ter Neuzen v. Korn, [1995] 3 S.C.R. 674

ter Neuzen v. Korn, [1995] 3 S.C.R. 674

Kobe ter Neuzen Appellant

v.

Dr. Gerald Korn Respondent

and

HIV‑T Group (Blood Transfused), the Canadian

Association of Transfused Hepatitis C Survivors,

the Canadian Hospital Association and

the Canadian Red Cross Society Interveners

Indexed as: ter Neuzen v. Korn

File No.: 23773.

1995: February 2; 1995: October 19.

Present: La Forest, L'Heureux‑Dubé, Sopinka, Gonthier, Cory, McLachlin and Iacobucci JJ.

on appeal from the court of appeal for british columbia

Torts ‑‑ Negligence ‑‑ Physicians and surgeons ‑‑ Patient undergoing artificial insemination procedure and contracting HIV through infected semen of donor ‑‑ Whether physician could be found negligent notwithstanding conformity with standard medical practice ‑‑ Whether trial judge erred in instructing jury that prevailing standard of practice could itself be found to be negligent.

Contracts ‑‑ Sale of goods ‑‑ Implied conditions ‑‑ Implied warranty ‑‑ Patient undergoing artificial insemination procedure and contracting HIV through infected semen of donor ‑‑ Whether implied conditions under Sale of Goods Act applicable ‑‑ Whether implied warranty exists at common law that semen would be of merchantable quality and fit for its purpose ‑‑ Sale of Goods Act, R.S.B.C. 1979, c. 370.

Damages ‑‑ Non‑pecuniary damages ‑‑ Limit ‑‑ Patient undergoing artificial insemination procedure and contracting HIV through infected semen of donor ‑‑ Whether trial judge erred in failing to charge jury on rough upper limit on non‑pecuniary damages ‑‑ If not, whether damages awarded for non‑pecuniary loss ought to be adjusted accordingly.

The appellant participated in the respondent obstetrician and gynaecologist's artificial insemination ("AI") program from 1981 until January 1985, and became infected by HIV as a result of the final AI procedure she underwent. The respondent had not warned her of the risk of HIV infection. The first documented case in the world of HIV transmission through AI was published in the lay media in July 1985 and in a medical journal in September 1985. None of the obstetric literature mentioned AI as a mode of transmission of HIV and no article summarized the disease risks of AI before 1986. As of January 1985, there was no test available for the detection of HIV in semen or blood in Canada. Although the respondent knew that HIV could be transmitted by heterosexual sex, he was not aware that HIV could be transmitted by AI until July 1985. There was expert evidence at trial that the respondent's AI practice was in keeping with general practices across Canada. Specifically, his practice of recruitment and screening of donors and semen accorded with standard practice. The trial judge instructed the jury that it was open to find the respondent negligent on the basis that he failed to comply with the standard medical practice at that time. Alternatively, they could find that the approved practice itself was negligent. With respect to the sale of goods issue, the trial judge instructed the jury to determine first whether the contract between the parties was primarily one for goods or primarily for services. If it was primarily for goods, then the Sale of Goods Act with its implied conditions applied. The trial judge also instructed the jury that aside from the Act, they had to consider whether there was a warranty of fitness and merchantability at common law, but that any common law warranty would simply amount to the same thing as negligence. The jury held that the contract between the parties was primarily one for services and therefore the Sale of Goods Act did not apply. They found the respondent negligent, however, and awarded damages totalling $883,800, including $460,000 for non‑pecuniary damages. The Court of Appeal set aside the verdict and ordered a new trial on the issue of liability as well as damages. With respect to the negligence claim, the court distinguished between two aspects of the respondent's practice: the conduct of the AI procedure, which reflected the current state of knowledge as to the risk of transmission of HIV involved in the use of that procedure, and the screening and follow‑up of donors. The court concluded that it was impossible to determine whether the jury found the respondent negligent on the first or second aspects of his practice. On the evidence, it was not available to the jury to find that the respondent ought to have known of the risk of HIV by AI. The trial judge charged the jury in this manner, however, and it may have been the basis upon which the jury found negligence. Therefore, the verdict could not stand. The court also concluded that while it was reasonable for the trial judge not to have instructed the jury on the rough upper limit for non‑pecuniary damages when neither counsel had asked for such a charge, the award must be corrected to conform with the jurisprudence if the damages exceed the limit.

Held: The appeal should be dismissed.

Per La Forest, Sopinka, Gonthier, Cory, McLachlin and Iacobucci JJ.: It is well settled that physicians have a duty to conduct their practice in accordance with the conduct of a prudent and diligent doctor in the same circumstances. In the case of a specialist, such as a gynaecologist and obstetrician, the doctor's behaviour must be assessed in light of the conduct of other ordinary specialists, who possess a reasonable level of knowledge, competence and skill expected of professionals in Canada, in that field. The conduct of physicians must be judged in the light of the knowledge that ought to have been reasonably possessed at the time of the alleged act of negligence. It was not possible for a jury acting judicially to have found that, in 1985, the respondent ought to have known of the risk of HIV by AI. While conformity with common practice will generally exonerate physicians of any complaint of negligence, there are certain situations where the standard practice itself may be found to be negligent. However, this will only be where the standard practice is fraught with obvious risks such that anyone is capable of finding it negligent, without the necessity of judging matters requiring diagnostic or clinical expertise. The question as to whether the trier of fact can find that a standard practice is itself negligent is a question of law to be determined by the trial judge irrespective of the mode of trial. The first aspect of the negligence claim was not one that the jury could decide without the aid of expert evidence. Accordingly, the trial judge erred in inviting them to do so.

With respect to the second aspect of the negligence claim, infection with HIV is within the same class of injury as other sexually transmitted diseases and the respondent could be liable for the damage caused notwithstanding that he did not foresee that a failure to take reasonable steps to protect his patients could result in HIV infection. It would be sufficient to found liability if it is established that the respondent ought to have foreseen the class of injury. The evidence as to standard practice with respect to the screening and follow‑up of donors was sketchy and it would be open to the jury to find that there was at the relevant time no standard practice. On the other hand, the jury could also find that the standard practice was not to screen donors or carry out follow‑up interviews beyond what the respondent did. The jury could determine the appropriate standard without reliance on expert evidence.

The Information Sheet provided by the respondent did not amount to a warranty that the donor would not be homosexual or a drug abuser. There is no indication that there was any intention on the part of either party that there should be contractual liability in respect of the statement in the Information Sheet. Moreover, it appears that the appellant did not raise this issue in either court below.

In order for the Sale of Goods Act to apply, a contract must be primarily for the purpose of selling goods. If the sale of a good is merely incidental to what is primarily a contract for services, then the statute will not imply a warranty. Assuming that there was in fact a sale of semen between the appellant and respondent, it cannot be contended that the contract was one primarily for the sale of semen such as to attract the application of the Sale of Goods Act. Apart from the Act, a court must consider whether a common law warranty of fitness and merchantability should be implied into the contract which includes services as well as the provision of materials. However, such a warranty will not be implied in all circumstances. The court must examine the specific nature of the contract and the relationship between the parties in order to assess whether it was the intention of the parties that such a warranty be implied. Courts must be very cautious in their approach to implying contractual terms. A rationale for implying warranties in contracts of goods and services is that a supplier of goods generally has recourse against the manufacturer under the Sale of Goods Act as a result of the statutory conditions imposed. While it is true that the primary purpose of the implied warranty is to hold the supplier of goods liable notwithstanding the absence of negligence, different considerations apply in the context of the medical profession than in the ordinary commercial context. The doctor cannot trace the liability back to the initial manufacturer. Moreover, it must be recognized that biological products such as blood and semen, unlike manufactured products, carry certain inherent risks. It would be inappropriate to imply a warranty of fitness and merchantability in the circumstances of this case. Moreover, any warranty would simply be to take reasonable care.

While the appellant has suffered immensely as a result of this tragedy, and AIDS is a dreadful disease which will eventually take her life prematurely, with respect to non‑pecuniary losses this case is no different from other tragedies and the adjusted rough upper limit on non‑pecuniary damages should accordingly apply. The trial judge should instruct the jury as to an upper limit if, after considering the submissions of counsel, he or she is of the opinion that the damages by reason of the type of injury sustained might very well be assessed in the range of or exceeding the upper limit. On the other hand, if the trial judge is of the view that the injuries involved will not likely produce an award approaching the rough upper limit, it is best that the trial judge not charge the jury on the matter. Whether the jury is or is not advised of the upper limit, if the award exceeds the limit the trial judge should reduce it to conform with the "cap" established, adjusted for inflation. In the present case it was reasonable for the trial judge not to give an instruction on the upper limit, but since the damages awarded for non‑pecuniary losses far exceeded that limit, he ought to have reassessed the award.

Per L'Heureux‑Dubé J.: Sopinka J.'s reasons were agreed with, except with respect to the trial judge's instruction to the jury on the upper limit of non‑pecuniary damages. The determination of the quantum of damages should be left to the jury without instructions as to the upper limit, subject to the judge's power to reduce excessive awards. It has always been held improper for the trial judge or counsel to express any views as to the quantum of damages. It is well established that the amount of damages is a question of fact for the jury to determine, and not a question of law for the judge. The establishment of an upper limit for non‑pecuniary damages only affects the quantum of the awards, not the way in which they are determined. The existence of an upper limit as a matter of law remains consistent with not instructing the jury as to such upper limit. By so instructing the jury, the status quo in novel areas of the law would be institutionalized. There is no evidence as to the inadequacy of the present system. Finally, from a pragmatic standpoint, to require the trial judge to charge the jury on the upper limit of non‑pecuniary damages may be inviting unmeritorious appeals on that part of the judge's instructions to the jury, given that, when it comes to damages, no figure will ever be "right".

Cases Cited

By Sopinka J.

Followed: Perlmutter v. Beth David Hospital, 123 N.E.2d 792 (1954); Fisher v. Sibley Memorial Hospital, 403 A.2d 1130 (1979); St. Luke's Hospital v. Schmaltz, 534 P.2d 781 (1975); not followed: Cunningham v. MacNeal Memorial Hospital, 266 N.E.2d 897 (1970); referred to: Andrews v. Grand & Toy Alberta Ltd., 1978 CanLII 1 (SCC), [1978] 2 S.C.R. 229; Lindal v. Lindal, 1981 CanLII 35 (SCC), [1981] 2 S.C.R. 629; Wilson v. Swanson, 1956 CanLII 1 (SCC), [1956] S.C.R. 804; Lapointe v. Hôpital Le Gardeur, 1992 CanLII 119 (SCC), [1992] 1 S.C.R. 351; McCormick v. Marcotte, 1971 CanLII 52 (SCC), [1972] S.C.R. 18; Roe v. Ministry of Health, [1954] 2 All E.R. 131; Vancouver‑Fraser Park District v. Olmstead, 1974 CanLII 196 (SCC), [1975] 2 S.C.R. 831; Roberge v. Bolduc, 1991 CanLII 83 (SCC), [1991] 1 S.C.R. 374; Waldick v. Malcolm, 1991 CanLII 71 (SCC), [1991] 2 S.C.R. 456; Anderson v. Chasney, 1949 CanLII 236 (MB CA), [1949] 4 D.L.R. 71 (Man. C.A.), aff'd 1950 CanLII 336 (SCC), [1950] 4 D.L.R. 223 (S.C.C.); R. v. Côté, 1974 CanLII 31 (SCC), [1976] 1 S.C.R. 595; Gee v. White Spot Ltd. (1986), 1986 CanLII 776 (BC SC), 7 B.C.L.R. (2d) 235; G. H. Myers and Co. v. Brent Cross Service Co., [1934] 1 K.B. 46; Young & Marten Ltd. v. McManus Childs Ltd., [1969] 1 A.C. 454; G. Ford Homes Ltd. v. Draft Masonry (York) Co. (1983), 1983 CanLII 1719 (ON CA), 43 O.R. (2d) 401; Arnold v. Teno, 1978 CanLII 2 (SCC), [1978] 2 S.C.R. 287; Crosby v. O'Reilly, 1974 CanLII 173 (SCC), [1975] 2 S.C.R. 381.

By L'Heureux‑Dubé J.

Referred to: Gray v. Alanco Developments Ltd. (1967), 1967 CanLII 221 (ON CA), 61 D.L.R. (2d) 652; Hill v. Church of Scientology of Toronto, 1995 CanLII 59 (SCC), [1995] 2 S.C.R. 1130; Andrews v. Grand & Toy Alberta Ltd., 1978 CanLII 1 (SCC), [1978] 2 S.C.R. 229; Thornton v. Board of School Trustees of School District No. 57 (Prince George), 1978 CanLII 12 (SCC), [1978] 2 S.C.R. 267; Arnold v. Teno, 1978 CanLII 2 (SCC), [1978] 2 S.C.R. 287.

Statutes and Regulations Cited

Negligence Act, R.S.B.C. 1979, c. 298, s. 6.

Sale of Goods Act, R.S.B.C. 1979, c. 370, ss. 17, 18.

Authors Cited

British Columbia. Law Reform Commission. Report on Compensation for Non‑Pecuniary Loss. Vancouver: The Commission, 1984.

Charles, W. H. R. Charles Handbook on Assessment of Damages in Personal Injury Cases, 2nd ed. Toronto: Carswell, 1990.

Fleming, John G. The Law of Torts, 7th ed. Sydney: Law Book Co., 1987.

Fridman, G. H. L. Sale of Goods in Canada, 2nd ed. Toronto: Carswell, 1979.

APPEAL from a judgment of the British Columbia Court of Appeal (1993), 1993 CanLII 2881 (BC CA), 81 B.C.L.R. (2d) 39, 103 D.L.R. (4th) 473, [1993] 6 W.W.R. 647, 29 B.C.A.C. 1, 48 W.A.C. 1, 16 C.C.L.T. (2d) 65, setting aside the jury's verdict, which found the respondent negligent and awarded damages to the appellant, and ordering a new trial. Appeal dismissed.

Sandra J. Harper and Kathleen Birney, for the appellant.

Christopher E. Hinkson, Q.C., and M. Lynn McBride, for the respondent.

Written submissions only by Kenneth Arenson, for the interveners HIV‑T Group (Blood Transfused) and the Canadian Association of Transfused Hepatitis C Survivors.

Written submissions only by Daniel A. Webster, Q.C., for the intervener the Canadian Hospital Association.

Written submissions only by Peter K. Boeckle and Anil Varma, for the intervener the Canadian Red Cross Society.

//Sopinka J.//

The judgment of La Forest, Sopinka, Gonthier, Cory, McLachlin and Iacobucci JJ. was delivered by

1 Sopinka J. -- This appeal raises issues concerning the liability of the respondent physician for conducting an artificial insemination ("AI") procedure which resulted in his patient, the appellant, contracting the Human Immunodeficiency Virus ("HIV") through the infected semen of the donor. Specifically, this Court must address whether the respondent physician could be found negligent, notwithstanding conformity with standard medical practice, and whether the trial judge erred in instructing the jury that the prevailing standard of practice could itself be found to be negligent.

2 As well, this appeal also raises for consideration the issue of whether the implied conditions under the Sale of Goods Act, R.S.B.C. 1979, c. 370, are applicable in the present circumstances or whether there exists an implied warranty at common law that the semen would be of merchantable quality and fit for its purpose.

3 Finally, this appeal requires the Court to consider whether the rough upper limit for non-pecuniary damages, established by this Court in Andrews v. Grand & Toy Alberta Ltd., 1978 CanLII 1 (SCC), [1978] 2 S.C.R. 229, ought to have been applied by the jury in the present case. First, it must be decided whether the trial judge erred in failing to charge the jury on the rough upper limit. If the trial judge was not in error, this Court must decide whether the damages awarded for non-pecuniary loss, which far exceeded the upper limit established in Andrews, ought nonetheless to be adjusted in accordance with those principles.

4 I should note that the interveners HIV-T Group (Blood Transfused) and Canadian Association of Transfused Hepatitis C Survivors also raised the issue of whether the medical profession ought to be held liable for the distribution of defective products, such as semen, based on a theory of strict liability in tort. This approach has been used in some American cases. However, I am in agreement with the respondent that this is not the appropriate case to address this important issue which could have far-reaching implications for the medical profession and the Canadian system of public health insurance in general. The appellant did not raise this issue in her pleadings at trial, nor at the Court of Appeal. Thus, the issue has been raised for the first time at this Court by the interveners. In these circumstances, the applicability of strict liability in tort is best left for another day.

I. Facts

5 The facts and the evidence of the expert witnesses at trial were thoroughly summarized in the reasons of the Court of Appeal. I do not intend to review the evidence of each expert individually again in these reasons. However, in order to resolve this appeal, it is necessary to review the relevant facts in some detail and highlight some of the key aspects which emerge from the evidence.

6 The respondent is an obstetrician and gynaecologist and has practised AI since 1974. The appellant was a psychiatric nurse. She participated in the respondent's AI program from 1981 until January 21, 1985, during which time she underwent about 35 AI procedures. It was agreed that the appellant became infected by HIV as a result of the AI procedure on January 21, 1985. The respondent did not warn the appellant of the risk of HIV infection resulting from the AI procedure (also referred to as "the risk").

7 The first report of HIV in female sex partners of men with AIDS appeared in early 1983. At that time, heterosexual intercourse was seen as a potential source of infection. However, there was no link between AI and HIV. In December 1983, for the first time, the risk of contracting AIDS was related to blood transfusions. In October 1983, an important letter was published by Dr. Mascola in the New England Journal of Medicine suggesting that there was a risk of transmitting sexually transmitted diseases ("STDs") through AI. However, this prestigious journal was not widely read by gynaecologists and the respondent did not read this letter. It appears that Dr. Mascola was the first person in the world to express a concern about the possibility of transmitting HIV through AI. The first documented case in the world of HIV transmission through AI was published in the lay media in July 1985 and in a medical journal in September 1985. None of the obstetric literature mentioned AI as a mode of transmission of HIV and no article summarized the disease risks of AI before 1986.

8 The evidence reveals that the knowledge about HIV was rapidly growing in the period of 1984-85; however, it was still quite limited and confusing in the early days. Generally, research scientists and health care officials in the AIDS field were expected to be more up to date on medical developments concerning HIV than clinical practitioners and the general medical community. In 1984, the Elisa test was developed in the United States which enabled doctors to test for HIV in blood and tissues. However, as of January 21, 1985, there was no test available for the detection of HIV in semen or blood in Canada. The Elisa test did not become available in Canada until later in 1985. Therefore, it was impossible to test the semen of donors for HIV at the time the appellant became infected.

9 At this time, it was widely believed that HIV was an STD; however, it was hoped that the atraumatic AI procedure was free of risk. It was believed that the small abrasions caused by intercourse were necessary in order for the virus to get into the bloodstream. Thus, in late 1984, a possibility of the transmission of HIV through heterosexual intercourse was known, but AI was not seriously considered to put anyone at risk of being infected.

10 Some of the experts testified that an analogy could be drawn between Hepatitis B and HIV, since both were STDs. It was not until late 1986 that it was recognized that Hepatitis B could be transmitted by AI and it was common ground that there were no published reports of Hepatitis B transmissions by AI until the fall of 1987.

11 In mid-1984, four babies in Australia were found to have acquired AIDS by blood transfusions. Since it was known that the Elisa test would eventually be available for clinical use, this led to a ministerial decree to impose a moratorium on all body fluid and tissue transfers. As a result, all AI clinics were closed at this time, although this was not supported by the whole medical profession in Australia. This event was not widely publicized in North America and the respondent, as well as the general medical community in North America, were unaware of the closures in Australia until later in 1985. Apparently, there was little exchange of medical information on the subject of HIV between Australia and North America. In September 1985, an article in the British medical journal Lancet revealed that four Australian women had contracted HIV from AI. When the respondent learned of this he immediately discontinued his AI program and recommended that his donors and the appellant be tested.

12 The American Fertility Society published guidelines for the use of semen donor insemination for the first time in 1986 and these guidelines were revised in 1988. The new guidelines recommended the use of frozen semen only, which is stored for at least six months. Blood samples for HIV antibodies are taken at the time of semen donation and six months later. The semen is only to be used if both samples are HIV negative. These guidelines were not published by the American Center for Disease Control until February 1988. The respondent did not start testing for Hepatitis B until sometime in 1987-88, when the guidelines were issued.

13 Although the respondent knew that HIV could be transmitted by heterosexual sex, as well as through blood transfusions, he was not aware that HIV could be transmitted by AI until July 1985, when he heard about the experience in Australia. None of the medical publications he read prior to January 1985 nor any of the medical meetings he attended suggested that there was a risk of contracting HIV through AI. The respondent did not read Dr. Mascola's letter which appeared in the New England Journal of Medicine. He did not make any connection between the fact HIV was an STD and the possibility that it could be passed on through AI since there was nothing recorded to this effect.

14 There was expert evidence that the respondent's AI practice was in keeping with general practices across Canada. Specifically, his AI practice of recruitment and screening of donors and semen accorded with standard practice across Canada. The respondent used only fresh semen prior to January 1985. He personally interviewed all his donors instead of using a questionnaire. In the interview, the respondent obtained a complete medical history of the prospective donor including whether the donor was heterosexual or homosexual. The respondent rejected homosexuals because he considered that they had practices that could potentially transmit certain diseases to recipients. The respondent also inquired into the levels of sexual activity since active donors were more likely to be exposed to STDs.

15 The respondent also conducted a general physical examination. He tested for gonorrhoea, syphilis and other bacteria and conducted blood and semen tests for Rh blood factor. Other than on a random basis, there was no repeat testing of semen nor re-interviewing of donors. Of the 28 active donors used by the respondent in his AI practice, only the donor whose semen was administered to the appellant was infected by HIV. That donor had been asked whether he was heterosexual or homosexual and he replied that he was heterosexual. After the donor tested HIV positive, he told the respondent that he was bisexual.

16 At trial, the trial judge instructed the jury that it was open to find the respondent negligent on the basis that he failed to comply with the standard medical practice at that time. Alternatively, the trial judge stated that the jury could find that the approved practice itself was negligent.

17 With respect to the sale of goods issue, the trial judge instructed the jury to first determine whether the contract between the parties was primarily one for goods or primarily for services. If it was primarily for goods, then the Sale of Goods Act applied. However, the jury held that the contract was primarily one for services and therefore the Act did not apply. The trial judge also instructed the jury that aside from the Act, they had to consider whether there was a warranty at common law. However, in this context, the trial judge stated that any common law warranty would simply amount to the same thing as negligence. In other words, under the common law warranty in a contract for medical services, the respondent undertakes to meet the standards of a reasonably competent person practising in his field. The doctor does not warrant that the services will be effective.

18 In the result, the jury found the respondent negligent. However, it is unknown on what basis the jurors reached their decision. The damages awarded by the jury totalled $883,800, including $460,000 for non-pecuniary damages. On appeal, however, the verdict of the jury was set aside and the Court of Appeal ordered a new trial on the issue of liability as well as damages: (1993), 1993 CanLII 2881 (BC CA), 81 B.C.L.R. (2d) 39, 103 D.L.R. (4th) 473, [1993] 6 W.W.R. 647, 29 B.C.A.C. 1, 48 W.A.C. 1, 16 C.C.L.T. (2d) 65.

II. Relevant Statutory Provisions

Sale of Goods Act, R.S.B.C. 1979, c. 370:

17. Where there is a contract for the sale of goods by description, there is an implied condition that the goods shall correspond with the description. If the sale be by sample, as well as by description, it is not sufficient that the bulk of the goods correspond with the sample if the goods do not also correspond with the description.

18. Subject to this Act and any statute in that behalf, there is no implied warranty or condition as to the quality or fitness for any particular purpose of goods supplied under a contract of sale, except as follows:

(a)where the buyer, expressly or by implication, makes known to the seller the particular purpose for which the goods are required, so as to show that the buyer relies on the seller's skill or judgment, and the goods are of a description which it is in the course of the seller's business to supply, whether he is the manufacturer or not, there is an implied condition that the goods are reasonably fit for such purpose; except that in the case of a contract for the sale of a specified article under its patent or other trade name, there is no implied condition as to its fitness for any particular purpose;

(b)where goods are bought by description from a seller who deals in goods of that description, whether he is the manufacturer or not, there is an implied condition that the goods shall be of merchantable quality; but if the buyer has examined the goods there is no implied condition as regards defects which such examination ought to have revealed;

(c)an implied warranty or condition as to quality or fitness for a particular purpose may be annexed by the usage of trade; and

(d)an express warranty or condition does not negative a warranty or condition implied by this Act unless inconsistent with it.

III. Judgment of Court of Appeal (1993), 1993 CanLII 2881 (BC CA), 81 B.C.L.R. (2d) 39

19 After a review of the evidence, including an exhaustive review of the evidence of standard practice, the court noted that the law was well established that "physicians are required to conduct themselves at least in accordance with the standard of their professional peers, but they are not expected to guarantee the success of their procedures or the perfect safety of their patients" (p. 61).

20 With respect to the negligence claim, the Court of Appeal distinguished between two aspects of the defendant's practice:

(1)the conduct of the AI procedure which reflected the current state of knowledge as to the risk of transmission of HIV involved in the use of that procedure;

(2) the screening and follow-up of donors.

21 With respect to the first aspect of the case, the court stated (at p. 67):

Thus, it is apparent that for the Defendant to be found liable on this part of the case, it would be necessary for the jury to be satisfied that the Defendant failed to have the knowledge and understanding expected in accordance with the standards of his profession. That, of course, was a question for the jury, and it was essential that the jury be carefully instructed upon the standard they should apply in deciding whether the Defendant was guilty of negligence. [Emphasis added.]

22 The court doubted that it could reasonably be found that the defendant was negligent in using fresh semen. It found, however, that the instructions to the jury were deficient in that the trial judge instructed the jury that it was open for them to find the general practice itself was negligent. In this regard, the court stated (at p. 73):

The only proper instruction to be given on at least a part of the case was that the jury should decide whether the Defendant conducted himself as a reasonable physician would in similar circumstances. In our judgment, that required the jury to confine itself to prevailing standards of practice.

The Court of Appeal also stated (at p. 74):

It was not possible on the evidence for the jury to find that it was negligent of the Defendant not to have known of the risk of HIV by AI and that he should have ceased performing AI. That was the principal argument of the Plaintiff and it may be the ground upon which the jury based its finding of negligence. If that is the case the verdict cannot stand.

23 With respect to the second aspect of the defendant's practice relating to the screening and follow-up of donors, the Court of Appeal stated (at p. 69):

What is more significant is that the Defendant made specific enquiries about homosexuality, not out of concern about HIV or AIDS, but because he regarded such persons as higher risks for the more common STDs. As a result of his initial enquiry, he felt confident in assuring the Plaintiff that her donor was not a homosexual person. A follow-up interview, or closer questioning, may or may not have disclosed that the donor was bisexual.

Keeping in mind that infection in a donor could be a "day-to-day" thing as described by Dr. Mascola, and that lifestyle changes can occur quickly, it was necessary for the jury to consider whether the Defendant's screening procedures were deficient. Again, of course, it was essential that the jury be given a proper instruction on the principles governing such matters. [Emphasis added.]

24 With respect to the instructions on this aspect, the Court of Appeal stated (at p. 73):

On the other part of the case, relating to the screening of donors, and follow-up interviews, the case is not so clear because less scientific questions are involved. In addition, we have identified some overlap between the two parts of the case. Screening homosexual persons out of an AI program because of their believed greater risk of infection was recognized by the Defendant as a desirable aim. The elimination of this particular donor for any reason would have protected the Plaintiff from HIV. It must be remembered, however, that it was impossible to guarantee safety because an inaccurate answer or a changed lifestyle after one or more follow-up interviews could have left the Plaintiff exposed to infection. There is no evidence suggesting that donors should be interviewed by the responsible physician before every donation.

On the other hand, if the jury found that the Defendant breached this duty of care to the Plaintiff then she may have an actionable claim against the Defendant because a defendant cannot always escape liability even though he does not foresee all the specific harm his patient may suffer from such a breach.

25 The court concluded that it was impossible to determine whether the jury found the respondent negligent on the first or second aspects of the defendant's practice. On the evidence, it was not available to the jury to find that the respondent ought to have known of the risk of HIV by AI. However, the trial judge charged the jury in this manner and it may have been the basis upon which the jury found negligence. Therefore, the verdict could not stand. A new trial was ordered on the issues other than those based on the first aspect of the case.

26 With respect to the sale of goods issue, the court assumed without deciding that there was a sale of semen. The trial judge instructed the jury to decide whether the contract between the parties was primarily for the sale of a good or whether it was a contract for services. Since the jury decided that it was not a sale of goods, the Sale of Goods Act did not apply. The court held that there was evidence upon which a jury could conclude that the arrangement was primarily one for professional services.

27 However, that did not resolve the issue of whether there should be an implied warranty at common law for the provision of medical services which included a good. The jury did not decide this issue because the trial judge instructed the jury that the warranty essentially amounted to the same thing as negligence. The court did not wish to go as far as to say that a physician would only be liable in warranty at common law if there was negligence in failing to detect or remove a defect in the good. Instead, the court was of the view that this was not a proper case for the implication of a warranty. In its opinion, it is a question of fact in each case whether the circumstances are sufficient to exclude common law warranties. It therefore dismissed this claim.

28 The court next considered the respondent's argument that the award of $460,000 for non-pecuniary damages, in excess of the rough upper limit prescribed by the Supreme Court of Canada, was inordinately high. It held that it was reasonable for the trial judge not to have instructed the jury on a rough upper limit when neither counsel had asked for such a charge. However, this did not mean that the appellant's damages should not be properly assessed in light of the jurisprudence. Relying on Lindal v. Lindal, 1981 CanLII 35 (SCC), [1981] 2 S.C.R. 629, the court concluded (at p. 87):

While recognizing that the [appellant's] injuries are markedly different in kind from those under consideration in the trilogy, we cannot conclude that the principle adopted by the Supreme Court of Canada, aimed at controlling the social burden of damage awards, should not be applied in this case. It follows, in our view, that the [appellant's] non-pecuniary damages cannot exceed the adjusted rough upper limit.

29 It was held that, although the trial judge did not err in his instruction to the jury, if the damages exceed the rough upper limit, the Court of Appeal must correct the award in conformity with the governing principles. Since there was to be a new trial on liability, the court was of the view that there should also be a new trial ordered on the question of damages generally.

IV. Issues

1.Did the Court of Appeal err in holding that the trial judge ought not to have charged the jury that, even if the respondent conformed to standard medical practice, it was open to the jury to find that the prevailing medical practice with respect to the AI procedure was negligent?

2.Did the Court of Appeal err in rejecting the appellant's claim based on a warranty under the Sale of Goods Act?

3.Did the Court of Appeal err in holding that a warranty of quality or fitness should not be implied under the common law, in the circumstances of this case?

4.Did the Court of Appeal err in holding that the rough upper limit on non-pecuniary damages applies in the present case?

5.If the rough upper limit of non-pecuniary damages applies in the present case, must the jury be instructed accordingly or should it be left to the Court of Appeal to adjust a jury award which exceeds the limit?

V. Analysis

A. Professional Negligence

30 I agree with the Court of Appeal that there are two aspects to the claim of professional negligence:

(1)breach of duty arising from the failure to be aware of the risk of HIV infection through the use of AI; and

(2)breach of duty with respect to the screening and follow-up of donors.

31 The alleged departures from the applicable standard of care were:

(1)in respect of the first aspect, failure to discontinue the practice of AI or, in the alternative, to warn the patients of the risk;

(2)in respect of the second aspect, the failure to adequately screen donors so as to eliminate those in a high risk category with relation to the transmission of STDs and to re-interview donors periodically to detect changes in lifestyle. There is also the claim that frozen semen rather than fresh semen should have been used.

32 In order to properly address the issues relating to professional negligence, it is useful to consider what were the matters which the jury was obliged to decide. In each aspect of the claim the jury was bound to consider whether the evidence established that a standard of practice existed. If the answer was in the affirmative, the next question was whether the defendant conformed to that practice. An affirmative answer to this question would result in a finding of no negligence in favour of the respondent unless the jury was entitled to consider and hold that the standard practice was itself below the required legal standard and that conduct below that standard constituted negligence. Similarly, if the jury found that no standard practice was established by the evidence, the appellant would have failed to prove her case unless the jury was entitled to fix the standard without the necessity of expert evidence. Finally, if the respondent failed to comply with the standard of care established by the evidence or by the jury, in the absence of or insufficiency of evidence, then the respondent would be found to be negligent.

(1) Standard of Care and Evidence of Standard Practice

33 It is well settled that physicians have a duty to conduct their practice in accordance with the conduct of a prudent and diligent doctor in the same circumstances. In the case of a specialist, such as a gynaecologist and obstetrician, the doctor's behaviour must be assessed in light of the conduct of other ordinary specialists, who possess a reasonable level of knowledge, competence and skill expected of professionals in Canada, in that field. A specialist, such as the respondent, who holds himself out as possessing a special degree of skill and knowledge, must exercise the degree of skill of an average specialist in his field: see Wilson v. Swanson, 1956 CanLII 1 (SCC), [1956] S.C.R. 804, at p. 817, Lapointe v. Hôpital Le Gardeur, 1992 CanLII 119 (SCC), [1992] 1 S.C.R. 351, at p. 361, and McCormick v. Marcotte, 1971 CanLII 52 (SCC), [1972] S.C.R. 18.

34 It is also particularly important to emphasize, in the context of this case, that the conduct of physicians must be judged in the light of the knowledge that ought to have been reasonably possessed at the time of the alleged act of negligence. As Denning L.J. eloquently stated in Roe v. Ministry of Health, [1954] 2 All E.R. 131 (C.A.), at p. 137, "[w]e must not look at the 1947 accident with 1954 spectacles". That is, courts must not, with the benefit of hindsight, judge too harshly doctors who act in accordance with prevailing standards of professional knowledge. This point was also emphasized by this Court in Lapointe, supra, at pp. 362-63:

... courts should be careful not to rely upon the perfect vision afforded by hindsight. In order to evaluate a particular exercise of judgment fairly, the doctor's limited ability to foresee future events when determining a course of conduct must be borne in mind. Otherwise, the doctor will not be assessed according to the norms of the average doctor of reasonable ability in the same circumstances, but rather will be held accountable for mistakes that are apparent only after the fact.

No issue is taken with this proposition which was applied both in the trial judge's charge to the jury and by the Court of Appeal.

35 The Court of Appeal, after a thorough review of the evidence, held that it was not possible for a jury acting judicially to have found that, in 1985, the respondent ought to have known of the risk. This is a power to review a jury verdict which a court of appeal clearly possesses. See Vancouver-Fraser Park District v. Olmstead, 1974 CanLII 196 (SCC), [1975] 2 S.C.R. 831. I agree with this finding and can find no basis upon which it can be questioned. Indeed my review of the evidence leads to the same conclusion. The evidence of standard practice on the first aspect of the case was based entirely on the state of knowledge required of the reasonable practitioner in 1985 and it would have been equally impossible for a jury acting judicially to have found that, given the state of knowledge, the reasonable practitioner ought to either have discontinued AI or warned the patients of the risk. It having been admitted that the respondent continued AI and did not warn his patients, there was no issue concerning his conformity with the standard practice.

36 The appellant, therefore, can only support a favourable finding on this aspect of the case on the basis that the jury was entitled to find that the standard established by the evidence itself departed from that of a prudent and diligent physician and that the respondent, in failing to conform with a higher standard, was guilty of negligence. This raises the issue as to the correctness of the trial judge's charge to the jury to the effect that the jury was so entitled.

37 With respect to the second aspect of the claim in professional negligence, the Court of Appeal considered that a verdict for the appellant was open to the jury. It is, however, by no means clear that the evidence establishes a standard practice with respect to the screening and follow-up of donors. This was a matter for the jury to determine. If the jury found that the evidence fell short of establishing the existence of a standard practice, the question arises as to whether the jury could determine the applicable standard without the aid of expert evidence. This is a legal issue upon which the Court of Appeal did not pronounce but which is closely related to the issue raised by the trial judge's instruction referred to above.

38 It is generally accepted that when a doctor acts in accordance with a recognized and respectable practice of the profession, he or she will not be found to be negligent. This is because courts do not ordinarily have the expertise to tell professionals that they are not behaving appropriately in their field. In a sense, the medical profession as a whole is assumed to have adopted procedures which are in the best interests of patients and are not inherently negligent. As L'Heureux-Dubé J. stated in Lapointe, in the context of the Quebec Civil Code (at pp. 363-64):

Given the number of available methods of treatment from which medical professionals must at times choose, and the distinction between error and fault, a doctor will not be found liable if the diagnosis and treatment given to a patient correspond to those recognized by medical science at the time, even in the face of competing theories. As expressed more eloquently by André Nadeau in "La responsabilité médicale" (1946), 6 R. du B. 153, at p. 155:

[TRANSLATION] The courts do not have jurisdiction to settle scientific disputes or to choose among divergent opinions of physicians on certain subjects. They may only make a finding of fault where a violation of universally accepted rules of medicine has occurred. The courts should not involve themselves in controversial questions of assessment having to do with diagnosis or the treatment of preference. [Emphasis added.]

39 In The Law of Torts (7th ed. 1987), Professor Fleming observed the following with respect to the role of standard practice, at p. 109:

Conformity with general practice, on the other hand, usually dispels a charge of negligence. It tends to show what others in the same "business" considered sufficient, that the defendant could not have learnt how to avoid the accident by the example of others, that most probably no other practical precautions could have been taken, and that the impact of an adverse judgment (especially in cases involving industry or a profession) will be industry-wide and thus assume the function of a "test case". Finally, it underlines the need for caution against passing too cavalierly upon the conduct and decision of experts.

All the same, even a common practice may itself be condemned as negligent if fraught with obvious risks. [Emphasis added.]

40 With respect to the medical profession in particular, Professor Fleming noted, at p. 110:

Common practice plays its most conspicuous role in medical negligence actions. Conscious at once of the layman's ignorance of medical science and apprehensive of the impact of jury bias on a peculiarly vulnerable profession, courts have resorted to the safeguard of insisting that negligence in diagnosis and treatment (including disclosure of risks) cannot ordinarily be established without the aid of expert testimony or in the teeth of conformity with accepted medical practice. However there is no categorical rule. Thus an accepted practice is open to censure by a jury (nor expert testimony required) at any rate in matters not involving diagnostic or clinical skills, on which an ordinary person may presume to pass judgment sensibly, like omission to inform the patient of risks, failure to remove a sponge, an explosion set-off by an admixture of ether vapour and oxygen or injury to a patient's body outside the area of treatment. [Emphasis added. Footnotes omitted.]

41 It is evident from the foregoing passage that while conformity with common practice will generally exonerate physicians of any complaint of negligence, there are certain situations where the standard practice itself may be found to be negligent. However, this will only be where the standard practice is "fraught with obvious risks" such that anyone is capable of finding it negligent, without the necessity of judging matters requiring diagnostic or clinical expertise.

42 In Roberge v. Bolduc, 1991 CanLII 83 (SCC), [1991] 1 S.C.R. 374, this Court had the opportunity to address this issue in the context of the civil responsibility of a notary under the Quebec Civil Code. In that case, it was recognized that where a custom of a profession ignores the elementary dictates of caution, it is open to a court to find the professional person negligent. Thus, even if a doctor practises in accordance with common professional practice, he will be liable if that practice is wanting. As L'Heureux-Dubé J. stated at pp. 436-37:

This brief overview of both doctrine and jurisprudence indicates that courts have discretion to assess liability despite uncontradicted evidence of common professional practice at the relevant time. The standard, in regard to the particular facts of each case, must still be that of a reasonable professional in such circumstances.

It may very well be that the professional practice reflects prudent and diligent conduct. One would hope that if a certain practice has developed amongst professionals in regard to a particular professional act, such practice is in accordance with a prudent course of action. The fact that a professional has followed the practice of his or her peers may be strong evidence of reasonable and diligent conduct, but it is not determinative. If the practice is not in accordance with the general standards of liability, i.e., that one must act in a reasonable manner, then the professional who adheres to such a practice can be found liable, depending on the facts of each case. [Emphasis in original.]

43 The foregoing principles were also endorsed by this Court in Waldick v. Malcolm, 1991 CanLII 71 (SCC), [1991] 2 S.C.R. 456. Thus, it is apparent that conformity with standard practice in a profession does not necessarily insulate a doctor from negligence where the standard practice itself is negligent. The question that remains is under what circumstances will a professional standard practice be judged negligent? It seems that it is only where the practice does not conform with basic care which is easily understood by the ordinary person who has no particular expertise in the practices of the profession. That is, as Professor Fleming suggests, where the common practice is fraught with danger, a judge or a jury may find that the practice is itself negligent.

44 As was observed in Lapointe, courts should not involve themselves in resolving scientific disputes which require the expertise of the profession. Courts and juries do not have the necessary expertise to assess technical matters relating to the diagnosis or treatment of patients. Where a common and accepted course of conduct is adopted based on the specialized and technical expertise of professionals, it is unsatisfactory for a finder of fact to conclude that such a standard was inherently negligent. On the other hand, matters falling within the ordinary common sense of juries can be judged to be negligent. For example, where there are obvious existing alternatives which any reasonable person would utilize in order to avoid a risk, one could conclude that the failure to adopt such measures is negligent notwithstanding that it is the prevailing practice among practitioners in that area.

45 Such a case arose in Anderson v. Chasney, 1949 CanLII 236 (MB CA), [1949] 4 D.L.R. 71 (Man. C.A.), aff'd 1950 CanLII 336 (SCC), [1950] 4 D.L.R. 223 (S.C.C.). There, a doctor performed surgery on a child's throat. During the course of the surgery, sponges were used without any tape or strings attached to ensure that none was left in the throat. Nor was a nurse present to count the number of sponges used. One of the sponges was inadvertently left in the throat and after the operation, the child died of suffocation. The surgeon gave evidence that it was not his practice to use sponges with strings, nor to have anyone count the sponges used. This appeared to be standard practice in the hospital, notwithstanding that there were sponges available with strings and nurses could be made available for counting sponges. The Court of Appeal held that the surgeon acted negligently.

46 The principal reasons, written by McPherson C.J.M., did not specifically deal with the issue with respect to standard practice. However, McPherson C.J.M. did observe that the fact a sponge was left in a potentially dangerous position is one which "the ordinary man is competent to consider in arriving at a decision as to whether or not there was negligence" (p. 74). By failing to take either of the precautions which were readily available, the majority held that the surgeon was negligent.

47 In separate concurring reasons, Coyne J.A. discussed the issue of customary practice among surgeons. He noted that whether or not it is negligent to omit to use sponges with ties or fail to keep count of the sponges is not a matter requiring expert knowledge. Coyne J.A. observed that general practice is not a complete defence to negligence. I find it useful to quote fairly extensively from his reasons as I believe that they are directly pertinent to the issues in this case. At pages 81-82, he stated:

Dr. Chasney defends on the ground that it has not been his practice, and that it is not the usual practice of operators in the hospital in question and of some operators elsewhere, to take these precautions. But he took a chance in neglecting them. Whether to adopt them or not, he says, is a matter of surgical skill and experience and the general practice of practitioners is conclusive. This, however, was not peculiarly a matter of such skill and experience. His counsel argued that if a general practice of surgeons is followed, negligence cannot be attributed, and that expert evidence is conclusive. But if that were correct the expert witnesses would, in effect, be the jury to try the question of negligence. That question, however, must continue to be one for the petit jury empanelled to try the case, if it is a jury case, and for the Court where it is not. The experts remain witnesses to give their expert opinions in assistance of the jury or the Court to determine whether there was negligence or not. The opinions of the experts are not conclusive. But when an operation itself is a complicated and critical one, and acquaintance with anatomy, physiology or other subjects of expert medical knowledge, skill and experience are essential, jury or Court may not be justified in disregarding such opinions and reaching conclusions based on views contrary to those of the experts. That is not the case here. Effective antecedent precautions were not taken and ordinary experience of jurymen or Court is sufficient to enable them to pass upon the question whether such conduct constituted negligence. In my opinion it is clearly so in this case. [Emphasis added.]

48 Coyne J.A. then adds the following at pp. 85-86:

Whether or not it is negligence to omit to use sponges with ties or to have a count kept is not a matter which requires an expert to decide; it is not special surgical skill that is in question. Such skill is not necessary to answer the question. The point involved is negligence or no negligence. It is not a matter here which requires an expert to decide. General practice of the defendant and some others does not constitute a complete defence. It is some evidence to be taken into consideration on the question of negligence but it is not conclusive on Court or jury. If it were a defence conclusive on jury or Court, a group of operators by adopting some practice could legislate themselves out of liability for negligence to the public by adopting or continuing what was an obviously negligent practice, even though a simple precaution, plainly capable of obviating danger which sometimes might result in death, was well known.... If a practitioner refuses to take an obvious precaution, he cannot exonerate himself by showing that others also neglect to take it. [Emphasis added.]

49 As well, Coyne J.A. emphasized that the case involved no difficult or uncertain questions of medical or surgical treatment nor any matters of a scientific or highly technical character. It was simply a matter of whether obvious and simple precautions, easily understood by ordinary individuals, were required to be taken. Coyne J.A. remarked (at pp. 86-87):

Ordinary common sense dictates that when simple methods to avoid danger have been devised, are known, and are available, non-user, with fatal results, cannot be justified by saying that others also have been following the same old, less-careful practice; and that when such methods are readily comprehensible by the ordinary person, by whom, also, the need to use them or not is easily apprehended, it is quite within the competence of Court or jury, quite as much as of experts to deal with the issues; and that the existence of a practice which neglects them, even if the practice were general, cannot protect the defendant surgeon. [Emphasis added.]

50 In brief reasons, the Supreme Court of Canada affirmed the reasons of McPherson C.J.M. The Court left open the issue of when it is appropriate for a judge or jury to find a standard medical practice to be unacceptable in terms of taking reasonable care.

51 I conclude from the foregoing that, as a general rule, where a procedure involves difficult or uncertain questions of medical treatment or complex, scientific or highly technical matters that are beyond the ordinary experience and understanding of a judge or jury, it will not be open to find a standard medical practice negligent. On the other hand, as an exception to the general rule, if a standard practice fails to adopt obvious and reasonable precautions which are readily apparent to the ordinary finder of fact, then it is no excuse for a practitioner to claim that he or she was merely conforming to such a negligent common practice.

52 The question as to whether the trier of fact can find that a standard practice is itself negligent is a question of law to be determined by the trial judge irrespective of the mode of trial. It is, of course, for the jury to determine on the evidence what the standard practice is. If the evidence is conflicting on this issue, the jury will have to resolve the conflict. If, as in this case, the evidence is virtually conclusive, the trial judge should instruct the jury that failure to accept the evidence may very well result in an unreasonable verdict which will be set aside. Moreover, unless the nature of the issue is of a kind to bring it within the exception to the general rule, the jury should be instructed that once they have determined on the evidence what the standard is, the only remaining issue is whether the defendant conformed to the standard. On the other hand, if the case is one coming within the exception so that the jury can fix the standard on the basis of common sense and the ordinary understanding of the jury without the assistance of expert testimony, the trial judge must instruct the jury accordingly.

53 To avoid the problem encountered in this case due to the inscrutability of the jury's response to the question relating to negligence, and as a precaution to test the jury's understanding of the instruction, the question to the jury with respect to negligence should require that the jury specify in what respects the defendant was negligent. In a case in which the general rule applies, the answer will reveal whether the jury has understood and applied the judge's instruction that it must accept the standard practice as the legal standard against which the defendant's conduct must be measured. Additionally, in a case which falls within the exception to the general rule and where the jury can fix the standard irrespective of the expert evidence, the answer to the question will ensure that the standard which the jury has adopted is not unreasonable or unknown in law.

54 In this regard, it is noteworthy that the respondent's counsel suggested that the jury be asked to particularize the acts of negligence. This was not done, with the result that the Court of Appeal observed that it is not possible to decide whether the jury did decide the issue of negligence on the ground that the standard practice was itself deficient.

55 A related legal issue arises with respect to the second aspect of this case. If the jury finds that the evidence of standard practice does not establish that there indeed exist a standard practice, can the jury fix the standard without reliance on expert testimony? In my opinion, the answer to this question is the same as the answer previously given. If the alleged act or acts of negligence are such that the jury could reject expert evidence as to standard practice and set the appropriate standard without reliance on expert evidence, then it can do precisely that where the expert evidence fails to establish a standard practice. In either case, the basic question is whether the nature of the issue can be decided on the basis of the ordinary knowledge possessed by the jury or, on the contrary, the matter requires expert evidence because it is beyond the ken of the average juror. Finally, although this is not a matter which arises in this case, I do not foreclose the possibility that exceptionally a jury might find negligence in respect of an issue beyond its ordinary understanding based on expert evidence which it accepts which falls short of establishing a standard practice.

(2) Conclusion on Professional Negligence

56 After correctly charging the jury with respect to the general standard of care relating to physicians, the trial judge commented on the role of evidence of standard practice as follows:

In deciding what risks should have been known to Dr. Korn, evidence of medical experts of custom or general practice is one factor to be considered, but it is not conclusive. It is open to you as triers of fact to find the custom or general practice negligent. [Emphasis added.]

57 I agree with the following characterization by the Court of Appeal of the issue relating to the standard of care which the jury was called upon to decide in this case (at pp. 72-73):

In this case there were difficult, uncertain, highly technical scientific questions requiring information not ordinarily expected of a practising gynaecologist or obstetrician. No jury is capable of deciding on its own what understanding of recent developments the Defendant should bring to his practice. As already described, both Dr. Stewart and Dr. Mascola partially exonerated the Defendant....

Moreover, the evidence established that the state of medical knowledge about AIDS and HIV was highly variable even between highly qualified scientists. There were differences of opinion between public health authorities and practitioners in different medical communities. In our judgment, this was not the kind of case where a judge could properly instruct the jury that it could decide that a practice that conformed to what other practitioners similarly situated were following was negligent. The only proper instruction to be given on at least a part of the case was that the jury should decide whether the Defendant conducted himself as a reasonable physician would in similar circumstances. In our judgment, that required the jury to confine itself to prevailing standards of practice.

58 In light of this characterization, I agree with the conclusion reached by the Court of Appeal that with respect to the first aspect of the case the question of the standard of care was not one which the jury could decide without the aid of expert evidence. The Court of Appeal went on to hold that, since it was impossible to tell from the answer given by the jury whether or not it found that the standard practice followed by the respondent and other AI practitioners in Canada was negligent, a new trial was necessary on this basis alone. In my opinion, however, this is not the reason why a new trial is required in this case. If the claim had been restricted to the first aspect of the case in negligence, a new trial would not have been necessary notwithstanding the misdirection with respect to the standard of care. A new trial would only have been necessary if a finding of negligence, supportable on the evidence, was open to the jury. As I have pointed out above, the Court of Appeal correctly held that a finding of negligence on this branch of the case would not be supportable in that it was not a finding which a jury acting judicially could make. There was no reason to direct a new trial on this aspect of the case and the Court of Appeal, in fact, did not do so. On this aspect the appellant's claim simply fails. The reason that a new trial is necessary is because the jury may have found negligence on the second aspect of the case, namely the failure to adequately screen and follow-up donors.

59 The main contention with respect to this aspect of the claim was that closer questioning and follow-up interviews with donors might have revealed that the donor who infected the appellant was in a high risk category relating to the transmission of STDs. With respect to the foundation in law of this claim the Court of Appeal stated (at pp. 74-75):

HIV is one of a number of sexually transmitted diseases. It was the duty of the Defendant to take reasonable steps to protect his patients against sexually transmitted diseases. If he failed to do so he would be liable even if the specific disease was one which he did not actually foresee.

As we are not able to know upon what ground the jury's verdict rests, there must be a new trial on the issues raised in the pleadings other than those based upon the Defendant's lack of knowledge of the risk of HIV by AI.

60 I agree that infection with HIV is within the same class of injury as other STDs and that the respondent could be liable for the damage caused notwithstanding that he did not foresee that a failure to take reasonable steps to protect his patients could result in HIV infection. It would be sufficient to found liability if it is established that the respondent ought to have foreseen the class of injury. See R. v. Côté, 1974 CanLII 31 (SCC), [1976] 1 S.C.R. 595, at p. 604.

61 The evidence as to standard practice with respect to this aspect of the claim was sketchy and it would be open to the jury to find that there was at the relevant time no standard practice. On the other hand, the jury could also find that the standard practice was not to screen donors or carry out follow-up interviews beyond what the respondent did. In either case, this is an issue that is quite different from that involved in the first aspect of the case. In my opinion, the jury could determine the appropriate standard without reliance on expert evidence. Accordingly, if they decide that no standard practice existed, they can fix the appropriate standard based on their view as to what a prudent and diligent practitioner ought to have done if he or she foresaw or ought to have foreseen that failure to carry out reasonable measures to screen and follow-up donors could result in infection with an STD other than HIV. As well, if the jury finds that the evidence does establish a standard practice, it is entitled to consider whether that practice is itself consistent with the measures which the hypothetical reasonable prudent practitioner would take in the same circumstances.

62 There is one other aspect to the claim in negligence which does not fit neatly into either of the two aspects to which I have had reference. I refer to the allegation that the respondent should have used frozen rather than fresh semen. The Court of Appeal doubted that a jury could find for the appellant on this ground but did not exclude this issue from the matters to be dealt with at the new trial. It appears that this allegation was entirely dependent on the Elisa test being available which did not occur until after January of 1985. I find nothing else in the evidence to support a finding of negligence based on this ground. Accordingly, I would not have included this allegation in the matters to be dealt with at the new trial were it not for the fact that it appears to be included in the order in appeal against which the respondent did not cross-appeal. I would, therefore, simply affirm the order of the Court of Appeal in respect to the negligence claim.

B. Warranty Issues

63 As an alternative ground of appeal, the appellant contends that the respondent breached a warranty that the semen provided by the donor would be of merchantable quality and fit for its purpose. In other words, she claims that the respondent warranted that the semen would not be contaminated with any STDs, including HIV, such that it would injure the appellant. Under this theory, even absent any negligence, the respondent would be held strictly liable under contract for failing to provide uncontaminated sperm. The appellant puts forward three bases for the existence of a warranty. First, she claims that there was an express contractual warranty found in the Information Sheet provided to the appellant where the respondent promised that no donor was a homosexual or drug abuser. Secondly, the appellant relies on the Sale of Goods Act, s. 18, which imposes liability on a seller of a good if the good is not fit for its purpose. This warranty is statutorily implied into contracts of sale. Finally, the appellant argues that there was an implied warranty at common law that the good provided in a contract for goods and services would not be defective. I will deal with each ground in turn.

(1) Express warranty in the Information Sheet

64 The appellant asserts that, in the Information Sheet provided, the respondent promised that the donor would not be homosexual or a drug abuser and that these statements were made to assure her that the respondent had taken precautions to ensure that the semen would be free from contamination. The appellant claims that the doctor breached this express warranty.

65 In my view, this ground of appeal must fail. There is no indication that there was any intention on the part of either of the parties that there should be contractual liability in respect of the statement in the Information Sheet. In fact, it appears that the appellant did not have any recollection of receiving any written information from the respondent regarding the AI procedure. It is apparent that the purpose of the Information Sheet was precisely what the name suggests. That is, to provide general information to the patient. There was no intention that the statements contained therein constitute an express warranty that the donor would not be a homosexual.

66 Moreover, I note that, at trial, the appellant did not rely on the existence of an express warranty in the Information Sheet and the issue was not raised in the trial judge's charge to the jury. Nor did the appellant argue this issue at the Court of Appeal. Therefore, it appears that the appellant is relying on the express warranty in the Information Sheet for the first time in this Court. In any event, as I have stated, I do not believe there is any merit to this argument.

(2) Warranty under the Sale of Goods Act

67 In order for the Sale of Goods Act to apply, a contract must primarily be for the purpose of selling goods. If the sale of a good is merely incidental to what is primarily a contract for services, then the statute will not imply a warranty. As Legg J. observed in Gee v. White Spot Ltd. (1986), 1986 CanLII 776 (BC SC), 7 B.C.L.R. (2d) 235 (S.C.), in order to come within the Sale of Goods Act, a contract need not be one exclusively for the sale of goods. However, the sale of a good must be the primary purpose of the contract. Whether a contract is primarily one for the sale of goods or primarily one for services depends upon the essential character of the agreement. As G. H. L. Fridman wrote at p. 25 of Sale of Goods in Canada (2nd ed. 1979):

... if the primary object of the contract is the transference of property in something which was not originally the property of the "buyer", the contract will be one of sale of goods: but if the primary purpose of the parties is the performance of certain work, or the provision of services, incidentally to which property in goods is to pass from one party to the other, the contract will not be one of sale of goods.

68 Thus, the preliminary question that arises is whether the AI procedure performed by the respondent primarily involved a contract for the sale of semen, or was it primarily a contract for medical services. If the procedure is properly characterized as the latter, then the appellant's argument under the Sale of Goods Act must fail. As the Court of Appeal did, I intend to address this issue assuming that there was in fact a sale of semen between the appellant and respondent. However, I note that it is not entirely clear that this was the case.

69 At trial, the appellant submitted that the question of whether a contract is characterized as a sale of goods or one primarily for services is a question of fact which should be left to the jury. The trial judge instructed the jury that there was a contract for medical services as well as for goods. The jury was then charged that they must determine, as a question of fact, whether the contract was "primarily formulated to offer a sale of services, or primarily formulated to offer the sale of a good or goods". The jury was told that the Sale of Goods Act only applied if the contract was primarily for the sale of semen.

70 At the outset, I find it somewhat problematic that the characterization of the contract was a matter left to the jury to decide. It seems that determining whether a particular statute has application in a given case is a question of law which ought to be resolved by the trial judge. Although not a pure question of law, the characterization of the contract as one primarily for the sale of goods is certainly one of mixed law and fact. Only once it is determined that the Sale of Goods Act has any application in the circumstances should the jury be instructed to determine whether the statutory warranty was breached on the facts. In any event, I need not determinatively settle the issue of whether this preliminary matter ought to have been left to the jury, as I am of the view that the jury correctly resolved the issue.

71 The jury correctly concluded that the contract to perform the AI procedure on the appellant was primarily a contract for medical services and not a sale of semen. To hold otherwise would be to distort the true nature of the whole agreement between the parties. The provision of the semen was obviously an important component to the AI procedure; however the primary reason the appellant went to a gynaecologist was for professional medical services and expertise. As the respondent argues, he provided medical services to the appellant in order to assist her to become pregnant by way of AI. Although donor semen was a necessary component of this process, the contract was not primarily for a sale of semen.

72 It is not relevant that the actual AI procedure was relatively simple and quick. The appellant still relied on the respondent's expertise in the screening process for donors, the collection of the semen, the insemination procedure itself, and the provision of medical advice and information concerning any risks and the possibility of success of the AI procedure. It cannot be contended that the contract was one primarily for the sale of semen such as to attract the application of the Sale of Goods Act.

(3) Warranty implied by the common law

73 The fact that the contract was primarily one for services does not end the possibility that there was an implied warranty that the semen would not be contaminated with HIV. The appellant also bases her claim under the common law by virtue of which a warranty may be implied in a contract for goods and services.

74 In the charge to the jury, the trial judge instructed that under the common law, where a contract is primarily for medical services, the doctor simply warrants to meet the standards of a reasonably competent person practising in the field. The trial charge explained to the jury that the warranty under common law required the jury to determine whether the respondent was negligent in failing to detect or remove the defect from the semen. In other words, the charge to the jury was to the effect that the implied warranty under common law was the same as the test for negligence in these circumstances.

75 In order to determine whether a common law warranty ought to be implied in the circumstances of this case and if so, whether the trial judge erred by equating the content of that warranty to negligence, it is necessary to review the authorities in this area in some detail.

76 In England, strict liability is imposed for implied warranties of fitness for defective products supplied under contracts for work and materials as well as under contracts for sale. The purchaser has a remedy against the business seller, even absent negligence. The seller can always recover, up the chain of production, from the manufacturer.

77 A leading English case in this area is G. H. Myers and Co. v. Brent Cross Service Co., [1934] 1 K.B. 46. That case involved the installation of faulty connecting rods in an automobile. One of the rods had a latent defect that the repair dealer could not have detected by reasonable care. Du Parcq J. held that, at common law, the dealer was liable under an implied warranty that in a contract for work done and material supplied, the material would be fit for its purpose. It was held that there should not be any distinction between a contract where goods are supplied only and a contract where goods are supplied in the course of a contract to perform services as this would be an arbitrary distinction. At pages 53-54, du Parcq J. held:

The view to which I have come is that the liability of the person supplying goods in the course of doing work and labour is certainly not less than the liability of the person selling goods ....

78 Thus, the court held that an individual who agrees to do work and provide goods has no lesser obligation than an individual who simply contracts to provide the goods. However, it was recognized that not all contracts for goods and services will contain an implied warranty that the materials will not be defective. At page 55, du Parcq J. noted the following:

That depends upon the terms of the contract, and I think that the true view is that a person contracting to do work and supply materials warrants that the materials which he uses will be of good quality and reasonably fit for the purpose for which he is using them, unless the circumstances of the contract are such as to exclude any such warranty. [Emphasis added.]

The foregoing passage contemplates that there may be certain circumstances such that a contract to supply goods and services will not contain an implied warranty that, regardless of any negligence, the goods will not be defective.

79 The decision in G. H. Myers was cited with approval by the House of Lords in Young & Marten Ltd. v. McManus Childs Ltd., [1969] 1 A.C. 454 (H.L.). That case dealt with contractors who were building a house and subcontracted the roofing work. The roofing tiles which were used contained a latent defect which was not apparent upon inspection. As a result, the contractor suffered damages and claimed that there was an implied warranty of quality or fitness. Lord Reid observed that, in such cases where there is no evidence of negligence by the manufacturer, unless the owners of the house could recover from the subcontractor, they would be without a remedy. However, if the subcontractors were held liable, they could generally recover from the manufacturer under the Sale of Goods legislation. The court adopted the general rule which was laid down by du Parcq J. in G. H. Myers. Lord Reid also made the following relevant observation (at p. 468):

It appears to me that less cogent circumstances may be sufficient to exclude an implied warranty of quality where the use of spare parts is only incidental to what is in essence a repairing operation where the customer's main reliance is on the skill of the tradesman, than in a case where the main element is the supply of an article, the installation being merely incidental.

80 In the result, the court affirmed that, in the circumstances of that case, the common law implied a warranty and the subcontractor was liable for the defective tiles. It was not conclusive that there was an element of services along with the supply of goods.

81 The principles espoused in the English case law have also been adopted in Canada. A leading case is G. Ford Homes Ltd. v. Draft Masonry (York) Co. (1983), 1983 CanLII 1719 (ON CA), 43 O.R. (2d) 401 (C.A.). In that case, the plaintiff supplied and installed staircases in two houses being constructed by the defendant. It turned out that the staircases contravened the Ontario Building Code and had to be replaced. The issue was whether there was an implied term in the contract that the staircases would conform to the Building Code. Cory J.A., as he then was, delivered the decision for the court. At page 403, he remarked:

When may a term be implied in a contract? A court faced with that question must first take cognizance of some important and time-honoured cautions. For example, the courts will be cautious in their approach to implying terms to contracts. Certainly a court will not rewrite a contract for the parties. As well, no term will be implied that is inconsistent with the contract. Implied terms are as a rule based upon the presumed intention of the parties and should be founded upon reason. The circumstances and background of the contract, together with its precise terms, should all be carefully regarded before a term is implied. As a result, it is clear that every case must be determined on its own particular facts. [Emphasis added.]

82 Cory J.A. reviewed the decisions in Young & Marten Ltd. and G. H. Myers and noted that the warranties implied at common law apply to contracts for goods and services. Thus, unlike the Sale of Goods Act, such warranties are equally applicable to contracts for the provision of work and materials. Secondly, unless the circumstances of the case are sufficient to exclude the warranty, there will be an implied term "that the materials used will be of merchantable quality and that those materials will be reasonably fit for the purposes for which they were intended" (p. 404).

83 Accordingly, it is apparent that apart from the Sale of Goods Act, a court must consider whether a common law warranty of fitness and merchantability should be implied into the contract which includes services as well as the provision of materials. However, such a warranty will not be implied in all circumstances. The court must examine the specific nature of the contract and the relationship between the parties in order to assess whether it was the intention of the parties that such a warranty be implied. As Cory J.A. observed, courts must be very cautious in their approach to implying contractual terms.

84 It is important to note that a rationale for implying warranties in contracts of goods and services is that a supplier of goods generally has recourse against the manufacturer under the Sale of Goods Act as a result of the statutory conditions imposed. Thus, one can always proceed up the chain of production and ultimately recover from the one who should bear responsibility for the production of faulty goods. From time to time, the supplier will be unable to recover from the manufacturer, for example, owing to insolvency or limitation periods. However, arguably it is better that the purchaser be compensated and the supplier occasionally bear the cost of defects than leaving the consumer without a remedy. It is important to keep this policy rationale in mind when considering whether such a warranty should be implied in the context of this case, which deals with biological substances.

85 Thus, the question that must be addressed in this appeal is whether, in the circumstances of this case, where there is a contract to conduct a medical procedure involving the use of biological material (semen), it is appropriate to imply a term that the semen was warranted to be without any defects (HIV contamination). Secondly, if it is appropriate to imply a term into the contract for medical services was the trial judge correct in holding that the respondent only warranted to take reasonable care to ensure that semen would not be used if it were contaminated with any STDs. In this regard, one must consider whether the nature of the contract at issue in this appeal is analogous to the commercial contracts which were dealt with in the English and Canadian cases discussed above. It is important to address any policy considerations and the implications of imposing warranties in these circumstances.

86 In answering these difficult questions, it is useful to survey some of the American jurisprudence which has dealt with this issue more frequently, in the context of warranties for the supply of blood. In my view, these cases are directly analogous to the supply of semen in an AI practice.

87 The leading American authority on this issue, which has been followed many times subsequently, is Perlmutter v. Beth David Hospital, 123 N.E.2d 792 (N.Y. 1954). In that case, a patient received a blood transfusion during the course of a medical procedure performed at the hospital. The blood contained jaundice viruses and as a result the plaintiff became infected. There was no question of any negligence on the part of the hospital as there were no means available for detecting the contamination in the blood. The plaintiff sought recovery under the Sales Act arguing that there was an implied warranty that the blood would be fit for its purpose and of merchantable quality. The issue turned on whether the transaction constituted a sale under the legislation. It must be observed at this point that the plaintiff did not argue that there was an implied warranty under the common law for material supplied under a contract for service and material. Nonetheless, much of the discussion on the Sales Act is relevant to the case at bar.

88 The majority of the court examined the nature of the contract between the hospital and held (at p. 794):

The essence of the contractual relationship between hospital and patient is readily apparent; the patient bargains for, and the hospital agrees to make available, the human skill and physical materiel of medical science to the end that the patient's health be restored.

Such a contract is clearly one for services, and, just as clearly, it is not divisible. Concepts of purchase and sale cannot separately be attached to the healing materials - such as medicines, drugs or, indeed, blood - supplied by the hospital for a price as part of the medical services it offers. That the property or title to certain items of medical material may be transferred, so to speak, from the hospital to the patient during the course of medical treatment does not serve to make each such transaction a sale. "Sale' and transfer' are not synonymous", and not every transfer of personal property constitutes a sale.... It has long been recognized that, when service predominates, and transfer of personal property is but an incidental feature of the transaction, the transaction is not deemed a sale within the Sales Act.

89 It should be observed that the majority focused on the sharp distinction between a contract of sale and one for services that exists in this context. While this is important for determining whether sale of goods legislation is applicable, it is not as crucial in the context of implied warranties under the common law which are also available for contracts of service where a good is furnished. Nonetheless, the remarks in Perlmutter are highly relevant since, as was noted by Lord Reid in Young & Marten Ltd., it will be less likely that such a warranty will be implied in a contract which is primarily for services where the transfer of the good is merely incidental.

90 In Perlmutter, it was held that the supplying of blood was entirely subordinate to the main purpose of the hospital, which was providing trained professionals and specialized facilities to care for the patient's health. The patient does not bargain for blood. Rather, it is the skill of the medical staff and the facilities which is sought. Fuld J. concluded that supplying the blood was entirely incidental to the services performed. In my view, the same can be said of the semen in the present case, as I discussed earlier in respect of the Sale of Goods Act.

91 In examining the nature of the contract at issue, Fuld J. noted some of the policy considerations which are relevant in this context. At page 795, the following was observed:

If, however, the court were to stamp as a sale the supplying of blood - or the furnishing of other medical aid - it would mean that the hospital, no matter how careful, no matter that the disease-producing potential in the blood could not possibly be discovered, would be held responsible, virtually as an insurer, if anything were to happen to the patient as a result of "bad" blood.... [Emphasis added].

92 Although these comments were made in the context of sale of goods legislation, they apply equally to the situation of an implied warranty under common law. In either situation, a medical practitioner would be held strictly liable for the biological products employed in the medical procedures, notwithstanding that it may be impossible for the doctor to detect any risks. This would have the effect of making physicians insurers of the biological substances that are used in medical procedures.

93 While it is true that the primary purpose of the implied warranty is to hold the supplier of goods liable notwithstanding the absence of negligence, different considerations apply in the context of the medical profession than in the ordinary commercial context. As Fuld J. observed (at p. 795):

The art of healing frequently calls for a balancing of risks and dangers to a patient. Consequently, if injury results from the course adopted, where no negligence or fault is present, liability should not be imposed upon the institution or agency actually seeking to save or otherwise assist the patient.

94 Furthermore, it should be noted that unlike in the ordinary commercial context, the doctor cannot trace the liability back to the initial manufacturer. Biological products are not manufactured goods in the same sense as commercial goods. The underlying rationale for the strict liability imposed under the Sale of Goods Act or by virtue of an implied warranty at common law does not apply to goods which are not manufactured in the ordinary sense. Absent negligence on the part of the donor (for example, if he knew he had AIDS), one would hardly expect that the respondent, in the present case, could recover from the donor for the semen contaminated with HIV, either under the Sale of Goods Act or under an implied warranty at common law. This is unlike a commercial contract where it would be open to a supplier to sue the manufacturer, even absent negligence on its part.

95 Moreover, it must be recognized that biological products such as blood and semen, unlike manufactured products, carry certain inherent risks. In some ways, these substances are inherently dangerous, although they are essential to medical procedures. Whether a doctor is trying to save a patient's life via a blood transfusion, or is simply attempting to assist a patient to become pregnant by AI, the physician cannot control the safety of these products beyond exhibiting the reasonable care expected of a professional to ensure that the biological substance is free from harmful viruses. By contrast, in the commercial world, the manufacturer has control over the goods. If they cannot be manufactured to be safe, then the products ought to be removed from the market. In medicine blood is essential to a variety of procedures in order to save lives. While arguably, AI is not in the same category as other life saving techniques, it is nonetheless a very important medical procedure. As long as the entire procedure does not amount to an unreasonable risk such that it ought not to be offered at all, the patient is entitled to weigh those risks and elect to proceed.

96 In Fisher v. Sibley Memorial Hospital, 403 A.2d 1130 (D.C. 1979), an action was commenced for injury suffered when a patient contracted hepatitis after a blood transfusion supplied by the hospital. It was not possible to detect the virus in the blood. The Court of Appeal held that the theories of implied warranty and merchantability and strict liability in tort had no place in the context of a hospital furnishing blood. At pages 1132-33, Gallagher J. held as follows:

"The activities involved in the transfusion of whole blood, a component of the living body, from one human being to another may be characterized as sui generis in that the sequence of events involve acts common to legal concepts of both a sale and a service. Moreover, it seems to us that under the facts in the case before us it would be unrealistic to hold that there is an implied warranty as to qualities of fitness of human blood on which no medical or scientific information can be acquired and in respect to which plaintiff's physician has the same information, knowledge, and experience as the supplier." (Balkowitsch v. Minneapolis War Memorial Blood Bank, Inc., 270 Minn. 151, 132 N.W.2d 805, 811 (1965).)

We agree with those courts which hold that the furnishing of blood is more in the nature of a service than of a sale of goods. Treating blood transfusions as an incidental service performed by hospitals comports with reality, and with the policies underlying merchantability liability. Although theoretically a seller's inability to discover defects in the goods he sells is not relevant to a warranty cause of action, we cannot ignore the difficulty of detecting hepatitis in blood given the current state of medical knowledge. To characterize as a sale the supplying of blood would mean that the hospital, no matter how careful, would be held responsible, virtually as an insurer, if the patient were harmed as a result of impure blood. After balancing the safety of the individual with the interests of the hospital (in light of the absence of an adequate test to determine the presence of hepatitis in the blood) and the public interest in assuring the ready availability of blood for medical treatments, we are reluctant to extend ... merchantability liability to a nonsale transaction by analogy or by characterizing the transaction as a sale. [Emphasis added.]

97 The court was of the view that it was unnatural to force a blood transfusion into the commercial sales mould since the main object of the hospital is health care and treatment. It was also noted that blood products are "unavoidably dangerous" (p. 1134) and the patient relies on the doctor's skill rather than any warranties of fitness.

98 I also find the case of St. Luke's Hospital v. Schmaltz, 534 P.2d 781 (Colo. 1975) to be persuasive. That case also involved a blood transfusion resulting in the patient contracting hepatitis. The court followed the reasoning in Perlmutter and rejected the imposition of strict liability in tort or on the basis of a warranty due to policy considerations. It was held that a realistic view of the relationship between a hospital and a patient is not that of a commercial transaction where a good is sold for a price. Rather, the patient bargains for the skill and materials of medical science in order to care for the patient's health. It is simply not realistic to view the furnishing of blood as a sale of a product.

99 Although the majority of American cases have followed Perlmutter, a few have criticized the policy analysis conducted. For example, in Cunningham v. MacNeal Memorial Hospital, 266 N.E.2d 897 (Ill. 1970), the court again considered the situation of a patient who contracted hepatitis through a blood transfusion. The court rejected the idea that an implied warranty did not arise because no "sale" was involved. Notwithstanding that blood is not a manufactured article of commerce, the court considered it a product which is distributed for consumption. The blood was sold in a container and the court felt it was unreasonable not to conclude that there was a sale of goods which was divisible from the contract for services.

100 However, the vast majority of American cases have agreed that the policy considerations dictate that, in the context of the provision of medical services, medical professionals should not be held strictly liable under warranty for goods used in the provision of those services. In my view, the reasoning of the majority in Perlmutter, and the line of cases which follow it, is more apt in the Canadian context with respect to implied warranties at common law. Although, the American decisions did not deal with exactly this issue, the remarks made with respect to warranties under sale of goods legislation provide support for the view that it is inappropriate to imply a warranty in these circumstances under the common law.

101 I am, therefore, in agreement with the Court of Appeal's conclusion in the present case that it would be inappropriate to imply a warranty of fitness and merchantability in the circumstances of this case. As Cory J.A. observed in G. Ford Homes, courts should be very wary about implying terms of contracts and there may be circumstances which ought to exclude such a warranty. In the present case, the action against the respondent for injury resulting from the AI procedure should be confined to negligence. I would adopt the following conclusion of the Court of Appeal at p. 85:

In the face of the American experience, we are unable to identify any policy reason why a physician should face stricter liability for "goods" which are furnished to a patient in the course of medical service than he or she would be for any lack of professional care and skill which must be brought to every healing or treating engagement.

102 I note that even if I am wrong in my conclusion that a warranty should not be implied in the circumstances of this case, I would hold, as the trial judge did, that any warranty would simply be to take reasonable care. In other words, if the parties intended that there be any contractual warranty, the nature of the warranty in this case would simply be to the effect that the respondent exercise diligence and care in performing the AI and selecting the donors. The contract was primarily one for medical services and parties would not have contemplated that the respondent would warrant the success of the procedure nor that the semen would not be contaminated with an STD. As the trial judge stated, the respondent "undertakes to meet the standards of a reasonably competent person practicing in his field". It would be unreasonable to hold the respondent to any higher standard.

103 In the result, the appellant's argument that there was an implied warranty at common law that the donor's semen would not be infected with HIV must also fail.

C. The Rough Upper Limit on Non-Pecuniary Damages

104 In Andrews v. Grand & Toy Alberta Ltd., 1978 CanLII 1 (SCC), supra, this Court established a rough upper limit for non-pecuniary damages of $100,000. In Lindal v. Lindal, 1981 CanLII 35 (SCC), supra, it was recognized that it was appropriate to take into account the effect of inflation and thus, the upper limit could vary with changing economic conditions, including the decrease in purchasing power caused by inflation. Thus, at the time of the present case, the rough upper limit was approximately $240,000. However, the jury awarded non-pecuniary damages of $460,000. The Court of Appeal held that the non-pecuniary damages could not exceed the rough upper limit and a new trial was ordered to assess damages in general. The Court of Appeal did not hold that the trial judge was necessarily in error for failing to charge the jury on the upper limit, in light of the views expressed by counsel. However, if a jury is not given such an instruction, the damages award must be corrected in order to conform with the jurisprudence.

105 There are two issues which must be addressed. First, whether this was one of those exceptional cases referred to in Andrews such that the rough upper limit ought not to have been applied. Secondly, whether the trial judge ought to have charged the jury to consider the upper limit when assessing non-pecuniary damages. The appellant submits that this case is unique and that the non-pecuniary damages should be permitted to exceed the limit.

106 In Andrews, this Court set out the rationale for the upper limit placed on the assessment of non-pecuniary damages for pain and suffering. It is simply impossible to put a money value on the non-pecuniary losses which have been suffered by the plaintiff. Therefore, the award of non-pecuniary damages "is a philosophical and policy exercise more than a legal or logical one" (p. 261). This Court has adopted the "functional" approach to assessing such damages. That is, rather than attempting to evaluate the loss of happiness, non-pecuniary damages seek to provide the plaintiff with reasonable solace for the misfortune suffered. Money acts as a substitute for the pleasure and enjoyment which has been lost and endeavours to alleviate, as far as possible, the pain and suffering that the plaintiff has endured and will have to endure in the future.

107 The amount of the award depends on the ability of money to ameliorate the condition of the victim in his or her particular situation. Non-pecuniary damages should only be awarded to the extent that they can serve a useful purpose by providing an alternative source of satisfaction.

108 However, as Dickson J. observed in Andrews, once non-pecuniary damages are viewed from a functional perspective, excessively large amounts should not be awarded once an individual is fully compensated in terms of future care and other pecuniary losses. Awards for non-pecuniary loss are inherently arbitrary or conventional. Therefore, once the future care of the plaintiff is adequately addressed, it is more appropriate to consider policy issues in limiting damage awards. In particular, the social burdens of excessive awards must be considered, as extravagant claims can pose a significant burden on society. In fact, in Arnold v. Teno, 1978 CanLII 2 (SCC), [1978] 2 S.C.R. 287, it was observed that the real social cost of exorbitant damage awards had been particularly evident in the United States with respect to medical malpractice cases. In Lindal, this Court reiterated the sentiments expressed in the Andrews trilogy, recognizing that the claim for non-pecuniary damages in the case of a severely injured plaintiff can potentially be limitless.

109 In Lindal, the Court refused to exceed the upper limit for non-pecuniary damages even though the injuries were arguably even more severe than those suffered in Andrews. Given that the upper limit is merely a rough and somewhat arbitrary cap on non-pecuniary damages, an award exceeding the limit is not justified simply because an injury is more serious that those in the trilogy of cases. As Dickson J. stated in Lindal (at pp. 642-43):

It is true that the Court in Andrews spoke of exceeding the limit of $100,000 in "exceptional circumstances". The variety of possible fact situations is limitless, and it would be unwise to foreclose the possibility of ever exceeding the guideline of $100,000. But, if the purpose of the guideline is properly understood, it will be seen that the circumstances in which it should be exceeded will be rare indeed. We award non-pecuniary damages because the money can be used to make the victim's life more bearable. The limit of $100,000 was not selected because the plaintiff could only make use of $100,000 and no more. Quite the opposite. It was selected because without it, there would be no limit to the various uses to which a plaintiff could put a fund of money. The defendant, and ultimately, society at large, would be in the position of satisfying extravagant claims by severely injured plaintiffs.

110 In my view, the above passage adequately addresses the arguments of the appellant in the present case. Essentially, she contends that the evidence demonstrated the uses to which the money could be put. However, this is not a proper rationale supporting an award of non-pecuniary damages in excess of the limit. There is no doubt that the appellant has suffered immensely as a result of this tragedy. It is also apparent that AIDS is a dreadful disease which will eventually take the life of the appellant prematurely. However, with respect to non-pecuniary losses, I do not believe that the present case is any different than other tragedies, such as those which befell the plaintiffs in the Andrews trilogy. I agree with the following comments of the Court of Appeal (at p. 87):

While recognizing that the Plaintiff's injuries are markedly different in kind from those under consideration in the trilogy, we cannot conclude that the principle adopted by the Supreme Court of Canada, aimed at controlling the social burden of damage awards, should not be applied in this case. It follows, in our view, that the Plaintiff's non-pecuniary damages cannot exceed the adjusted rough upper limit.

111 The second question which arises with respect to non-pecuniary damages is whether the trial judge ought to have charged the jury on the upper limit. This is a difficult question because, on the one hand, in a case in which a jury would not have awarded damages in the range of the upper limit, the jury will be unduly influenced by an instruction that refers to a higher figure. On the other hand, it seems wrong not to advise the jury in an appropriate case that as a matter of law and policy, there is a limit. The jury may struggle for nought with the assessment only to learn later that their labours were in vain. Moreover, judges are permitted, and indeed required, to give the jury guidance as to the proper conventional figure. In Crosby v. O'Reilly, 1974 CanLII 173 (SCC), [1975] 2 S.C.R. 381, Laskin C.J., for the Court, stated, at pp. 386-87:

I cannot agree with the Alberta Appellate Division that where a survival action for the benefit of a deceased's estate is tried by judge and jury the jury should be instructed as a matter of law that $10,000 is the present upper limit of an award. I do not think that damages can be so exactly defined by putting them on the basis of a legal limitation. At the same time, it is only common sense, where an appellate court is to have the final say on what is a proper conventional figure, that the jury be given careful guidance lest the result be, as here, an extravagant figure leading to successive appeals at a risk of costs that will eat up the ultimate award. Rather than fix the direction as one of law governing the upper limit of an award, the trial judge should direct the jury, in the light of the evidence respecting the deceased in all of his or her qualities, mode of life and prospects, in the light of age and physical condition, that a figure beyond a particular sum, which may be less than $10,000, may be regarded as excessive.

It would be impossible for a trial judge to comply with this instruction with any degree of intellectual honesty and not mention that there is an upper limit. Indeed, the statement by Laskin C.J. clearly implies that if an upper limit had existed, the jury should be told about it.

112 Accordingly, in my view, the trial judge should instruct the jury as to an upper limit, if, after considering the submissions of counsel, he or she is of the opinion that the damages by reason of the type of injury sustained might very well be assessed in the range of or exceeding the upper limit. The instructions may include an explanation of the reason for the cap.

113 On the other hand, if the trial judge is of the view that the injuries involved will not likely produce an award approaching the rough upper limit, it is best that the trial judge not charge the jury on the matter. The upper limit, like any other matter of law, need not be placed before the jury where the issue does not reasonably arise on the facts of the case.

114 Whether the jury is or is not advised of the upper limit, if the award exceeds the limit, the trial judge should reduce the award to conform with the "cap" set out in the trilogy and adjusted for inflation. While a trial judge does not sit in appeal of a jury award, the trilogy has imposed as a rule of law a legal limit to non-pecuniary damages in these cases. It would be wrong for the trial judge to enter judgment for an amount that as a matter of law is excessive. While it is true that the matter can be corrected on appeal, an appeal may be unnecessary if the correct amount is fixed at trial.

115 In the present case, counsel for the respondent submitted that this was not a case which would approach the upper limit. However, it was submitted that the jury should be instructed on the limit if the appellant's counsel or the trial judge felt otherwise. Counsel for the appellant did not request that the trial judge charge the jury on the upper limit. In these circumstances, I agree with the Court of Appeal that it was reasonable for the trial judge not to give an instruction on the upper limit. However, since the damages awarded for non-pecuniary losses far exceeded the upper limit, the trial judge ought to have reassessed the award in accordance with the Andrews trilogy.

116 In light of the fact that a new trial will be ordered with respect to the issues surrounding liability, I agree that the issue of non-pecuniary damages should also be dealt with by the jury. Given that the jury at the first trial assessed the damages in excess of the limit, it will be appropriate for the trial judge to charge the jury with respect to the upper limit at the new trial.

117 The Court of Appeal also ordered a new trial on the issue of damages generally. The Court merely stated that it was not satisfied that the damages for the past and future income losses and the future care costs were properly assessed. The Court of Appeal did not explain on what basis it found the assessment to be lacking. However, the appellant did not argue that the Court of Appeal was in error with respect to ordering a new trial on these pecuniary damages. In her factum, however, the appellant does seek an order that the jury's awards for non-pecuniary damages and for past and future wage loss be confirmed.

118 Given that the appellant has not demonstrated any error with respect to the Court of Appeal's conclusions on damages, I am not inclined to interfere with the Court of Appeal's holding that there should be a new trial on the issue of damages in general.

VI. Disposition

119 In the result, the appeal is dismissed with costs and the judgment of the Court of Appeal is affirmed. The new trial will be limited to: (1) a determination of the issue of negligence other than the allegation relating to the first aspect of the claim based on the defendant's lack of knowledge of the risk of HIV infection by A1, and (2) an assessment of damages in general in respect of the claim in negligence. With respect to non-pecuniary general damages, the trial judge should instruct the jury on the upper limit in accordance with the principles in the Andrews trilogy and Lindal.

120 It will also not be open to the appellant to raise the issues relating to warranties under the Sale of Goods Act, at common law or by virtue of the Information Sheet. These issues have been resolved in this appeal.

//L'Heureux-Dubé J.//

The following are the reasons delivered by

I. L'Heureux-Dubé J. -- I have read the reasons of my colleague Justice Sopinka and, save for one point, I generally agree with them as well as with his disposition of the appeal. The issue on which I part ways with my colleague concerns the trial judge's instruction to the jury on the upper limit of non-pecuniary damages.

II. According to my colleague, the trial judge should instruct the jury on such upper limit if he or she believes that the damages "might very well be assessed in the range of or exceeding the upper limit", as opposed to that not being "likely" (paras. 112-113). Two arguments are relied on to support this position: first, the possibility of wasting the jury's time absent such instruction, and second, the impossibility for the trial judge to guide the jury, in light of the evidence, as to a proper conventional figure of damages, without mentioning the upper limit. In my view, these arguments are not persuasive.

III. It has always been held improper for the trial judge or counsel to express any views as to the quantum of damages. This is echoed by the well- established principle that the amount of damages is a question of fact for the jury to determine, and not a question of law for the judge: Negligence Act, R.S.B.C. 1979, c. 298, s. 6. By holding that the judge should instruct the jury as to the upper limit of non-pecuniary damages, my colleague is indirectly making their assessment a question of law. In my view, it is impossible to reconcile this approach with the separate functions performed by judge and jury. For instance, as my colleague puts it, "in a case in which a jury would not have awarded damages in the range of the upper limit, the jury will be unduly influenced by an instruction that refers to a higher figure" (para. 111). Moreover, as a result, the benefit of the trial by jury is lost, namely the value of the independent judgment of lay people. In that light, economic considerations, such as the time and the cost involved in the jury's determination of the quantum of damages, are irrelevant. As well, as the Ontario Court of Appeal stated in Gray v. Alanco Developments Ltd. (1967), 1967 CanLII 221 (ON CA), 61 D.L.R. (2d) 652, at p. 656:

To permit the trial Judge to express such an opinion would be to sanction the admission of opinion evidence unsupported by qualified evidence but based solely upon the Judge's personal experience derived from the evidence or verdicts in other cases. This would be tantamount to countenancing his usurpation of functions committed exclusively to the jury.

IV. These considerations have not been challenged by the Andrews trilogy (Andrews v. Grand & Toy Alberta Ltd., 1978 CanLII 1 (SCC), [1978] 2 S.C.R. 229, Thornton v. Board of School Trustees of School District No. 57 (Prince George), 1978 CanLII 12 (SCC), [1978] 2 S.C.R. 267, and Arnold v. Teno, 1978 CanLII 2 (SCC), [1978] 2 S.C.R. 287). The establishment of an upper limit for non-pecuniary damages only affects the quantum of the awards, not the way in which they are determined. Nowhere was it mentioned that it is incumbent on the trial judge to instruct the jury as to such upper limit. Therefore, the existence of an upper limit as a matter of law remains consistent with not instructing the jury as to such upper limit.

V. Moreover, in light of my colleague's arguments, I am unable to see why the jury should be charged on the upper limit in this case, but not in Hill v. Church of Scientology of Toronto, 1995 CanLII 59 (SCC), [1995] 2 S.C.R. 1130. In Hill, this Court decided that there would not be a cap on damages for defamation. Particular emphasis was placed on the fact that the injury suffered by a plaintiff as a result of libellous statements is markedly different from the kind of injury the Court was concerned with in Andrews, supra, namely severe physical injuries resulting from a motor vehicle accident. In my opinion, the same reasoning must prevail in cases involving other types of injuries, including HIV contamination during an artificial insemination procedure, as in this case. By suggesting that the jury be instructed on the upper limit of non-pecuniary damages, my colleague is in fact muzzling the jury, and, in so doing, institutionalizing the status quo in that novel area of the law, independently of evolving economics and other considerations.

VI. Indeed, when it comes to damages, no figure will ever be "right": the trial judge is not in a better position than the jury to determine the quantum of damages, nor the range in which they might be assessed. It is not to say that jury awards are entitled to complete deference. If an award is excessive, I share my colleague's opinion that the judge should reduce it to an amount that he or she believes is appropriate on the facts of the case in order to avoid an otherwise unnecessary appeal. The jury's discretion in reaching that award, however, must be left unfettered. In my view, this system deals adequately with the guidelines established by the trilogy.

VII. In the present case, there is not an iota of evidence before the trial judge, the Court of Appeal, or this Court as to the inadequacy of this system. Before this Court, only a brief mention was made by the respondent in his factum, as an alternative argument, as to the desirability of adopting the system my colleague Sopinka J. suggests. At the hearing, the issue was not discussed. In his reasons, my colleague does not bring forward any data to demonstrate that there is a crisis in Canada with respect to extravagant jury awards for non-pecuniary damages, which would encourage this Court to modify the law. Nevertheless, on this fragile basis, he advocates a profound change of the law.

VIII. It is important to recall the context in which this issue must be resolved. While it is true that the upper limit was designed, in part, to avoid extravagant non-pecuniary awards, this justification is not restricted to jury awards, but also applies to trials before judge alone. Furthermore, the accuracy of this assumption has been challenged in the Law Reform Commission of British Columbia's Report on Compensation for Non-Pecuniary Loss (1984), at p. 13. In addition to statistics indicating that only a small number of cases are still tried by jury today, I find apposite the following passage by W. H. R. Charles, Charles Handbook on Assessment of Damages in Personal Injury Cases (2nd ed. 1990), at p. 75:

...in Ontario it is defence counsel who tend to request a jury trial for personal injury cases. Whether this is because the unpredictability of their awards, both as to liability and measure of damages, encourages a settlement or whether experience suggests that jury damage awards are less generous than those given by judges sitting without a jury, is difficult to say. If the unpredictable nature of jury verdicts is a major factor in inducing settlement, it seems an ironic justification for their continued existence. Obviously there are other reasons for retaining the jury in civil cases generally, not the least of which is the fact that the jury reflects public opinion.

IX. Finally, from a pragmatic standpoint, to require the trial judge to charge the jury on the upper limit of non-pecuniary damages may be inviting unmeritorious appeals on that part of the judge's instructions to the jury.

X. In conclusion, there is no evidence that the law as it stands is incapable of dealing with the upper limit of non-pecuniary damages, the assessment of which has always been the province of the jury. For the reasons previously set forth, I would leave to the jury the determination of the quantum of damages without instructions as to the upper limit, subject to the judge's power to reduce excessive awards.

XI. In the result, I would dispose of this appeal as suggested by my colleague Sopinka J.

Appeal dismissed with costs.

Solicitors for the appellant: McConnan, Bion, O'Connor & Peterson, Victoria.

Solicitors for the respondent: Harper Grey Easton, Vancouver.

Solicitor for the interveners HIV‑T Group (Blood Transfused) and the Canadian Association of Transfused Hepatitis C Survivors: Kenneth Arenson, Toronto.

Solicitors for the intervener the Canadian Hospital Association: Bull, Housser & Tupper, Vancouver.

Solicitors for the intervener the Canadian Red Cross Society: MacMillan, Rooke & Boeckle, Toronto.