1839) 6 Cl. & F. 193 [7 E.R. 670]. 4. (1823) 1 Bing. 347 [130 E.R. 140]. 5. (1858) 20 Dunlop 1077; 20 S.C. 1077. 6. (1845) 12 Cl. & F. 91 [8 E.R. 1332]. 7. (1895) 73 L.T. 98. 8. (1879) 14 L.J. 700. 9. (1885) 20 L.J. 318. 10. [1963] S.L.T. (Notes) 49. 11. (1860) 5 H. & N. 890 [157 E.R. 1436]. 12. (1791) Peake 131 [ 170 E.R. 104]. 13. (1860) 5 H. & N., at p. 920 [157 E.R., at p. 1448]. 14. (1863) 13 C.B. (N.S.) 677 [143 E.R. 268]. 15. (1842) 2 M. & Rob. 429 [174 E.R. 340].

Hedley Byrne & Co. Ltd. v. Heller & Partners Ltd. [27] showed that a duty of care is imposed, apart from any contract, upon persons exercising professional skills and this made it apparent that it was no longer sufficient to support a barrister's immunity from liability for negligence solely by reference to the absence of a contract between him and his client or to his inability to sue for his fees. See also now Hawkins v. Clayton [28] . The true foundation for the immunity was found in Rondel v. Worsley to lie in public policy. Liability for negligence arising from a professional relationship apart from contract was not novel - see Nocton v. Lord Ashburton [29] - and the public aspect of a barrister's function - perhaps duty more than policy - had always been recognized. However, it was the light thrown by Hedley Byrne & Co. Ltd. v. Heller & Partners Ltd. which required the true basis for a barrister's immunity from liability for professional negligence to be identified. And, of course, if the reason for that immunity lay in public policy, as was ultimately found to be so in Rondel v. Worsley, then there could be no justification for denying the same immunity to solicitors when performing the same functions as those performed by barristers.

1. [1964] A.C. 465. 2. (1988) 164 C.L.R. 539. 3. [1914] A.C. 932.

The immunity of solicitors for negligence as advocates in court was readily accepted in Rondel v. Worsley with little discussion. Indeed, no elaboration was necessary because the conclusion follows ineluctably from the reasoning in that case: see per Lord Reid [30] ; Lord Morris [31] ; Lord Pearce [32] ; Lord Upjohn [33] ; Lord Pearson [34] ; see also Saif Ali v. Sydney Mitchell & Co. [35] , per Lord Wilberforce; Lord Salmon [36] .

1. [1969] 1 A.C., at pp. 230-231. 2. [1969] 1 A.C., at pp. 243-244. 3. [1969] 1 A.C., at pp. 265-267. 4. [1969] 1 A.C., at pp. 284-285. 5. [1969] 1 A.C., at p. 294. 6. [1980] A.C., at p. 215. 7. [1980] A.C., at p. 227.

The question therefore becomes whether, apart from statute, Rondel v. Worsley and Saif Ali v. Sydney Mitchell & Co. represent the law in this country and, if so, whether those two decisions represent a shift in position from that which obtained in 1891.

Victoria inherited the common law through New South Wales and there can be no doubt that immediately before the 1891 Act came into force the position in Victoria with respect to the immunity of barristers from liability for professional negligence was the same as it was in England. Rondel v. Worsley did not alter the extent of the liability of barristers or solicitors in England. The extent of the immunity remained the same; the development of the law of negligence had merely made it plain in the meantime that it must rest upon public policy rather than the absence of contract. That in turn demonstrated that the immunity of solicitors from liability for negligence as advocates in court could be no different from that of barristers - a matter which until then had never been conclusively determined. In my view it cannot be said that Rondel v. Worsley or Saif Ali v. Sydney Mitchell & Co. represent anything other than the common law as it existed in 1891 explained, perhaps, upon a basis which had previously been given an insufficient emphasis.

Nevertheless, this Court is not compelled to follow the two English decisions. It could, if it thought that there was any force in it, take the view that the true basis, and only basis, at common law for the immunity of a barrister from liability for professional negligence has always been the absence of any contractual relationship between him and his client and that, in the light of Hedley Byrne & Co. Ltd. v. Heller & Partners Ltd. and Hawkins v. Clayton, it is no longer a sufficient basis. To take such a view would, of course, be to regard Rondel v. Worsley as having changed the common law in England since 23 November 1891 and that of itself would be sufficient under s. 10 of the Legal Profession Practice Act to require the liability of a solicitor for professional negligence to be determined without reference to the decision. It will be apparent from what I have already said that I do not regard Rondel v. Worsley as having changed the extent of the liability of a barrister or solicitor for professional negligence, but it is desirable that I go on and express the reasons why I think it is appropriate to find in public policy the basis for the immunity from liability for negligence of a barrister or solicitor acting as an advocate in court.

More than one reason was advanced in the speeches in Rondel v. Worsley and Saif Ali v. Sydney Mitchell & Co. A barrister, it was said, owes his primary duty to the court and not his client and in following his duty, which at times may involve the immediate determination of difficult questions, he ought not to be hampered by the prospect of liability to his client for a wrong decision. In Saif Ali v. Sydney Mitchell & Co. [37] , Lord Diplock did not think this a sufficient reason to differentiate barristers from other professional men who also have to make difficult decisions on the spot. Professional negligence consists of more than a mere error of judgment and I must confess that I find it difficult to see any inconsistency between the existence of a duty to the court and the existence of a duty of care to a client. Nor do I see any great force in the argument that immunity is to be justified by the cab-rank principle which requires a barrister to accept instructions to act on behalf of a client, provided he is free, a proper fee is tendered and he practises in the particular jurisdiction. Again like Lord Diplock (in Saif Ali v. Sydney Mitchell & Co. [38] ), I have never thought the principle to have as much practical operation as is sometimes suggested and it is hardly enough, to my mind, to differentiate the profession of a barrister from other professions.

1. [1980] A.C., at pp. 219-221. 2. [1980] A.C., at p. 221.

Of more weight in terms of public policy is the suggestion that a barrister would be impaired in the exercise of his professional judgment by the prospect of an action against him by a disappointed client. He would, the suggestion continues, be led out of caution to put unnecessary arguments, to take unnecessary defences, to ask unnecessary questions and to call unnecessary witnesses, thereby becoming not only prolix but also less effective in the art of advocacy. Not only might the interests of the client be adversely affected but the efficient conduct of the business of the courts would be likely to be impaired.

But there are weightier considerations than these. The first, to my mind, is that the availability of an action in negligence for the conduct of a case in court would subject the decision of the court to collateral attack by a client who sought to blame his barrister for his loss of the case. Not only would this mean relitigation of issues already decided, but the relitigation would be before a different tribunal after a lapse of time upon evidence which would not necessarily be the same. This would be bad enough after a decision in a civil case but would be intolerable after a criminal trial. The verdict of the jury would be impugned in a court of law and yet, assuming all avenues of appeal to have been exhausted, it would remain with all its consequences. True it is that the way in which a trial has been conducted by a practitioner appearing for an accused may afford a ground for appeal if it results in a miscarriage of justice (see, e.g. Re Knowles [39] ), but to contemplate an attack in collateral proceedings which would be incapable of affecting the verdict, raises quite different considerations. Nothing could be more calculated to destroy confidence in the processes of the courts or be more inimical to the policy that there be an end to litigation. If the decision of a court is wrong, the appeal process is the means by which it should be corrected. To allow the courts to be used to undermine its authority in other proceedings is clearly not in the public interest.

However, the most cogent consideration is, to my mind, more broadly based in policy which is not confined to immunity from liability for negligence in the presentation of a client's case in court. Fundamental to the administration of justice is the opportunity which the law affords to all those who are participants in proceedings in a court to speak and act freely, within the rules laid down, unimpeded by the prospect of civil process as a consequence of their having done so. This privilege against civil liability - for privilege it is - extends beyond the parties and their representatives to witnesses, the court officials and the judge himself. As Starke J. said in Cabassi v. Vila [40] :

No action lies in respect of evidence given by witnesses in the course of judicial proceedings, however false and malicious it may be, any more than it lies against judges, advocates or parties in respect of words used by them in the course of such proceedings or against juries in respect of their verdicts The law protects witnesses and others, not for their benefit, but for a higher interest, namely, the advancement of public justice Whilst the freedom spoken of in Cabassi v. Vila was freedom from actions for defamation or the like, the same considerations clearly apply to an advocate's immunity from actions in negligence. This appears strikingly in the judgment of Brett M.R. in Munster v. Lamb [41] , also a case dealing with an action for defamatory words uttered in court, on this occasion by a solicitor acting as an advocate. He said: If upon the grounds of public policy and free administration of the law the privilege be extended to judges and witnesses, although they speak maliciously and without reasonable or probable cause, is it not for the benefit of the administration of the law that counsel also should have an entirely free mind? Of the three classes - judge, witness, and counsel - it seems to me that a counsel has a special need to have his mind clear from all anxiety. Brett M.R. goes on to point out that it is for the benefit of the public that an advocate in court should be free from the constraints which the possibility of civil action would necessarily impose. I can see no relevant difference for this purpose whether the action be for damages for defamation or for negligence. And it is to be remembered that in either case it is protection against actions which are ill founded as much as against those which are soundly based which is necessary. It is the contemplation of either type of suit which must be likely to hamper an advocate's efforts so as to deprive the administration of justice of the assistance which it might otherwise expect. In this respect the practice of the profession of an advocate differs from the practice of other professions. To err on the side of caution is not only practicable but ordinarily the best course with other professions. With an advocate this may be fatal, not only to the interests of the client, but also to the proper determination of the case. 1. (1940) 64 C.L.R. 130, at pp. 140-141. 2. (1883) 11 Q.B.D. 588, at p. 603.

I should add that in New Zealand, where the two branches of the profession are amalgamated, Rondel v. Worsley has been applied. Indeed, it was in Rees v. Sinclair [42] , a decision of the Court of Appeal, that McCarthy P. [43] propounded the test which was applied in Saif Ali v. Sydney Mitchell & Co. to limit the extent of the immunity to matters intimately connected with the conduct of a cause in court. And in South Australia, where the profession is also amalgamated to a substantial degree, in Feldman v. A Practitioner [44] , Bray C.J. accepted the decisions of Rondel v. Worsley and Saif Ali v. Sydney Mitchell & Co. Despite the American and Canadian experience, which appear to have been different, I have reached the conclusion that the reasoning upon which the two English decisions are based has a clear application here and that those cases ought to be followed.

1. [1974] 1 N.Z.L.R. 180. 2. [1974] 1 N.Z.L.R., at p. 187. 3. (1978) 18 S.A.S.R. 238.

It is as well at the outset to identify the persons with whom these appeals are concerned.

There are four appellants. The first three appellants ("the Giannarellis") were earlier charged with criminal offences arising from answers given by them to the Royal Commission into the affairs of the Federated Ship Painters' and Dockers' Union. The action, the subject of this appeal, is an action in negligence brought by the Giannarellis against their legal representatives on the ground that the latter failed to raise in answer to the charges a point which was subsequently held by this Court to provide a complete defence: see Giannarelli v. The Queen [45] . The other appellant, Mr. Shulkes, acted as solicitor for the Giannarellis at their committal proceedings. He is an appellant by reason of a claim for indemnity or contribution made by him against the other defendants, he being one of the defendants in the action. The matter came before this Court in the form of two appeals, the Giannarellis being the appellants in one and Mr. Shulkes the appellant in the other.

There are three respondents. Mr. Wraith, of counsel, appeared for the Giannarellis at the committal proceedings. Mr. Francis, one of Her Majesty's counsel, was senior counsel for them at their trial. Mr. Hedigan, also one of Her Majesty's counsel, was senior counsel for the Giannarellis at their appeal to the Court of Criminal Appeal. There are other defendants to the action brought by the Giannarellis, who are not respondents to the appeal. Some were their counsel and some were their solicitors, at different stages of the proceedings against them.

As the matter came before the Full Court of the Supreme Court of Victoria (and as the matter now comes before this Court), three questions of law arose for determination. They were:

(a) As a matter of law is the first Appellant immune from all liability to the first, second and third respondents on the facts alleged against him in the Statement of Claim on the grounds that as a matter of public policy he is immune from all and any liability for negligence or breach of retainer arising out of any act or omission on his part during his conduct of the said Respondents' defence at the committal hearing?

(b) As a matter of law is the second Appellant immune from all liability to the first, second and third Respondents on the facts alleged against him in the Statement of Claim on the grounds that as a matter of public policy he is immune from all and any liability for negligence or breach of retainer arising out of any act or omission on his part during his conduct of the said Respondents' defence at their trial?

(c) As a matter of law is the third Appellant immune from all liability to the first, second and third Respondents on the facts alleged against him in the Statement of Claim on the grounds that as a matter of public policy he is immune from all and any liability for negligence or breach of retainer arising out of any act or omission on his part during his conduct of the said Respondents' application to the Court of Criminal Appeal for leave to appeal against their convictions and sentences at the trial?

The appellants referred to in the passage quoted are of course the respondents in the appeal to this Court; the first, second and third respondents mentioned are the Giannarellis.

The Full Court answered each of these questions in the affirmative; the appellants contend that in so doing the Court fell into error. In Victoria these questions cannot be answered solely by reference to the common law. There is an existing statutory provision, as there have been past statutory provisions, that bears directly on the answers. The current provision is s. 10 of the Legal Profession Practice Act 1958 Vict. which reads:

(1) Every barrister shall be entitled to maintain an action for and recover from the solicitor or client respectively by whom he has been employed his fees costs and charges for any professional work done by him.

(2) Every barrister shall be liable for negligence as a barrister to the client on whose behalf he has been employed to the same extent as a solicitor was on the twenty-third day of November One thousand eight hundred and ninety-one liable to his client for negligence as a solicitor.

The appellants argue that, by reason of s. 10, the respondents are liable in negligence whether that negligence occurred in court or out of court. The appellants submit in the alternative that the respondents are so liable at common law, a submission which can only succeed if certain decisions of the House of Lords are not followed in this country.

The origin of s. 10 is to be found in the Legal Profession Practice Act 1891 Vict., s. 5 of which reads:

Every barrister shall in future be entitled to maintain an action for and recover from the solicitor or client respectively by whom he has been employed his fees costs and charges for any professional work done by him. And every barrister shall in future be liable for negligence as a barrister to the client on whose behalf he has been employed to the same extent as a solicitor is now liable to his client for negligence as a solicitor.

As to later legislation, s. 10 of the Legal Profession Practice Act 1915 Vict. was in terms virtually identical with s. 10 of the current statute. So too was s. 10 of the Legal Profession Practice Act 1928 Vict..

To understand these various sections, it is necessary to go back to the 1891 Act and ask why s. 5 was introduced and what was its purpose. This in turn requires consideration of some other sections of that Act and some reference to the organization of the legal profession in Victoria in 1891. That profession was organized in much the same way as its English counterpart at the time. It was divided into barristers on the one hand and attorneys, solicitors and proctors on the other. "Solicitor" was the name generally given to the latter, a usage which was carried through into the statute law of Victoria. Barristers had exclusive right of audience before the Supreme Court of Victoria. Descriptions of the English profession may be found in Holdsworth, A History of English Law, vol. VI (1924), pp. 431-457, vol. XV (1965), pp. 242-244; Baker, Counsellors and Barristers, The Legal Profession and the Common Law (1986), Ch. 8 .

In Victoria there was, from 1880 onwards, a number of private members' bills for the amalgamation of the legal profession. This followed moves for fusion in England that gathered some impetus about the middle of the nineteenth century. Although the organization of the profession in the United States found favour with some eminent English lawyers who were or had been themselves barristers, nothing came of those moves. In Victoria amalgamation was achieved by the Legal Profession Practice Act 1891 which came into operation on 1 January 1892. The circumstances giving rise to this legislation are described at some length in Dean, A Multitude of Counsellors: A History of the Bar of Victoria (1968), Ch. 6 and Forbes, The Divided Legal Profession in Australia: History, Rationalisation Rationale (1979), Pt 1, Div. 4.

Having defined "barrister" to mean "a barrister of the Supreme Court" and "solicitor" to mean "an attorney solicitor and proctor of the said court" (s. 2), the 1891 Act went on to provide that every person who had been admitted as a barrister was thereby admitted as a solicitor and conversely every person who had been admitted as a solicitor was thereby admitted as a barrister (ss. 3, 4). After the passing of the Act, no person was to be admitted to practise "as a barrister or a solicitor solely, but every person admitted by the Supreme Court shall be admitted both as a barrister and solicitor": s. 10. To meet what the parliamentary debates of the time suggested may have been something of a problem, it was provided that any barrister and solicitor who did not give "substantial attendance" to a matter in court might lose all or part of his fee and that any contract to the contrary was of no effect: s. 8. Dean, p. 101, said of the legislation:

It seems to have been supposed that once the Act was passed there would no longer be barristers and there would no longer be solicitors. Every practitioner alike would be a barrister and solicitor and would practise in both capacities; every barrister would leave the Bar and join a solicitors' firm or start one of his own, and see clients, draw wills, do probate business and conveyancing business and appear in Court, as his practice required. But owing to the ingenuity, resource and the will of the Bar to survive the practical result was nil.

The Full Court pointed out in its judgment, speaking of s. 5 of the 1891 Act [46] :

It might be thought strange that in an Act which effectively abolished the distinction between barristers and solicitors, a section should be found dealing with the rights and liabilities of a barrister in the future. It was, however, common ground during the argument before us that the words "every barrister" wherever appearing in the section should be taken to mean "every barrister and solicitor acting as a barrister". This interpretation is obviously correct. None of the parties sought to resile from that position. 1. [1988] V.R. 713, at p. 719.

There were two limbs to s. 5 of the 1891 Act, as there are two limbs to s. 10 of the 1958 statute. The first conferred an entitlement on barristers to recover fees for professional work done. The second rendered barristers liable for negligence, though on what terms and in what circumstances is a matter which was debated at length in this Court. One matter which arises is whether and to what extent the two limbs are interrelated.

It may be taken for granted that in 1891 a barrister, as that term was understood outside the Legal Profession Practice Act of that year, could not be sued for negligence in the conduct of litigation. This immunity had been acknowledged in many decisions including Fell v. Brown [47] ; Swinfen v. Lord Chelmsford [48] ; and Kennedy v. Broun [49] . No distinction was drawn between a barrister and a certificated special pleader: Perring v. Rebutter [50] . It was equally well understood that a barrister could not sue for his fees. Lord Pearce suggested in Rondel v. Worsley [51] , that "the hypothesis that the immunity stems from the inability to sue for fees is unsound". His Lordship considered that the immunity lay in public policy, being derived from the special position of a barrister in the conduct of litigation and from the duties he owed, not only to his client but to the court. This position had been emphasized by Erle C.J. in Kennedy v. Broun. Again, to quote from Lord Pearce in Rondel v. Worsley [52] :

Such thoughts as these, based upon public policy, plainly underlay both the inability to sue and the inability to be sued. It is clear that one did not spring from the other. They both sprang independently from similar public considerations. 1. (1791) Peake 131 [ 170 E.R. 104]. 2. (1860) 5 H. & N. 890 [157 E.R. 1436]. 3. (1863) 13 C.B. (N.S.) 677 [143 E.R. 268]. 4. (1842) 2 M. & Rob. 429 [174 E.R. 340]. 5. [1969] 1 A.C. 191, at p. 261. 6. [1969] 1 A.C., at p. 262.

However, the notion that the immunity of the barrister had no connexion with his inability to sue for his fees is open to considerable argument. The inability of barristers to sue for their fees is a principle of long standing. It was certainly recognized in the seventeenth century: Baker, pp. 118-123. And in Swinfen v. Lord Chelmsford [53] , Pollock C.B., speaking for the members of the Court of Common Pleas sitting in banc, emphasized that "a barrister is to be considered, not as making a contract with his client, but as taking upon himself an office or duty " Whatever the origin of the barrister's immunity from suit, there can be little doubt that by the end of the nineteenth century it was closely linked with his inability to sue. Thus, in In re Le Brasseur and Oakley [54] , Lindley L.J. commented:

But I think it is of the utmost importance that the Court should not assist barristers to recover their fees. If they do so, the whole relation between a barrister and his professional client will be altered, and a door will be opened which will lead to very important consequences as regards counsel. The inevitable result will be to do away with that which is the great protection of counsel against an action for negligence by his client. In Rondel v. Worsley Lord Pearce's view as to the relationship between a barrister's immunity from negligence and his inability to sue for his fees was not shared by Lord Pearson who said [55] : It has been suggested that the reasoning of the Court of Appeal, especially Lord Lindley, in the case of In re Le Brasseur and Oakley was defective, in that it was being said that because a barrister could not sue for his fees therefore he must be immune from any liability for negligence. But I think that this criticism is based on a misunderstanding of the reasoning, which was entirely sound. The reasoning was that the relation between the barrister and the client (acting through his solicitor) is non-contractual and does not create legal rights or obligations on either side, so that the barrister can neither sue for his fees nor be sued for negligence; but if the courts were to decide that the barrister could sue for his fees, the decision would mean that there is a contractual relationship, and it would follow that he could be sued for negligence in the conduct of a case. 1. (1860) 5 H. & N., at p. 920 [157 E.R., at p. 1448]. 2. [1896] 2 Ch. 487, at p. 494. 3. [1969] 1 A.C., at p. 293.

The implications of the 1891 Act do not seem to have arisen for any detailed consideration by the courts. There is a passing comment by Hood J. in G. v. G. [56] . Criticizing the conduct of the respondent's proctors who, in divorce proceedings, withdrew a denial of adultery a few days before the hearing but persisted in countercharges of cruelty, connivance and misconduct conducing to the adultery (if any), his Honour said: "The only excuse for this line of procedure was that counsel so advised. I doubt if counsel's advice is any protection to a solicitor since the amalgamation Acts, even on points of law " See also Levy v. Union Bank of Australia Ltd. [57] where Hodges J. said of the Act (referred to as Act No. 1216): "It has of course altered the relation of counsel to his client. It has made such relation a contractual one, with legal responsibilities on both sides "

1. [1908] V.L.R. 420, at p. 426. 2. (1896) 21 V.L.R. 683, at p. 685.

It is necessary to turn now to s. 10(2) of the 1958 Act, bearing in mind that the sub-section has its origin in the 1891 statute and that it identifies the liability for negligence of a barrister as "to the same extent as a solicitor was on [23 November 1891] liable to his client for negligence as a solicitor".

The terminology used is far from clear. What is meant by "to the same extent as"? The Full Court rejected the argument that "extent" is a reference to a measure of liability. In the view of the Court [58] :

It would be meaningless to enact that a barrister was to be liable to the same measure of liability as a solicitor, for the measure of the liability of a solicitor in a given case would be the damage flowing from any negligence established. Further, the extent of a solicitor's liability to his client for negligence may be affected by the contract which he has with his client and although sub-section (1) creates what might be called a statutory contract between a barrister and the solicitor and the lay client nothing in it suggests that the statutory contract might be varied by agreement to exclude liability for negligence. 1. [1988] V.R., at p. 720.

There can be no argument with these comments nor, I think, with the view of the Full Court that the phrase "to the same extent" should be understood as meaning "on the same basis" or "in the same way". The inquiry then must be as to the basis on which or the way in which a solicitor was liable to his client for negligence as a solicitor on 23 November 1891. The Full Court was of the opinion that this inquiry required an answer to the question: Was a solicitor in 1891 liable to his client for negligent advocacy or the negligent handling of a case in Court? Whether this was the right question to ask lies at the heart of this appeal.

Dr. Pannam, senior counsel for the Giannarellis, submitted that the Full Court had misinterpreted his argument. He agreed that the phrase "to the same extent as" is not concerned with the measure of loss. Rather, he submitted, it is concerned with indicating the measure of a barrister's liability in the sense of identifying the relevant duty of care owed by a barrister to his lay client. He argued that the content or measure of a barrister's liability is defined by reference to the extent of the liability of solicitors for negligence as solicitors on 23 November 1891. Then, as now, solicitors owed a duty of care to their clients. The phrase "to the same extent as" in s. 10(2) thus identifies the measure of a barrister's liability by reference to this same duty of care. On this interpretation, s. 10(2) operates by imposing a statutory liability upon a barrister for negligence in relation to his activities as a barrister. What is that liability? That, counsel submitted, is answered by the words "to the same extent as" which identify the measure of a barrister's liability by reference to the liability of a solicitor for negligence as a solicitor. In other words, solicitors and barristers both owe a duty of care to their clients which, if breached, renders them liable for negligence.

Dr. Pannam's submission as to the meaning of the phrase "to the same extent as" was supported by Mr. Heerey, senior counsel for Mr. Shulkes, though Mr. Heerey's primary submission was that the phrase related to the standard of care owed rather than the area of liability. In other words, the phrase ensured that barristers owed the same standard and duty of care as solicitors and not some higher standard.

In my view Dr. Pannam's submission should be accepted. But even if the Full Court did ask the right question, I do not think that it arrived at the right answer.

Following an examination of nineteenth century cases in England, the Full Court reached the conclusion that in 1891 solicitors were not liable for "in-court negligence" and that s. 10(2) of the Legal Profession Practice Act 1958 did "not alter the common law position of barristers in relation to similar negligence". It is not surprising that no authority could be found in support of the proposition that a solicitor was liable for "in-court negligence" because solicitors had no right of audience in superior courts. The Court then proceeded to examine a number of authorities, in particular Rondel v. Worsley, and concluded that Victorian public policy supported the common law immunity from suit of barristers, and solicitors acting as barristers, "in respect of anything done in court in the course of the conduct of a case".

If the intention of the legislature in 1891 was to do no more than equate the position of barristers as advocates with that of solicitors as advocates, s. 5 of the 1891 Act was a curious way of achieving that end. It is clear from a reading of the section, and from its place in the context of the statute generally, that the intention was to impose some liability on barristers that hitherto had not existed. Section 5 provided that "every barrister shall in future be liable for negligence " That is the language of s. 10(2) of the 1958 Act, save that the words "in future" no longer appear. Merely to have given a barrister a right to recover fees would not necessarily have attracted any liability. There is no doubt that in the exercise of his functions as a solicitor, a solicitor was liable in negligence to his client. It is logical enough that in a statute seeking to amalgamate the two branches of the profession, but recognizing that the functions of each were different and might continue to be different, the legislature should have provided that a barrister too would be liable in negligence while exercising his functions as a barrister. Put another way, the barrister was to be liable in negligence for what he did as a barrister, just as the solicitor was liable in negligence for what he did as a solicitor. That is not to say that there are not difficulties in attaching some precision to this liability in negligence. Thus, in Purves v. Landell [59] , it was said that an attorney was only responsible in damages to his client for gross negligence. But these difficulties only point up the rather unsatisfactory way in which this matter comes before the Court, seeking answers to questions shorn of any findings of fact. Nevertheless the questions must be answered.

It may not be right, as Dr. Pannam suggested, that the price to be paid by barristers for the right to recover fees was to lose their immunity from suit. One consequence was not necessarily dependent upon the other. In any event that suggested interrelationship is not essential to the notion that the immunity from suit hitherto enjoyed by barristers came to an end in Victoria in 1891. Many cases were cited to Marks J. (the primary judge), to the Full Court and to this Court concerning the liability of solicitors. One thing is tolerably clear; there is no authority to support the proposition that solicitors were immune from suit for negligence in the conduct of litigation. There is a wealth of authority tending the other way though there is room for debate as to whether any of the cases cited are truly cases of "in-court negligence". But, as Marks J. pointed out, in those cases "no distinction was made between conduct in or associated with Court and other work". I borrow his Honour's summary of those decisions:

See Reece v. Righey [60] , failure to have witnesses in court; Swannell v. Ellis [61] , failure to appear before arbitrators; Price v. Bullen [62] , failure to procure attendance of witnesses; Montriou v. Jeffreys [63] , failure to give proper advice; Godefroy v. Dalton [64] , failure properly to conduct case, not making sufficient proof; Godefroy v. Jay [65] , allowing judgment to go by default; Hart v. Frame [66] , instituting legal proceedings under wrong statutory provision; Ritchie v. McCrosty [67] , negligent conduct of case, failure to make points and assemble witnesses; Urquhart v. Grigor [68] Privy Council, misconduct of litigation; Griffiths v. Murphy [69] , failure to make sufficient proof of case; Smith v. Grant and Leslie [70] , failure to have proper authentication of a warrant of imprisonment of a servant under Stat. 4 Geo. IV c.34; Mercer v. King [71] , engaging an alcohol-prone surveyor to make essential measurements; Fray v. Voules [72] , settling without client's authority even though on advice of counsel; Hatch v. Lewis [73] , failure to call witnesses; In re Spencer [74] , application to court on wrong basis for payment of money out of a fund; Bullen v. Hooper [75] , bringing action in wrong court; Ibbotson v. Shippey [76] , advice that claim could not be made in bankruptcy although it could; Fergusson v. Lewis [77] , a recorder's decision rejecting solicitors being on same footing as barristers; failure to attend court to appear for client; Clarke v. Couchman [78] , handing over conduct of case to another solicitor; Manley v. Palache [79] Privy Council, misconduct of litigation, failure to examine opponent's witnesses. 1. (1821) 4 B. & Ald. 202 [106 E.R. 912]. 2. (1823) 1 Bing. 347 [130 E.R. 140]. 3. (1824) 3 L.J. (K.B.) 39. 4. (1825) 2 Car. & P. 113 [172 E.R. 51]. 5. (1830) 6 Bing. 460 [130 E.R. 1357]. 6. (1831) 7 Bing. 413 [131 E.R. 159]. 7. (1839) 6 Cl. & Fin. 193 [7 E.R. 670]. 8. (1854) 16 Dunlop 554. 9. (1857) 19 Dunlop 853. 10. [1856] Vic. Law Times 3 (F.C.). 11. (1858) 20 Dunlop 1077. 12. (1859) 1 F. & F. 490 [175 E.R. 822]. 13. (1859) 1 El. & El. 839 [120 E.R. 1125]. 14. (1861) 2 F. & F. 467 [175 E.R. 1145]. 15. (1870) 39 L.J. (Ch.) 841. 16. (1871) 2 V.L.R. (L.) 108 (Full Court). 17. (1879) 23 Sol. Jo. 388. 18. (1879) 14 L.J. 700. 19. (1885) 20 Law Jo. 318. 20. (1895) 73 L.T. 98.

These decisions and the standard texts of the nineteenth century relating to solicitors, referred to by Marks J., clearly establish the liability of a solicitor for failing adequately to carry out his functions as a solicitor. The texts to which his Honour referred are Pulling, A Summary of the Law and Practice Relating to Attorneys, 3rd ed. (1862), p. 424; Cordery, Law Relating to Solicitors of the Supreme Court of Judicature, 2nd ed. (1888), pp. 112-115; Beven, Principles of the Law of Negligence (1889), p. 780; White, A Treatise on the Constitution and Government of Solicitors, Their Rights and Duties (1894), p. 175; Halsbury's Laws of England, 1st ed. vol. 26 (1914), pars 1251, 1253.

Cordery, writing in 1888 some three years before the 1891 statute in Victoria said, at p. 115:

In the colonies, where the union of the two characters of solicitor and advocate is frequent, the liability of the solicitor in his character of solicitor is clearly recognised.

The authority cited in support of that proposition is Leslie v. Ball [80] , a Canadian decision. It is of interest, bearing in mind the availability of Cordery to legislators and lawyers at the time of passing of the Victorian statute, that of the report Adam Wilson J. commented [81] :

I am not, therefore, prepared to say that a counsel in this country, even although he is not the attorney also, is exempt from liability to his client for such negligence on his part of the conduct of the cause as would make the attorney liable for negligence in his particular portion of it. But I think there is no doubt that a counsel who is also the attorney in the cause is certainly liable for his neglect as counsel, in like manner and to the same extent as an attorney is. 1. (1863) 22 U.C.Q.B. 512. 2. (1863) 22 U.C.Q.B., at p. 519.

The similarity between the concluding words of this passage and s. 5 of the 1891 statute is marked.

It is of further interest that in Feldman v. A Practitioner [82] Bray C.J. reached the same conclusion as Cordery had reached nearly one hundred years earlier. After referring to the immunity for barristers recognized in Rondel v. Worsley and in the judgments in the Court of Appeal in Saif Ali v. Sydney Mitchell & Co. [83] and the extension of that immunity in a fused profession to a solicitor-barrister when acting as a barrister (Rees v. Sinclair [84] ), Bray C.J. said [85] :

Nevertheless, of course, a solicitor-barrister remains liable to an action for negligence for what he does while acting as a solicitor. Clearly he cannot by assuming the dual role acquire an immunity that he would not have had if he had acted as solicitor alone and briefed other counsel. And when he performs both roles it will often be a question of some nicety whether any particular act, omission or decision emanated from the solicitor or from the barrister. 1. (1978) 18 S.A.S.R. 238. 2. [1978] Q.B. 95. 3. [1974] 1 N.Z.L.R. 180. 4. (1978) 18 S.A.S.R., at p. 239.

In my view, no assistance in resolving this matter can be gained from the position of the advocate who publishes defamatory statements in court. His immunity in that regard was accepted as virtually axiomatic in Munster v. Lamb [86] . As it happened, the defendant in that case was a solicitor advocate who was sued in defamation for words uttered by him during the course of a hearing in a court of petty sessions. As appears from the judgments of Brett M.R. and Fry L.J. in the Court of Appeal, the basis of their decision dismissing an appeal from a dismissal of the action against the solicitor was not related to the immunity of an advocate from an action in negligence. Rather, the question was, as formulated by Brett M.R. [87] : "whether an action of any kind will lie for defamation uttered in the course of a judicial proceeding". It is a question which may be directed to any of the participants in the judicial process whether judge, juryman, witness or advocate. The answer in each case is the same and, as was said by the Court of Exchequer Chamber in Dawkins v. Lord Rokeby [88] , it is the "occasion" on which the statements are made that gives their author protection from suit. It follows that Munster v. Lamb is no authority, express or implied, for any assertion that a solicitor advocate may not be sued in negligence.

1. (1883) 11 Q.B.D. 588. 2. (1883) 11 Q.B.D., at p. 602. 3. (1873) 8 Q.B. 255, at p. 268.

In Demarco v. Ungaro [89] , Krever J. rejected for Ontario any notion of an immunity, based on public policy, for a lawyer in relation to his conduct, as an advocate, of a case in court. In doing so Krever J. said [90] :

The last consideration to be dealt with is the perceived anomaly related to the absolute privilege enjoyed in respect of anything said in Court by a lawyer. I confess that I am unable to appreciate why it should follow from the existence of that privilege that a lawyer may not be sued by his or her client for the negligent performance of the conduct of the client's case in Court. The privilege, a fundamental aspect of the law of slander, is not concerned with relationships among persons. It relates to legal proceedings in open Court. The special relationship of lawyer and client is not involved as it is, of course, when one is considering the law of negligence. 1. (1979) 95 D.L.R. (3d) 385. 2. (1979) 95 D.L.R. (3d), at pp. 407-408.

In Saif Ali v. Sydney Mitchell & Co. [91] , both Lord Wilberforce [92] and Lord Russell of Killowen [93] rejected the principle of immunity for statements in court as a basis for the immunity enjoyed by a barrister in the conduct of litigation. It must be said, however, that others of their Lordships have placed some reliance upon the principle in Munster v. Lamb to support the conclusion that the public interest supported the immunity of barristers from suit in the conduct of litigation: see Rondel v. Worsley, per Lord Reid [94] ; Lord Pearce [95] ; Lord Upjohn [96] ; Saif Ali, per Lord Diplock [97] ; Lord Salmon [98] .

1. [1980] A.C. 198. 2. [1980] A.C., at p. 214. 3. [1980] A.C., at p. 233. 4. [1969] 1 A.C., at p. 230. 5. [1969] 1 A.C., at pp. 266-267. 6. [1969] 1 A.C., at p. 284. 7. [1980] A.C., at p. 222. 8. [1980] A.C., at p. 230.

When s. 10(2) of the 1958 Act equates the liability for negligence of a barrister "as a barrister" with that of a solicitor "as a solicitor", there is no good reason for fastening on to the liability of solicitors for "in-court negligence" as it existed on 23 November 1891. That is not the equation expressed in the sub-section. The equation is with the liability of a solicitor in the performance of his duties as a solicitor. As Marks J. commented in the course of his judgment:

The words which equate the extent of liability for negligence to that of solicitors was something of a political statement to emphasize that the two kinds of practitioners were to be on the same footing: see Lowe J. in Re Long [99] . That liability sprang in 1891, as it springs today, from a duty to the client to exercise care in carrying out the functions of a solicitor. I agree with counsel for the appellants that it would be a curious result if the Act empowered a barrister to sue for his fees but at the same time precluded the client from raising any defence, set-off or cross-claim based upon the negligence of the barrister in and about the work for which fees had been rendered and were sought to be recovered. In that respect it should be noted that s. 12 of the present statute (s. 8 of the 1891 Act) provides that any barrister or solicitor who does not give substantial attendance to a matter in court shall be deprived of or shall return to the client the whole or such portion of the fee received or agreed to be received as the taxing master may determine. The question whether the barrister and solicitor have given substantial attendance is decided by the taxing master, subject to an appeal to a judge in chambers. But any such decision shall not affect any liability which such barrister and solicitor "has incurred" for negligence: s. 12(2); "may have incurred" in s. 8(2) of the 1891 Act. The reference to liability for negligence that has been or may have been incurred reads somewhat strangely unless the section contemplates that an action for negligence lies in regard to the conduct of the matter in court. 1. [1929] V.L.R. 318, at pp. 319-320.

Mr. Charles, senior counsel for the respondents, challenged the correctness of the appellants' construction of s. 10(2) as a fixed time statute. If the effect of s. 10(2) was to fix the liability of a barrister by reference to the liability of a solicitor on 23 November 1891 then it followed, Mr. Charles argued, that a barrister's liability (unlike that of a solicitor) would not be subject to changes in the law of negligence. This would mean that the liability of a solicitor could become enlarged while that of a barrister would be ever fixed as at 1891. Such a result, it was said, could never have been intended in legislation that was designed to place barristers and solicitors on the same footing.

This argument may have an immediate attraction; but the appellants' construction of s. 10(2) does not have the effect suggested by Mr. Charles. It is true that the Legal Profession Practice Act 1958 is a fixed time Act; see Bennion, Statutory Interpretation (1984), p. 368. And it is true that a barrister's liability is to be determined by reference to the liability of a solicitor in the performance of his duties as a solicitor in 1891. But this means no more than that a Victorian barrister today owes a duty to the client to exercise care in carrying out his functions as a barrister, just as in 1891 a solicitor owed a duty to his client to exercise care in carrying out his functions as a solicitor. The Act thus merely posits a duty of care owed by a barrister to his client. It does not follow from this construction that the duty is not capable of development or that the categories of negligence for which a barrister is liable are forever fixed as at 23 November 1891.

But it does follow that s. 10 of the 1958 statute imposed on a barrister, in the conduct of his practice as a barrister, a liability in negligence. In terms of the questions asked of the Court, it removed the immunity that hitherto existed. That conclusion makes it unnecessary to examine the broader questions of public policy that were argued before us in support of the barrister's immunity from suit in the conduct of litigation. It is however advisable to say two things about those questions, particularly as they were extensively canvassed in Rondel v. Worsley. Assuming the correctness of that decision and assuming further its application to the common law as it existed in 1891, the immunity it upholds is that of a barrister in the performance of his duties in court; it is not that of a solicitor in the performance of his duties as a solicitor. The equation in s. 10(2), as I have said earlier, is with the liability of the solicitor "as a solicitor". The second thing is this. In a non-statutory context matters of public policy to which Rondel v. Worsley refers have some cogency, not the least being the obvious problems arising from the relitigation of issues already decided. Even in the statutory context, as I have construed it, some of these matters have a bearing on the barrister's duty of care. The advocate's obligations to the court, for instance, are relevant to the duty of care he exercises though this is not to suggest that there is anything irreconcilable between the duty owed to the client and that owed to the court. In this connexion, see the illuminating article by Smith, Liability for the Negligent Conduct of Litigation: The Legacy of Rondel v. Worsley, Saskatchewan Law Review, vol. 47 (1982) 211, esp. at pp. 228-246 .

The questions to be answered by this Court fall to be answered by reference to the statute law of Victoria. Each of the questions should be answered "No". It follows that the appeals should be allowed.