St-Jean v. Mercier, [2002] 1 S.C.R. 491, 2002 SCC 15

Frédéric St-Jean Appellant

v.

Denis Mercier Respondent

Indexed as: St-Jean v. Mercier

Neutral citation: 2002 SCC 15.

File No.: 27515.

2001: April 18; 2002: February 21.

Present: McLachlin C.J. and L’Heureux-Dubé, Gonthier, Iacobucci, Major, Bastarache, Binnie, Arbour and LeBel JJ.

on appeal from the court of appeal for quebec

Civil responsibility – Physicians and surgeons – Duty of prudence and diligence – Victim of car accident bringing action in civil responsibility arising out of treatment of his injuries by orthopaedic surgeon – Trial judge finding surgeon to be without fault and ruling that causation had not been established – Whether Court of Appeal justified in intervening in trial judge’s decision – Whether presumption of causation applies.

Appeals – Standard of review – Victim of car accident bringing action in civil responsibility arising out of treatment of his injuries by orthopaedic surgeon – Trial judge finding surgeon to be without fault and ruling that causation had not been established – Whether Court of Appeal justified in intervening in trial judge’s decision – Whether Supreme Court of Canada has reason to intervene in Court of Appeal’s decision.

The appellant was hit by an automobile and was transported by ambulance to a hospital. He had open fractures in both legs and was bleeding from the head. The appellant was examined by the respondent, an orthopaedic surgeon, later that evening. The respondent was aware of a suspected fracture at the T7 vertebra, but he nonetheless concluded that the fracture was stable or that it was actually only a benign anomaly, thus not impeding his ability to operate. The operation was extremely urgent as it was necessary to treat the open wounds in the legs. Two days later, the respondent decided that the appellant’s condition had sufficiently improved to allow another operation on the leg, which was carried out the next day. That evening, a nurse entered a note that the appellant was unable to move his legs or toes, a condition the respondent attributed to sciatica. A week later, a third operation was carried out. During a visit to the out-patient clinic after the appellant was discharged from the hospital, the respondent witnessed spasms in the appellant’s right leg, a symptom of spastic paraplegia. He consulted a neurologist, who confirmed a condition of spastic paraplegia arising from a medullary contusion. No x-ray of the back was taken. Subsequent x-rays and a myelogram revealed a fracture at the T8 vertebra with a subluxation (partial dislocation) of 9 millimetres in relation to T9. A more detailed review of the x-rays taken following the accident revealed that the fracture was visible but that the subluxation was only 3 millimetres. The respondent proposed surgery to decompress the spinal cord, but the appellant asked to be transferred to another hospital, where the treating neurosurgeon decided not to operate. The appellant was diagnosed with paraparesis, a type of paralysis where the motor activity of the lower limbs is undermined but sensitivity is still retained.

The appellant’s action against the respondent went to trial. After the case had been heard and while the trial judge was deliberating, he fell ill and so the case was transferred to another judge, to be decided on the basis of the evidence entered at trial. The new trial judge found the respondent to be without fault, concluding that he could not be reproached for not having diagnosed a medullary contusion rather than a sciatic condition. He also found that the appellant had not discharged his burden to show causation. The Court of Appeal found that the respondent had committed a fault, but affirmed the trial judge’s ruling that causation had not been established.

Held: The appeal should be dismissed.

The inquiry to be made in this case is the traditional one: was the Court of Appeal correct in intervening in the trial judge’s decision because there was an error of law, an error of mixed law and fact, or an overriding and palpable error of fact? The Court of Appeal concluded that the trial judge made a mistake in the analysis of fault by asking the wrong question in determining whether a fault had been committed. The correct inquiry to be made in assessing whether a professional committed a fault is to ask whether the defendant behaved as would a reasonably prudent and diligent fellow professional in the same circumstances. To ask, as the principal question in the general inquiry, whether a specific positive act or an instance of omission constitutes a fault is to collapse the inquiry and may confuse the issue. What must be asked is whether that act or omission would be acceptable behaviour for a reasonably prudent and diligent professional in the same circumstances. The erroneous approach runs the risk of focussing on the result rather than the means. Here, the Court of Appeal was concerned that a proper inquiry was not made and the wrong standard of fault was applied by the trial judge. This concern entitled the Court of Appeal to intervene.

The Court of Appeal also faulted the trial judge’s decision for not taking a position on the scientific debate as to causation. A trial judge must assess whether the evidence supports, on a balance of probabilities, a finding of causation. It is not enough to say that there are opposing medical theories on causation and that it is not up to the court to decide between them. It was thus open to the Court of Appeal to be concerned that the trial judge had erred in this manner and accordingly undertake its own analysis of causation.

Findings of fault in the law of delict are questions of mixed law and fact, and the standard of review in this instance is one of correctness. Here, this Court is justified in modifying certain findings of fault made by the Court of Appeal, all the while allowing the finding of fault to stand. There can be no reproach against the respondent for not verifying the fracture at T7 with the radiologist. As far as T7 itself was concerned, the respondent had exercised his professional judgement and decided that whatever fracture there may be at that level, it was stable. However, the suspected fracture at T7 together with the violence of the impact, as well as the consequent concern of the respondent about back pain later on, should have led the respondent to examine more fully the state of the spine once the urgent first operation was done. The fault lies in not conducting the requisite neurological tests, and not examining the back properly with the available tests such as more localized x-rays and scans. There is also no reason to find a fault for not taking cognizance of the nurses’ notes, since the respondent personally checked to see how the appellant was recovering after the second operation and asked specifically about back pain. Finally, a lack of immobilization, more specifically the second operation, constituted a fault, but only in conjunction with the first fault of not making further inquiries.

The Court is not clearly satisfied that there was an error in the Court of Appeal’s findings of fact as to causation. The Court of Appeal concluded that notwithstanding the faults committed by the respondent, the accident was the legal cause of the injury suffered by the appellant resulting in paraparesis. This conclusion was amply supportable by the evidence. The Court of Appeal ultimately gave its sanction to the respondent’s experts’ theory that it was at the point and moment of impact that there occurred a medullary contusion, which set in motion the chain of physiological phenomena which would lead to the subsequent paraparesis. The Court of Appeal committed no error in its interpretation of the evidence and its conclusion that given the initial harm caused by the accident, no causation can be attributed to the respondent’s faults. The initial harm of the accident simply outweighed any kind of effect the faulty treatment might have had, to the point where it cannot be said on a probabilistic basis that the faulty treatment had any causal effect. The chances of recuperation in this case were not significant enough on a balance of probabilities to establish that the faulty treatment caused the harm suffered.

The Court of Appeal made no error in its decision not to apply presumptions of causation in favour of the appellant. The refusal to make presumptions is as much an evidentiary decision as is any other acceptance or non-acceptance of methods of proof. There was evidentiary material for the appellant to make his case. Article 2849 C.C.Q. provides that only serious, precise, and concordant presumptions are to be taken into consideration. These criteria were not fulfilled in this case. The evidence pointed in different and sometimes opposite directions. The Court of Appeal appropriately said that it is insufficient to show that the defendant created a risk of harm and that the harm subsequently occurred within the ambit of the risk created.

The rule of imposing solidary liability where a wrong causing injury has been jointly committed should not be applied in favour of the appellant. It should only be applied in cases where there is a true impossibility to determine the author of the delict, which impossibility did not exist here.

Cases Cited

Applied: Canada (Director of Investigation and Research) v. Southam Inc., 1997 CanLII 385 (SCC), [1997] 1 S.C.R. 748; Laferrière v. Lawson, 1991 CanLII 87 (SCC), [1991] 1 S.C.R. 541; approved: Stéfanik v. Hôpital Hôtel-Dieu de Lévis, 1997 CanLII 8479 (QC CS), [1997] R.J.Q. 1332; distinguished: Lapointe v. Hôpital Le Gardeur, 1992 CanLII 119 (SCC), [1992] 1 S.C.R. 351; considered: Dorval v. Bouvier, 1968 CanLII 3 (CSC), [1968] S.C.R. 288; Beaudoin-Daigneault v. Richard, 1984 CanLII 15 (SCC), [1984] 1 S.C.R. 2; Schwartz v. Canada, 1996 CanLII 217 (SCC), [1996] 1 S.C.R. 254; Morin v. Blais, 1975 CanLII 3 (SCC), [1977] 1 S.C.R. 570; referred to: P. (D.) v. S. (C.), 1993 CanLII 35 (SCC), [1993] 4 S.C.R. 141; Moge v. Moge, 1992 CanLII 25 (SCC), [1992] 3 S.C.R. 813; R. v. Nova Scotia Pharmaceutical Society, 1992 CanLII 72 (SCC), [1992] 2 S.C.R. 606; Quebec (Public Curator) v. Syndicat national des employés de l’hôpital St-Ferdinand, 1996 CanLII 172 (SCC), [1996] 3 S.C.R. 211; Stein v. The Ship “Kathy K”, 1975 CanLII 146 (SCC), [1976] 2 S.C.R. 802; Demers v. Montreal Steam Laundry Co. (1897), 1897 CanLII 57 (SCC), 27 S.C.R. 537; Pelletier v. Shykofsky, 1957 CanLII 65 (SCC), [1957] S.C.R. 635; Scotsburn Co-operative Services Ltd. v. W. T. Goodwin Ltd., 1985 CanLII 57 (SCC), [1985] 1 S.C.R. 54; Cie de volailles Maxi Ltée v. Empire Cold Storage Co., [1995] Q.J. No. 731 (QL); Cooke v. Suite, 1995 CanLII 4836 (QC CA), [1995] R.J.Q. 2765; Toneguzzo-Norvell (Guardian ad litem of) v. Burnaby Hospital, 1994 CanLII 106 (SCC), [1994] 1 S.C.R. 114; Ontario (Attorney General) v. Bear Island Foundation, 1991 CanLII 75 (SCC), [1991] 2 S.C.R. 570; Hébert v. Lamothe, 1973 CanLII 173 (SCC), [1974] S.C.R. 1181; Bonenfant v. O.T.J. de la Rédemption, [1994] R.R.A. 225; Snell v. Farrell, 1990 CanLII 70 (SCC), [1990] 2 S.C.R. 311; Labelle v. Charette, [1960] Que Q.B. 770; Massignani v. Veilleux, [1987] R.R.A. 541; Cook v. Lewis, 1951 CanLII 26 (SCC), [1951] S.C.R. 830.

Statutes and Regulations Cited

An Act respecting the implementation of the reform of the Civil Code, S.Q. 1992, c. 57, ss. 9, 85.

Civil Code of Lower Canada, arts. 1053, 1238, 1242.

Civil Code of Québec, S.Q. 1991, c. 64, arts. 1480, 2846, 2847, 2849.

Authors Cited

Baudouin, Jean-Louis, et Patrice Deslauriers. La responsabilité civile, 5e éd. Cowansville, Qué.: Yvon Blais, 1998.

Baudouin, Jean-Louis, et Yvon Renaud. Code civil du Québec annoté, t. 2, 3e éd. Montréal: Wilson & Lafleur, 2000.

Gibbens, R. D. “Appellate Review of Findings of Fact” (1991-92), 13 Adv. Q. 445.

Jutras, Daniel. “Expertise scientifique et causalité”. Dans Congrès annuel du Barreau du Québec (1992). Montréal: Service de la formation permanente, Barreau du Québec, 1992, 897.

Karim, Vincent. Commentaires sur les obligations, vol. 1. Cowansville, Qué.: Yvon Blais, 1997.

Kerans, Roger P. Standards of Review Employed by Appellate Courts. Edmonton: Juriliber, 1994.

Mayrand, Albert. “L’énigme des fautes simultanées” (1958), 18 R. du B. 1.

Québec. Ministère de la Justice. Commentaires du ministre de la Justice: Le Code civil du Québec: Un mouvement de société, t. II. Québec: Publications du Québec, 1993.

Royer, Jean-Claude. La preuve civile, 2e éd. Cowansville, Qué.: Yvon Blais, 1995.

Woods, Thomas S. “Overturning Findings of Fact on Appeal: A Justifiably Narrow Jurisdiction” (1998), 56 The Advocate 61.

APPEAL from a judgment of the Quebec Court of Appeal, 1999 CanLII 13888 (QC CA), [1999] R.J.Q. 1658, [1999] Q.J. No. 2584 (QL), affirming a decision of the Superior Court, [1998] Q.J. No. 234 (QL). Appeal dismissed.

Frédéric St-Jean and Benoît Mailloux, for the appellant.

Gérald R. Tremblay, Q.C., Louis Terriault and David E. Platts, for the respondent.

English version of the judgment of the Court delivered by

Gonthier J. –

I. Introduction

1 This tragic case raises issues in the areas of medical liability under Quebec civil law and appellate review of lower court decisions. The appellant, who was the victim of a serious car accident, appeals against a decision of the Quebec Court of Appeal dismissing his appeal of Morin J.’s decision in the Superior Court, which rejected his action in civil responsibility arising out of the treatment of his injuries by the respondent orthopaedic surgeon as well as another defendant physician against whom the appellant did not pursue his action in this Court. The general issue raised by this appeal is whether the Court of Appeal was justified in intervening in the Superior Court’s decision, and likewise whether this Court has reason to intervene in the Court of Appeal’s decision. An incidental question is whether this case warrants an application in favour of the appellant of a presumption as to causation between the respondent’s alleged fault and the harm suffered by the appellant. For the reasons that follow, I have concluded that the Court of Appeal was entitled to intervene in the Superior Court’s decision and find fault but nevertheless confirm the latter’s ruling that there was no causation. This Court is justified in modifying certain findings of fault made by the Court of Appeal, all the while allowing the finding of fault to stand. The latter court’s conclusion that there was no causation is affirmed. On the incidental question, there is no reason to apply a presumption in favour of the appellant since, given the divergent facts and expert opinion on the matter, there were no serious, precise, and concordant presumptions to be made in this case.

II. Facts

2 In the evening of August 11, 1986, the appellant was hit by an automobile travelling at 90 km/h an hour while he was hitchhiking along highway 20 in Saint-Jean-Chrysostome, in the Quebec City area. He was transported by ambulance to Hôtel-Dieu de Lévis Hospital where he came under the care of the attending emergency doctor, Dr. Couture. The doctor found that the appellant was conscious but agitated, had open fractures in both legs and abrasions on the abdomen, and was bleeding from the head.

3 The appellant was hovering between life and death upon arrival at the hospital. He was losing a lot of blood from the lower limbs and his pulse was very rapid. The appellant suffered from partial amnesia so his memory of events between noon on August 11 and the evening of August 18 is confused at best. Dr. Couture treated him for approximately three hours. The doctor asked for x-rays to be done on the lungs, abdomen, thorax, and the dorsolumbar column, indicating in the x-ray requisition that this was a case of a “polytrauma”. These x-rays showed nothing abnormal except that Dr. Couture suspected a fracture in the spine at the T7 vertebra. He made a note of this as well as the appellant’s complaint of back pain which he testified meant general back pain.

4 The appellant was examined by the respondent Dr. Mercier, an orthopaedic surgeon, later that evening. The respondent was aware of the suspected fracture at T7 and noted it as follows: [translation] “Fx T7? Will see with radiologist”. He nonetheless concluded that the fracture was stable or that it was actually only a benign anomaly, thus not impeding his ability to operate. The operation was extremely urgent as it was necessary to treat the open wounds in the legs.

5 The appellant was brought to the operating room that evening to treat the leg fractures. The respondent noted that the procedure was well tolerated. There are two notes describing the appellant’s condition written by the respondent that evening. The first indicates that the neurovascular exam was satisfactory. The respondent said at trial that this exam was basically a vascular one, despite the terminology used. The second note, which is undated but considered to have been written at the same time, reads: [translation] “From a neurological point of view, drowsy and agitated at times. No neurovascular deficit but difficult to assess on lower limbs given the open fractures. No dorsolumbar pain.” The respondent said at trial that what he meant by the latter observation was that there was no specific back pain.

6 In the afternoon of August 12, a nurse entered a note that there was good warmth and colour to the two feet as well as good movement of the toes of both feet. She observed that both feet, but especially the left one, were edematous.

7 On August 13, the respondent decided that the appellant’s condition had sufficiently improved to allow another operation on the leg the next day. Unlike the first operation, this one was not urgent though it was desirable that it be performed promptly so as to avoid prolonging immobilization.

8 The second operation was carried out on August 14. A nurse’s note states that the appellant was carried to the operating room in a stretcher, contrary to the usual practice of taking the patient to the operating room in his own bed. The appellant had earlier in the day complained of pains in the back and abdomen, as well as being thirsty and hot. The operation took place on an orthopaedic table and it went well.

9 In the evening of August 14, a nurse entered a note that the appellant could feel physical contact to his legs but that he was unable to move his legs and toes. The same observation was made in the course of the next few days.

10 On August 15, the respondent attributed this condition to sciatica on the right side consequent to the surgery. The respondent noted on August 18 that there was still no recuperation of the sciatic nerve. A similar observation was made on August 21, the day a third operation was carried out to modify a traction device attached to the leg. Since the appellant was able to feel his toes when they were touched, the respondent decided that there was nonetheless an amelioration in the sciatica.

11 The respondent left for vacation on August 22 and was replaced by Dr. Nolin, who was given a treatment plan that called for progressive mobilization of the appellant. The appellant started moving around in a wheelchair and was able to move around in his bed with the aid of supports.

12 The appellant was discharged from the hospital on August 27 and was scheduled to meet with the respondent in the out-patient clinic on September 12. It was during this visit, upon removing the plaster covering the right leg, that the respondent witnessed spasms to his great surprise, a symptom of spastic paraplegia. He consulted a neurologist, Dr. Bergeron, who confirmed a condition of spastic paraplegia arising from a medullary contusion.

13 No x-ray of the back was taken on September 12. The appellant returned to school while waiting for more thorough tests, which were fixed for September 19. The x-rays and myelogram revealed a fracture at the T8 vertebra with a subluxation (partial dislocation) of 9 millimetres in relation to T9. A more detailed review of the x-rays taken on August 11 revealed that the fracture at T8 was visible but that the subluxation was only 3 millimetres. Tomography tests were also done on September 22.

14 The respondent ordered the immobilization of the appellant and proposed surgery in order to decompress the spinal cord. On September 24, the appellant requested to be transferred to the Enfant-Jésus Hospital, where the treating neurosurgeon decided not to operate.

15 The appellant was diagnosed with paraparesis, a type of paralysis where the motor activity of the lower limbs is undermined but sensitivity is still retained. He underwent physiotherapy and continued with his education, obtaining a law degree and being called to the Quebec Bar.

16 The appellant applied for and received compensation from the Société de l’assurance automobile du Québec (“SAAQ”). The SAAQ informed him in July 1991 that it would no longer grant him full compensation as it found no causal relationship between the paraparesis and the automobile accident of August 11, 1986. However, after the Superior Court’s ruling finding the car accident to be the cause of the appellant’s injuries, the SAAQ reversed its decision.

17 The appellant’s action against the respondent and Dr. Couture, the emergency-room doctor, went to trial on February 10, 1997 before Rioux J. The appellant’s action against Hôtel-Dieu de Lévis Hospital was settled out of court on March 23, 1994. Five medical experts appeared for each side. After the case had been heard and while Rioux J. was deliberating, he fell ill and so the case was transferred to Morin J., to be decided on the basis of the evidence entered at trial.

III. Relevant Statutory Provisions

18 An Act respecting the implementation of the reform of the Civil Code, S.Q. 1992, c. 57

9. Proceedings pending continue to be governed by the former legislation.

. . . A further exception is made for all matters concerning proof and procedure in such proceedings.

85. The conditions of civil liability are governed by the legislation in force at the time of the fault or act which causes the injury.

Civil Code of Lower Canada

1053. Every person capable of discerning right from wrong is responsible for the damage caused by his fault to another, whether by positive act, imprudence, neglect or want of skill.

1054. Presumptions are either established by law or arise from facts which are left to the discretion of the courts.

1055. Presumptions not established by law are left to the discretion and judgment of the court.

Civil Code of Québec, S.Q. 1991, c. 64

2846. A presumption is an inference established by law or the court from a known fact to an unknown fact.

2847. . . .

A presumption concerning presumed facts is simple and may be rebutted by proof to the contrary; a presumption concerning deemed facts is absolute and irrebuttable.

2849. Presumptions which are not established by law are left to the discretion of the court which shall take only serious, precise and concordant presumptions into consideration.

IV. Judgments Below

A. Quebec Superior Court, [1998] Q.J. No. 234 (QL)

19 Morin J. found that Dr. Couture had committed no fault in the manner that he treated the appellant and accordingly rejected the action against him.

20 Morin J. also found the respondent Dr. Mercier to be without fault. He began with the suspected fracture at T7 and concluded that either it was stable or it was not a fracture at all. This was the assessment of all of the expert witnesses as well as the respondent himself. In those circumstances, there can be no reproach for not discussing T7 with the radiologist since it would not have led to anything and there were no injuries resulting from this supposed fracture.

21 The trial judge found that reviewing the x-rays with the radiologist would not have shed any light on the T8-T9 fracture-luxation since the radiologist himself testified that he could not see it on the x-rays taken on August 11, 1986. Similarly, the respondent could not be blamed for not seeing it.

22 On the question of back pain, Morin J. accepted the respondent’s testimony that the reason he wrote [translation] “no dorsolumbar pain” was because the appellant indicated no precise back pain in response to questions posed by the respondent on that subject. Since there was no indication to suspect a fracture of the spinal column, the respondent had no reason to immobilize the patient to stabilize the spine.

23 The respondent diagnosed the appellant with a sciatic problem when, in retrospect, it was a case of paraparesis stemming from a medullary (spinal) contusion. Morin J. listed some factors explaining why this could not be characterized as a fault. By August 15, when the respondent made the diagnosis of sciatica, he had the radiologist’s report which made no mention of the fracture-luxation at T8-T9. The appellant did not complain to the respondent of any specific back pain. Even though the nurses took note of back pain, they testified that this meant that it was of a generalized nature. The appellant himself testified that from what he remembers of that time, he had back pain but of a tolerable nature. Morin J. found that it was not unreasonable to think that the appellant did not consider it important to mention it to the respondent. Even though the appellant’s friend, Jocelyn Richard, did testify that the former complained to him of back pain, there was no indication that the respondent was witness to this. Moreover, the friend did not mention the complaints of back pain to any medical personnel. Richard also claims to have frequently seen the appellant with priapism (which would have been indicative of medullary trauma) but no nurse ever noticed this, and they testified that they would have taken note of such a state had they seen it. There is no evidence that the respondent knew of this condition. Finally, the state of the appellant’s lower limbs made an adequate neurological exam very difficult.

24 For all of these reasons, Morin J. concluded that the respondent could not be reproached for not having diagnosed a medullary contusion rather than a sciatic condition. This conclusion is reinforced by the fact that many experts testified that the nature of the injuries suffered by the appellant was very unusual, which made diagnosis even more difficult.

25 Even though Morin J. found there to be no fault and therefore no civil liability, he canvassed the arguments regarding the causal link between the purported fault and the injury. He indicated that the respondent’s experts’ theory seemed more probable, but that even if it was not, the court found itself confronted by two opposing medical theories regarding causation among which it could not choose one against the other. He consequently found that the appellant had not discharged his burden to show causation and thus there was a second reason to dismiss the action.

B. Quebec Court of Appeal, [1999 CanLII 13888 (QC CA)](https://www.canlii.org/en/qc/qcca/doc/1999/1999canlii13888/1999canlii13888.html), [1999] R.J.Q. 1658

26 The Court of Appeal (Gendreau, Proulx and Rousseau-Houle JJ.A.) concluded that Morin J. had asked himself the wrong question in determining whether there was a fault. Morin J. had asked whether the respondent had committed a fault by not diagnosing the neurological deficit and the unstable fracture at the T8 and T9 vertebrae. The true question to be asked according to the Court of Appeal was whether the respondent [translation] “had abided by the rules of art in the treatment of a seriously injured patient who has the potential, if not the probability, of having a back fracture” (p. 1662). This error on the part of the trial judge led him to apply the wrong standard in evaluating the conduct of the respondent and therefore permitted the Court of Appeal to make its own assessment as to fault.

27 The Court of Appeal found that in not following up on the suspected fracture at the T7 vertebra, in not doing more complete neurological exams, in not consulting the nurses’ notes alerting to the possibility of a neurological deficit, and in not ordering immobilization of the spinal column, the respondent did not behave like a prudent and diligent orthopaedic surgeon. The respondent had thus committed a fault.

28 On the issue of causation, the Court of Appeal corrected what it considered to be a mistaken approach by Morin J. The trial judge had refused to take sides in the experts’ dispute as to the cause of the neurological deficit, saying that it was not his role to judge between two opposed medical theories. The Court of Appeal found that such an approach to causation would unduly favour defendants in all cases where the establishment of the causal link depends on contradictory expert scientific evidence. Even if the scientific conclusions are not absolute or not incontestable, a judge must convert scientific conclusions to a legal conclusion since scientific causation is not the same as legal causation. The latter is established on the balance of probabilities based on all of the evidence.

29 The Court of Appeal rejected the appellant’s claim for a reversal of the burden of proof. It said that there was enough evidence to maintain the traditional rule of the burden of proof. It specified that this rule should be set aside in only the most exceptional cases, such as the one where two defendants negligently fire in the same direction and at the same time, their delictual conduct eliminating any means of proof available to the plaintiff.

30 On the question of causation, the Court of Appeal concluded that the medullary injury was already complete on the evening of the accident and that the manifestations of that injury presented themselves gradually. It took note that even the appellant’s principal experts did not exclude the hypothesis that a subluxation already occurred in the evening of August 11 and that this entailed a certain medullary contusion. Their only contention was that only external mobilization could account for the subluxation of T8. The Court of Appeal accepted the respondent’s experts’ evidence regarding the ligaments’ rupture, compression, and MRI results to conclude that the medullary lesion was probably complete at the time of the accident. It acknowledged that an early immobilization resulting in a better recuperation was more than a simple possibility, but that this was not significant enough to give rise to civil liability given the extent of the initial harm caused.

V. Issues

31 The questions raised by this appeal are the following:

(1) What is the standard of review of a lower court decision?

(a) Did the Court of Appeal properly intervene in the Superior Court decision?

(b) Is there a basis for this Court to intervene in the Court of Appeal decision?

(2) Is there a reason to apply a presumption of causation in favour of the appellant?

VI. Analysis

A. Transitional Law

32 The events in this case all took place before the coming into force of the new Civil Code of Québec in 1994. Sections 9 and 85 of An Act respecting the implementation of the reform of the Civil Code make it clear that the substantive law governing this case is that of the Civil Code of Lower Canada, more specifically art. 1053 of that Code. While questions of evidence such as presumptions are governed by the new code, there has been no break in the approach towards presumptions from the previous regime. The relevant provisions in the Civil Code of Lower Canada did not provide a definition and did not specify the kind of presumptions to be taken into account. However, the case law has always made those specifications and the new code has simply codified them in art. 2849 (see Ministère de la Justice, Commentaires du ministre de la Justice (1993), vol. II, at p. 1783; J.-L. Baudouin and Y. Renaud, Code civil du Québec annoté (3rd ed. 2000), vol. 2, at pp. 3355-58).

B. Reviewing Lower Court Decisions

33 Questions about the correct legal test are questions of law (Canada (Director of Investigation and Research) v. Southam Inc., 1997 CanLII 385 (SCC), [1997] 1 S.C.R. 748, at para. 35). In the case at bar, the question of the correct inquiry or legal test in determining a delictual fault is a question of law.

34 An appellate court must review a lower court’s judgment on questions of law to determine their correctness (P. (D.) v. S. (C.), 1993 CanLII 35 (SCC), [1993] 4 S.C.R. 141, at p. 189 (a contrario); Moge v. Moge, 1992 CanLII 25 (SCC), [1992] 3 S.C.R. 813, at p. 833; R. v. Nova Scotia Pharmaceutical Society, 1992 CanLII 72 (SCC), [1992] 2 S.C.R. 606, at p. 647; Lapointe v. Hôpital Le Gardeur, 1992 CanLII 119 (SCC), [1992] 1 S.C.R. 351, at p. 373 (a contrario); and R. P. Kerans, Standards of Review Employed by Appellate Courts (1994), at p. 90).

35 Questions of fact are “questions about what actually took place between the parties” (Southam, supra, at para. 35). The question of whether the respondent did this or that is a question of fact, as is the question of whether the accident or any lack of immobilization caused the appellant’s state of paraparesis or hindered his recovery.

36 It is established law that an appellate court such as a provincial court of appeal should not interfere with a trial judge’s findings of fact, absent a palpable and overriding error in the understanding of the evidence (see Quebec (Public Curator) v. Syndicat national des employés de l’hôpital St-Ferdinand, 1996 CanLII 172 (SCC), [1996] 3 S.C.R. 211, at para. 46, where L’Heureux-Dubé J. cites the relevant cases going back to Stein v. The Ship “Kathy K”, 1975 CanLII 146 (SCC), [1976] 2 S.C.R. 802, and Dorval v. Bouvier, 1968 CanLII 3 (CSC), [1968] S.C.R. 288). The rule is partly grounded in the recognition of the fact that the trial judge had the opportunity to observe the witnesses and hear the testimonies first-hand, and is therefore better able to assess the facts.

37 The principle of non-intervention on questions of fact is also applicable to a second appellate court such as this Court vis-à-vis a first appellate court (Demers v. Montreal Steam Laundry Co. (1897), 1897 CanLII 57 (SCC), 27 S.C.R. 537, at pp. 538-39; Pelletier v. Shykofsky, 1957 CanLII 65 (SCC), [1957] S.C.R. 635, at p. 638; Dorval, supra, at p. 294; Scotsburn Co-operative Services Ltd. v. W. T. Goodwin Ltd., 1985 CanLII 57 (SCC), [1985] 1 S.C.R. 54, at pp. 63-64).

38 In Demers, supra, at pp. 538-39, Taschereau J. said:

For it is settled law upon which we have often acted here, that where a judgment upon facts has been rendered by a court of first instance, and a first court of appeal has reversed that judgment, a second court of appeal should interfere with the judgment on the first appeal, only if clearly satisfied that it is erroneous. . . .

This statement was endorsed in Pelletier, supra, and both of these cases were discussed in Dorval, supra, where Fauteux J. examined the principle that must govern this Court’s review of an appeal court’s judgment on questions of fact once the latter has examined the facts and arrived at a conclusion after justifiably intervening in the trial judgment. He said, at p. 294:

[translation] Thus, in order to intervene in this case, it would be necessary to be clearly satisfied that the judgment of the Court of Appeal is erroneous, either with respect to the reason for its intervention or with respect to its assessment of the evidence in the record.

Fauteux J. said that he was not “clearly satisfied” that the Court of Appeal erred and therefore dismissed the appeal.

39 Similarly in Scotsburn, supra, this Court was asked to review the decision of a court of appeal where the latter justifiably intervened in the trial judgment. Dickson J. (as he then was) applied the rule in Dorval, supra, and held, at p. 64, that he was not “clearly satisfied” that there was an error by the Court of Appeal in its interpretation of the evidence.

40 In Beaudoin-Daigneault v. Richard, 1984 CanLII 15 (SCC), [1984] 1 S.C.R. 2, even though this Court did disturb the findings of fact made by the Court of Appeal, it did so only after affirming the validity of the principle expressed in Dorval, supra, and finding that the Court of Appeal erroneously intervened in the trial judgment. This Court intervened for the purpose of restoring the trial decision, which simply confirms the general principle of appellate non-intervention.

41 A peculiarity of the case at bar is that the first trial judge, Rioux J., fell ill after hearing all the evidence and was consequently substituted by Morin J. The latter made his decision based on the evidence entered at trial. There is seemingly less of a basis for deference where the trial judge did exactly what the Court of Appeal and, by extension, this Court do to arrive at a decision. However, the Quebec Court of Appeal has upheld the principle of deference in cases similar to the one at bar where the judge of first instance is not the one who presided over the trial: Cie de volailles Maxi Ltée v. Empire Cold Storage Co., [1995] Q.J. No. 731 (QL); Cooke v. Suite, 1995 CanLII 4836 (QC CA), [1995] R.J.Q. 2765.

42 Even where observation of the witnesses and hearing the testimonies are not of any particular importance for the disposition of the case, the principle of non-intervention is still important since the trial judge is entrusted with the task of making a decision based on the facts, and is presumed to have diligently discharged that duty. In Toneguzzo-Norvell (Guardian ad litem of) v. Burnaby Hospital, 1994 CanLII 106 (SCC), [1994] 1 S.C.R. 114, the British Columbia Court of Appeal justified its intervention on the basis that it was in as good a position to draw inferences from the expert evidence as was the trial judge since there was no issue of credibility. In this Court, McLachlin J. (as she then was), at p. 122, agreed that the principle of non-intervention does not apply with the same force where the credibility of witnesses is not in issue. However, she did observe that the weight to be assigned to the various pieces of evidence under our trial system is essentially the province of the trier of fact. More broadly it is correct to say that where a trial judge has been given the mandate to make a decision based on questions of fact, there is still a need for appellate judicial restraint given that the assessment of facts is the province of the trier of fact. Also, the autonomy and integrity of the trial process as well as resource allocation militate in favour of the finality of judgments (see Schwartz v. Canada, 1996 CanLII 217 (SCC), [1996] 1 S.C.R. 254, at para. 32; T. S. Woods, “Overturning Findings of Fact on Appeal: A Justifiably Narrow Jurisdiction” (1998), 56 The Advocate 61, at pp. 63 and 65-67; Kerans, supra, at pp. 10-16; and R. D. Gibbens, “Appellate Review of Findings of Fact” (1991-92), 13 Adv. Q. 445, at pp. 445-47).

43 Schwartz, supra, brought an element of refinement to the established case law that a second appellate court should only disturb findings of fact if it is clearly satisfied that an error has occurred in the first appellate judgment. La Forest J. said, at para. 37:

In my view, nothing justifies a second appellate court in showing that kind of deference to the assessment of the balance of probabilities made by the first appellate court. If the second appellate court agrees that the trial judge made some kind of error that justifies intervention, it should be free to reconsider the evidence and substitute its own findings of fact for that of the first court of appeal’s if disagreement occurs.

44 Schwartz, supra, is a refinement on the previous case law in the sense that where the first appellate court reverses the trial judge on findings of fact related to credibility, this Court should not feel constrained in assessing the evidence on a balance of probabilities since the first appellate court was no better placed in assessing credibility, and this Court is faced with a trial level decision going in one direction and an appellate decision going in the opposite direction. This Court can in that context make its own assessments on questions of fact such as whether something has been established on a balance of probabilities.

45 In contrast, where there are concurrent findings of fact at the lower courts, this Court will be hesitant to intervene and disturb findings of fact. This Court has already said in Ontario (Attorney General) v. Bear Island Foundation, 1991 CanLII 75 (SCC), [1991] 2 S.C.R. 570, at p. 574, that the principle of non-intervention “is all the stronger in the face of concurrent findings of both courts below” (emphasis added).

46 In either case, the second appellate court must be “clearly satisfied that [the judgment of the Court of Appeal] is erroneous” (Dorval, supra, at p. 294). I do not consider La Forest J.’s statement as being a departure from the previous case law on the role of second appellate courts on questions of fact. He cited some of the same cases as I have here (Beaudoin-Daigneault, Demers, and Dorval), and did not express any kind of opposition to that case law, not to mention any desire to overturn those cases. Despite the freedom with which this Court can reconsider the evidence and “substitute its own findings of fact for that of the first court of appeal’s if disagreement occurs” (Schwartz, supra, at para. 37), that disagreement must nonetheless stem from a clear satisfaction that an error has occurred in the first appellate court’s assessment of the facts.

47 In the case at bar, the Court of Appeal did not really disturb the Superior Court’s findings on causation since it substantially agreed with the ultimate conclusion of Morin J. The Court of Appeal did intervene on the causation issue since Morin J. had framed the legal analysis of causation in an erroneous manner, which I discuss in greater detail below. But after reassessing the evidence, the Court of Appeal came to the same conclusion as Morin J. that no causation was proven on the balance of probabilities. Thus, on the issue of causation, there is substantive agreement between the Superior Court and the Court of Appeal, which takes us away from the pattern described in Schwartz, supra, and brings us closer to the scenario in Bear Island, supra.

48 A question “about whether the facts satisfy the legal tests” is one of mixed law and fact. Stated differently, “whether the defendant satisfied the appropriate standard of care is a question of mixed law and fact” (Southam, at para. 35).

49 Generally, such a question, once the facts have been established without overriding and palpable error, is to be reviewed on a standard of correctness since the standard of care is normative and is a question of law within the normal purview of both the trial and appellate courts. Such is the standard for medical negligence. There is no issue of expertise of a specialized tribunal in a particular field which may go to the determination of facts and be pertinent to defining an appropriate standard and thereby call for some measure of deference by a court of general appeal (Southam, supra, at para. 45; and Nova Scotia Pharmaceutical Society, supra, at p. 647).

50 The inquiry to be made here is the traditional one. Was the Court of Appeal correct in intervening in the Superior Court’s decision because there was an error of law, an error of mixed law and fact, or an overriding and palpable error of fact? Do the appropriate bases present themselves for this Court to intervene in the Court of Appeal’s decision?

C. Errors of Law in the Reasons of the Superior Court

51 The Court of Appeal found two errors of law in the Superior Court judgment. First, it concluded that Morin J. made a mistake in the analysis of fault. Second, it found that Morin J. made a mistake in the analysis of causation. I shall deal with these in turn.

52 The Court of Appeal said that Morin J. had asked the wrong question in determining whether a fault had been committed and thereby applied the wrong standard in evaluating the conduct of the respondent. According to the Court of Appeal, instead of asking whether the respondent was at fault in failing to diagnose or badly diagnosing the medullary contusion and the fracture-luxation at T8-T9, Morin J. should have asked whether the respondent acted according to the professional rules of art in the treatment of this patient. The Court of Appeal essentially said that the question to be answered should be cast in general terms according to the appropriate standard of conduct in the analysis of fault under Quebec civil law, rather than incorporating the specific facts of the case into the question.

53 The correct inquiry to be made in assessing whether a professional committed a fault is indeed to ask whether the defendant behaved as would a reasonably prudent and diligent fellow professional in the same circumstances (J.-L. Baudouin and P. Deslauriers, La responsabilité civile (5th ed. 1998), at p. 850). To ask, as the principal question in the general inquiry, whether a specific positive act or an instance of omission constitutes a fault is to collapse the inquiry and may confuse the issue. What must be asked is whether that act or omission would be acceptable behaviour for a reasonably prudent and diligent professional in the same circumstances. The erroneous approach runs the risk of focussing on the result rather than the means. Professionals have an obligation of means, not an obligation of result.

54 In deciding that the wrong question had been asked, the Court of Appeal was concerned that a proper inquiry was not made and the wrong standard of fault was applied by the trial judge. This concern entitled the Court of Appeal to intervene. While it is not entirely clear that Morin J. was unaware of the appropriate inquiry to be made, his conclusion, at para. 78, that [translation] “Dr. Mercier cannot be faulted for failing to diagnose the medullary contusion his patient suffered . . . and having thought instead that he was suffering from sciatica on the right side” raises some doubt as to whether he was sufficiently focussed on a delictual analysis of means rather than one of result.

55 The Court of Appeal also faulted the decision by the trial judge for not taking a position on the scientific debate as to causation. Morin J. wrote, at para. 90:

[translation] [T]he court concludes that it is not up to it to decide between the opposing medical theories it is finally presented with concerning the causal link between the defendants’ actions and the paraplegia the plaintiff is suffering from.

The trial judge must assess whether the evidence supports, on a balance of probabilities, a finding of causation. If this standard of proof is not met, the plaintiff has simply been unsuccessful in discharging his burden of proof. It is not enough to say that there are opposing medical theories on causation and that it is not up to the court to decide between them.

56 A trial judge must reach a legal conclusion based on the scientific evidence and other evidence presented. Not to come to a definitive conclusion on a balance of probabilities amounts to an undue advantage granted to the defendant, who might simply need to come up with a plausible but contrary scientific theory in order to negative the plaintiff’s claim. Such an approach is tantamount to an alteration of the standard of proof since the trial judge is no longer looking at which scientific theory is most probable. It is an error of law in the analysis of causation for a trial judge to conclude that he or she does not have the authority to make a final legal determination in the face of competing theories.

57 There is room for the principle of judicial neutrality in the analysis of fault. It is appropriate for the courts not to take a position when there are competing but recognized medical theories on what is competent professional practice or an appropriate diagnosis. In Lapointe, supra, at p. 363, L’Heureux-Dubé J. said that “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” (emphasis added). However, it is not appropriate to be similarly neutral on matters of causation and thereby refrain from determining whether causation has been established on the legal standard of balance of probabilities.

58 It was open to the Court of Appeal to be concerned that the trial judge had erred in this manner and accordingly to undertake its own analysis of causation. There is no reason for this Court to question that intervention. At issue are the Court of Appeal’s findings as to fault and causation. I will now turn to these.

D. Fault

59 The Court of Appeal identified four principal faults committed by the respondent: not verifying with the radiologist the fracture at T7 and not ensuring that there were no other fractures in the spinal column; not asking for more complete neurological tests; not taking cognizance of the nurses’ notes indicating a neurological deficit; and not ordering the immobilization of the spinal column. Because of all these failings, it was possible to conclude that the respondent did not behave as a prudent and diligent orthopaedist would in similar circumstances.

60 As I discussed above, findings of fault in the law of delict are questions of mixed law and fact. They are the application of a legal concept onto a set of facts. The standard to be applied in this instance of a question of mixed law and fact is one of correctness. It is open to this Court to vary the findings of fault of the Court of Appeal if it judges that these were done erroneously.

61 The Court of Appeal, at p. 1664, identified as a fault the omission by the respondent to verify with the radiologist the possible fracture at T7 and to assure himself that there were no other fractures, given the violence of the impact during the accident:

[translation] Dr. Mercier’s presumption concerning the relative importance of the fracture at T7 did not exempt him from examining, after the first emergency operation, the appellant’s neurological state in order to verify the injury at T7 and to ensure that there were no other spinal column fractures, given the violence of the impact during the accident.

There are two elements to this finding of fault: the suspicion about a fracture at T7 and the violence of the impact. The majority of the expert opinion concluded that there was no fracture at T7 but rather a benign anomaly. In any case, at the time of diagnosis by Dr. Couture and the respondent, there was only a suspicion of a fracture at T7. The Court of Appeal undoubtedly meant this implicitly as it reproached the respondent not for being ignorant of the fracture at T7 (there was none to be discovered), but rather, for not inquiring further, so long as there was a remaining doubt, into the possibility of a fracture at T7 or elsewhere, given the extreme violence of the accident.

62 The fault of the respondent lies in making neither further examinations of the spine based on the suspected fracture at T7 nor neurological examinations. Even though it was generally accepted among the experts that there was no fracture at T7, the suspicion of a fracture at one level should have motivated the respondent to ensure that there were no fractures at any other level, and to ensure that there was no other injury in the back. By his own testimony, the respondent was concerned about T7 even after the second operation, and expressed that concern by asking the appellant about back pain, though by that time he could no longer move his toes.

63 There can be no reproach against the respondent for not verifying the fracture at T7 with the radiologist. As far as T7 itself was concerned, the respondent had exercised his professional judgement and decided that whatever fracture there may be at that level, it was stable.

64 However, the suspected fracture at T7 together with the violence of the impact, as well as the consequent concern of the respondent about back pain later on, should have led the respondent to examine more fully the state of the spine once the urgent first operation was done. Further investigation would have permitted him to discover the true fracture at T8. The fault lies in not conducting the requisite neurological tests, and not examining the back properly with the available tests such as more localized x-rays and scans. The respondent admitted at trial that these tests were available on August 11, and so they clearly could have been done after the first operation. Dr. Bouvier for the appellant testified that there is a general need upon admission to the emergency ward to do tests frequently to assess whether there is deterioration or amelioration. Both Dr. Lévesque and Dr. Drouin, also for the appellant, testified that neurological tests were possible despite the limitations posed by the injuries to the lower limbs. While these experts said that the exams could have been done upon admission, which was contested by the respondent’s experts, Dr. Jodoin and Dr. Séguin, they certainly could have been done after the first operation, once the appellant’s injuries in the lower limbs were stabilized. Dr. Charlebois for the respondent did say that he would have inquired into the patient’s ability to withstand further tests, and would have required more specific information before conducting further tests. The respondent’s other experts did not contest the possibility for doing more localized x-rays and scans, but said that there was no indication to the respondent for the need to do them. For the reasons mentioned above, I find no error in the Court of Appeal’s concluding that a reasonably prudent and diligent orthopaedic surgeon would have done them.

65 The Court of Appeal considered it a fault for the respondent not to take cognizance of the nurses’ notes made after the second operation. These notes indicated a neurological deficit as well as back pain. However, it is difficult to see this as a fault given that the respondent personally checked to see how the appellant was recovering after the second operation and asked specifically about back pain. There is no reason to find a fault on this point.

66 The Court of Appeal found the lack of immobilization to be a fault. It did not make clear whether this fault was contingent on the first fault. I cannot see how it could be otherwise. Obviously, if T7 was considered to be stable and there was no reason to suspect any other fracture, there was no reason to immobilize. It is only upon learning or at least being suspicious that there is an unstable fracture that there would be a reason to immobilize. I do consider that a lack of immobilization, more specifically the second operation, could be a fault, but only in conjunction with the first fault of not making further inquiries.

67 As for the particulars of that fault of not immobilizing, the Court of Appeal reproached the respondent for not making an immobilization order. However, the testimony of two nurses was quite explicit to the effect that there is immobilization in cases such as these until an explicit order for mobilization is given. This is corroborated by the fact that the respondent made a mobilization order on August 14, after the second operation.

68 On the other hand, there is documentary evidence in the hospital record stating that the respondent had been turned over several times, was rubbed in the back, and was given a bath in bed between August 12 and 13. The appellant’s expert, Dr. Bouvier, considered this to be an example of wrongful mobilization.

69 There was thus evidence for the Court of Appeal to conclude as it did on this point. There is no reason for this Court to disturb that conclusion.

70 The Court of Appeal found the use of a stretcher in carrying the appellant to the operating room to be a fault. It made this finding on the basis of nurse Couture’s note that said that the patient was moved on a “civière” (stretcher) for the second operation on August 14.

71 Nurse Couture testified at trial that she found it odd that she should have written such a thing since the normal practice was to take orthopaedic patients in their own beds, especially when there is a traction device on the patient, as was the case here. She said that it normally never happens that such a patient is taken on a stretcher. The respondent testified that all patients in traction on whom he has operated are brought to the operating room in their beds. Dr. Jodoin, expert for the respondent, insisted that it is impossible that a stretcher was used instead of bringing the appellant into the operating room in his own bed since he was attached to a traction device. This device with its pulley mechanism is fixed to the bed with the result that a patient is never put on a stretcher when traction is needed. Dr. Jodoin said that he was [translation] “morally certain” that the appellant was not put on a stretcher to be taken to the operating room.

72 It was open to the Court of Appeal to rely on nurse Couture’s written note regarding use of the stretcher, even though the author of that note as well as other medical professionals cast serious doubt about its accuracy.

73 The second operation conducted on August 14 is an instance of a lack of immobilization. The respondent admitted at trial that the second operation was not urgent. Carrying it out without making the appropriate examinations to ensure adequate stability was clearly a fault. There was a need to ensure prior stability through appropriate examinations because the operation, with its attendant curarization resulting in the neutralization of the muscular support around the spine, posed an unacceptable danger to the stability of the spine. The respondent did not behave as a reasonably prudent and diligent orthopaedic surgeon, and thereby committed a fault.

E. Causation

74 The Court of Appeal concluded, at p. 1669, that, notwithstanding the faults committed by the respondent, the accident was the legal cause of the injury suffered by the appellant resulting in paraparesis,

[translation] While an early immobilization and a diminution of the oedema in the spinal cord are more than mere possibilities with regard to a better recovery of the neurological deficit, they are not significant enough, given the magnitude of the initial damage caused by the accident, to hold Dr. Mercier liable in part.

It canvassed the principal points of the expert evidence and reached a conclusion. This conclusion was amply supportable by the evidence. I am not clearly satisfied that there was an error in the Court of Appeal’s findings of fact as to causation that would permit me to intervene in the latter’s decision.

75 The general theory of the appellant’s experts is that the injury took place due to hospital mobilization that related principally to the second operation. The appellant’s toes were moving after the first operation, but were not after the second. It is argued that the paraparesis is due to an absence of adequate immobilization of the appellant. According to these experts, it was the faulty treatment that aggravated the injury caused by the accident. Immobilization would have decompressed the spinal cord. According to Dr. Bouvier, the paraparesis is due to the displacement of the vertebrae caused by the mobilization of the appellant, such as the back rub given between the first and second operations, the turning over of the appellant, the giving of baths in bed, the transfer from bed to stretcher before the second operation, and the second operation itself. Dr. Guimond ascribed greatest probability of injury to the second operation. He specified that the curarization of the second operation had the effect of relaxing the back muscles, which resulted in a lack of support for the spine. It was this instability that caused the vertebral displacement at T8-T9 to go from 3 millimetres ascertained upon arrival to 8 or 9 millimetres discovered a month later. Dr. Guimond acknowledged that some neurological deficit to the lower limbs may have been caused by the accident, but asserted that the deficit was not complete. He attributed 25 percent of the injury to the accident, the remainder to the respondent’s fault of not properly immobilizing the appellant. The appellant’s experts pointed to the spinal compression shown on the myelogram and MRI as proof that the hospital mobilizations caused the harm.

76 The general theory of the respondent’s experts is that the violence of the accident was so great that there was a fracture-luxation at T8-T9 of 8 or 9 millimetres, which was spontaneously reduced by virtue of elastic recoil to 3 millimetres. The vertebrae assumed their normal position because these bone structures are linked not only by ligaments but also muscles that are all around the spinal column. The initial displacement caused a contusion to the spinal cord, setting off a chain of physiological phenomena that caused irreversible damage to the spinal cord. A contusion is an injury produced by an instantaneous blunt force. The spinal column ultimately stabilized at the 8 or 9 millimetre mark which it had reached at the peak of the impact. Dr. Jodoin observed in support of the recoil theory that nothing other than a violent impact making the spinal column stretch could have caused the ligaments to tear. Given the violence of the impact, the subsequent movements in the hospital had a negligible effect on an injury that was definitive and final in its amplitude, leaving no room for recuperation. In support of this conclusion, the respondent’s experts drew a distinction between primary and secondary compression, saying that the tests only show the latter, which indicates that the harm was caused by a contusion rather than a compression.

77 The uncontested facts are that there was a massive shearing between the thoracic vertebrae upon impact, causing fractures to a vertebra and a rib, and causing all but one of the surrounding ligaments to tear. The one ligament that did not tear was attached to a rib that did fracture, such was the violence of the impact. Experts for both the appellant and respondent acknowledged the tearing of the ligaments. Dr. Milette for the appellant stated in his report, at p. 9:

[translation] Several vertebral ligaments were probably partially or completely torn at T8-T9, including the posterior ligamentary complex. Certain costo-vertebral ligaments resisted, and a fracture occurred at the origin of the ninth rib on the left side and probably on the right side.

Dr. Milette also stated numerous times in his testimony that the ligaments were torn: [translation] “all those ligaments were torn”; “that means tearing of many structures which, normally, are very solid”; “the violence causes the tearing, causes the fractures and the torn ligaments . . .”. This assessment was confirmed by both Dr. Jodoin and Dr. Charlebois, the latter saying in his testimony that [translation] “the subsequent displacement shows that there was in fact a rupture of all those ligaments and we see that among other things on the tomography, the scan, that was subsequently carried out . . .”.

78 The Court of Appeal ultimately gave its sanction to the respondent’s experts’ theory that it was at the point and moment of impact that there occurred a medullary contusion, which set in motion the chain of physiological phenomena such as swelling, intramedullary oedema, ischemia by compression, and destruction of nerve cells, all of which would lead to the subsequent paraparesis. There then occurred a recoil, hence explaining the mere 3 millimetre subluxation of T8 upon admission to the hospital, which was universally considered to be within normal range. The later subluxation of 8 to 9 millimetres was simply a return to the state of the spine as it existed immediately upon impact. This happened because the ligaments started healing in their elongated form. The fact that the toes were moving on August 12 but were not moving on August 14 after the second operation was merely a delayed manifestation of the injury that occurred at the time of the accident.

79 The appellant’s experts have no alternative explanation for the torn ligaments. The fact that the ligaments were torn lends much credence to the respondent’s theory that there was a violent extension and recoil action of the spine. Nothing less than this would have caused such tough ligaments to tear. This in turn supports the view that the sudden displacement of the vertebrae caused a contusion to the spinal cord, setting in motion the destructive phenomena that resulted in paraparesis.

80 As part of their general contestation of the respondent’s experts’ theory that all the damage was done upon impact, the appellant’s experts provide a contrary affirmation on three points in support of their own theory that any injury from the accident could have been contained and reversed without much lasting adverse effect. The first and most significant point is the extent of the immobilization in the hospital. The second is the significance to be assigned to compression of the spinal cord as shown in the tests. The third is the significance to be assigned to spinal shock and the ostensible lack of it in this case.

81 The appellant alleges that inadequate immobilization, as evidenced by the back rub, the baths, the turning over, and the second operation and attendant curarization, was the cause of the harm he suffered. The evidence shows that there was no explicit order for the immobilization of the appellant. However, nurse Plante testified that immobilization of patients such as the appellant is done as a matter of course due to the extent of the injuries suffered until an order of mobilization is given. This appears to have occurred given that an express order for the progressive mobilization of the patient was made after the second operation. Nurse Couture testified that patients in intensive care are moved by at least two people, and care is taken to use a pillow between the legs to guard against torsion. Hospital records indicate that this was the care taken even after the mobilization order was given. There is no reason to suspect that it was otherwise before the order was given. The respondent testified that when the appellant was transferred onto the operating table, he was moved en bloc with at least four people. Thus, even if we accept that the appellant was first moved onto a “civière” and then onto the operating table, it was done with minimal possibility of torsion.

82 As for the second operation itself, an orthopaedic operating table was used which was described as having a post in the centre that immobilized the appellant’s back during the operation. While curarization would have weakened the muscular support around the spine, the care taken in transporting the patient to and from the operation and the immobilization of his back on the operating table left little or no possibility of its having any harmful effect.

83 The respondent testified that after the operation, the appellant would have been moved in the same way as he was prior to the operation, namely en bloc. While the respondent’s testimony about how the appellant was moved before and after the operation, and how the appellant was placed on the orthopaedic table during the operation, was introduced under reserve of an objection to its admissibility as descriptive of general practice rather than specific recollection, it is of lesser probative value but nevertheless useful in assessing probability. It is properly admissible.

84 The appellant was indeed allowed to move after the second operation. However, even the appellant’s experts limit the cause of the paraparesis to movements up to the second operation and immediately after (i.e. the transfer from the operating table to the “civière” and then back to the bed), but not the progressive mobilization subsequent to the second operation. Dr. Bouvier explicitly admitted that he had no indication that aggravation took place after the second operation.

85 The appellant’s experts argued that it was the hospital movements that caused the subluxation of 8 or 9 millimetres. As Dr. Jodoin for the respondent testified, the relative impact of the accident in contrast to the hospital movements would be a ratio of 100:1. Given that it takes a violent trauma for the spinal ligaments to tear, and given that they were in fact torn in this case, it is hard to believe that the accident caused a mere 3 millimetre displacement but the hospital movements caused an 8 or 9 millimetre displacement. The Court of Appeal was entitled to weigh the evidence and conclude that the harm caused by the accident was so great and definitive in its effect that any lack of immobilization was not significant enough to impinge upon a better recovery. Even though there obviously was some mobilization of the appellant between the first and second operations, this was minimal in light of the general level of immobilization and the precautions taken.

86 A significant part of the evidence, and a major point of contention between the experts as to its diagnostic value, concerned the existence of spinal cord compression as shown by the myelogram and MRI. Compression here means the pathological and sustained vertebral pressure on the spinal cord. It was accepted by both groups of experts that compression can cause paraparesis and that evidence of compression would be an indication that hospital movements had caused damage. The respondent’s experts added the caveat that it is only direct compression that would be such an indication, whereas indirect or secondary compression would mean that there was another cause for the medullary damage, namely contusion to the spinal cord with the consequent damaging physiological phenomena.

87 The appellant’s experts, Dr. Lévesque and Dr. Drouin, pointed to the compression depicted by the tests as an indication that the hospital movements had caused damage. Dr. Laroche and Dr. Séguin, experts for the respondent, distinguished between a direct and an indirect compression of the spinal cord. Dr. Laroche said that while the dural sac that contains the spinal cord was compressed, the spinal cord itself was not. In other words, there was no direct compression, only a secondary compression. Dr. Séguin said that secondary compression was brought on by the oedema, swelling of the spinal cord, and destruction of the nerve cells.

88 There is one slight error committed by the Court of Appeal in understanding the expert evidence, but this does not bear in any way on the result that it ultimately arrived at. The Court of Appeal says that Dr. Séguin stated that if there was a medullary compression caused by the subluxation of the already fractured and unstable vertebrae, this compression would have been visible on the myelogram even 5 to 6 weeks after the hospitalization.

89 What Dr. Séguin actually said was that if there was a direct compression caused by the accident, there would have been a residual appearance of it in the myelogram. However, what he said is as true about compression caused by the lack of immobilization as it is about compression caused by the accident. In other words, if there had been direct compression of the spinal cord itself, regardless of the cause, it would have appeared on the myelogram.

90 The Court of Appeal was entitled to accord more weight to Dr. Séguin’s interpretation of the myelogram and Dr. Bélanger’s and Dr. Charlebois’s interpretations of the MRI, and come to the conclusion that there was a contusion to the spinal cord at impact but not a compression brought on by subsequent hospital mobilization. Both Dr. Séguin and Dr. Laroche were quite detailed at trial in their explanations of how the dye in the myelogram still managed to flow through the suspected point of compression, thus indicating that there was no direct compression. The contusion to the spine, which was an instantaneous blow to the tissue by the vertebral structure, was the injury that sparked the physiological chain reaction that ultimately resulted in paraparesis.

91 There was much discussion before this Court about the spinal shock syndrome. Neither the Superior Court nor the Court of Appeal discussed this. The appellant’s position is that both lower courts failed to appreciate the significance of the absence of spinal shock in this case.

92 The appellant asserted that spinal shock is a syndrome that is recognized in medicine and that manifests itself in cases where there has been a major trauma to the spinal cord, even when this trauma leads only to partial disability in terms of mobility, and that spinal shock manifests itself by the observable lack of mobility and sensitivity on the part of the patient.

93 The appellant’s expert, Dr. Guimond, stated that, as a rule, serious trauma to the spine is usually accompanied by spinal shock, where the patient is rendered completely immobile for a temporary period after the traumatic incident. The appellant argued that since no deduction of a state of temporary immobility could be made from the hospital record, there was no spinal shock. Since there was no spinal shock, the neurological damage had not been irreversibly caused at the moment of the car accident.

94 Counsel for the respondent argued that the absence of spinal shock was not inconsistent with a severe but partial medullary lesion causing paraparesis. The expert evidence, notably that of Dr. Jodoin and Dr. Séguin which remained uncontradicted in this respect, supports the respondent’s position.

95 The significance, if any, of the absence of spinal shock is a matter of fact which has to be appreciated in the context of all of the evidence. Despite the cogent argument advanced by the appellant on this issue, the evidence relating to spinal shock is inconclusive, and does not justify overlooking all the evidence that points in the other direction.

96 There was ample evidence that all of the damage suffered was caused by the accident and that there was no aggravation or independent damage caused by the faults of the respondent. The profusely bleeding and life-threatening open wounds on the legs, the fractures at the T8 vertebra and the ninth rib, the tearing of all but one of the ligaments attaching the vertebral column to the ribs, the bleeding head wounds and brain concussion, and the partial memory loss all speak to the extreme violence of the appellant being hit by a car travelling at 90 km/h. It is also important to keep in mind that the appellant was an Olympic-level wrestler whose muscular structure was exceptionally strong, as evidenced by the infliction of an extremely unusual injury rather than a complete and fatal fracture of the spine.

97 Most of the appellant’s experts conceded that the accident did cause at least some harm. Dr. Guimond considered the second operation as the most probable cause of the vertebral subluxation since the appellant’s toes were still moving after the first operation. Dr. Guimond and the appellant’s other experts did not consider the progressive mobilization permitted after the second operation to be a causal factor. Dr. Guimond apportioned the causation as 25 percent due to the accident and 75 percent due to the respondent’s faults because the toes were moving upon arrival and there was no spinal shock. Dr. Bouvier also considered the turning over of the appellant, and the rubbing and baths given to the appellant between the two operations as causal elements of the paraparesis. However, to do so he relied on the respondent’s written note upon the appellant’s arrival in emergency that the neurovascular exam was all right, despite the fact that the respondent testified that “neurovasc” was simply a shorthand manner of saying “vascular”. The limited neurological examination cannot be taken to mean that there was no neurological problem until the second operation. Dr. Guimond did not justify his apportionment of causation. It is inconsistent with the fact that the accident was far more violent than the subsequent movements during the hospitalization, more specifically the second operation.

98 Causation here is a question of fact. There seems to be some confusion as to whether this element in the analysis of civil liability is a question of fact or a question of mixed law and fact. Baudouin and Deslauriers, supra, at p. 348, observe that the Quebec Court of Appeal has a tendency to characterize the issue as one of fact and leaves the determination to the sovereign appreciation of the trial judge. However, they also say that this Court in Morin v. Blais, 1975 CanLII 3 (SCC), [1977] 1 S.C.R. 570, considered the matter to be a question of law. While there is nothing explicit to that effect in Morin, supra, Beetz J. did seem to be aware of the need to justify intervention since he said, at p. 578, that “the determination of a causal relation between one or more facts evincing fault and the damage presupposes a choice, an appraisal and a characterization of the facts”. He then cited Hébert v. Lamothe, 1973 CanLII 173 (SCC), [1974] S.C.R. 1181, as an example of how “this Court did not hesitate to intervene and divide the liability between the two drivers”. The issue was one of the points of contention between the majority judgment and the dissenting opinion of Ritchie and de Grandpré JJ., where de Grandpré J. said, at p. 574, that he found nothing “clearly erroneous” in the judgment of the Court of Appeal, citing Dorval, supra, as well as Pelletier, supra, where this Court held that it could only intervene in the judgment of the Court of Appeal on questions of inference (whether the defendant’s employee was in the performance of his duties) if it was clearly satisfied that an error had occurred.

99 It is important to note that this Court in Morin, supra, sought to correct an approach by the Quebec Court of Appeal where the latter “appeared to see no choice except between two mutually exclusive hypotheses, namely respondent’s fault or appellant’s fault” (p. 578). The Court also laid down a legal principle for a particular circumstance in which a presumption of causation can be made (pp. 579-80).

100 The Quebec Court of Appeal in Bonenfant v. O.T.J. de la Rédemption, [1994] R.R.A. 225, at p. 227, cited Morin, supra, as authority for the proposition that it can intervene on a question of causation since it is an issue [translation] “where the facts and the law intersect, a question of characterization”. I have stated above the peculiarity of the Morin judgment.

101 V. Karim, Commentaires sur les obligations (1997), vol. 1, at p. 271, supports this characterization of causation as a question of mixed law and fact. He says that the intervention of an appellate court is justified when

[translation] the trier of fact does not give reasons for his or her decision in justifying the application of the law to the facts. The same applies when the trial court’s decision does not conform to the evidence adduced, or its conclusions as to the existence of causation cannot be reconciled with the established facts.

102 It is true that the reasons Karim lists are grounds for intervention by an appellate court. However, not giving reasons for a decision, not making the decision conform to the proven facts, and not having conclusions that are reconcilable with the established facts have nothing to do with the nature of the causation inquiry per se. They are general grounds for appellate intervention.

103 The confusion on this issue perhaps stems from an inability to distinguish between cause in the pure physical sense and cause as it is cognizable in law. The latter is a question of law only insofar as we are looking at facts through the lens of the law. However, the inconsequentiality of this observation is made obvious by the truth that everything in judicial decision making is looked at through the lens of the law. This does not make everything a question of law. For example, even questions of pure fact like whether a person was present in a certain place, or whether a person committed a certain act are determined according to the probability of that being so (or according to certainty beyond a reasonable doubt in the criminal domain). This use of the legal rule of evidence of proof on a balance of probabilities to ascertain facts does not transform the question of fact into one of law.

104 In the determination of fault one applies norms of behaviour required by law to a set of facts. This obviously makes the question one of mixed law and fact. In contrast, in the determination of causation one is inquiring into whether something happened between the fault and the damage suffered so as to link the two. That link must be legally significant in an evidentiary sense, but it is rendered no less a question of fact.

105 The difficulty on this point may also arise from the fact that the analysis of causation sometimes depends on presumptions that can be drawn. However, as indicated by the legislator in the Civil Code of Québec in making “Presumptions” a chapter in the title on proof, and as L’Heureux-Dubé J. discussed in Quebec (Public Curator), supra, at para. 47, these too are means of proof and so are properly within the realm of fact.

106 I am not clearly satisfied that the Court of Appeal committed an error in its interpretation of the evidence and its conclusion that given the initial harm caused by the accident, no causation can be attributed to the respondent’s faults. The court found that while an early immobilization and a diminution of the oedema in the spinal cord offered chances of recuperation that were more than simple possibilities, these were not significant enough given the magnitude of the initial harm caused by the accident. The court here touched upon the likelihood of the treatment preventing recuperation. It concluded that the treatment administered by the respondent did not cause the harm suffered. By saying [translation] “more than simple possibilities” and [translation] “not significant enough”, the Court of Appeal basically found the likelihood of an early immobilization leading to recuperation to be somewhere on the spectrum in between the poles of possibility and probability: greater than the realm of what is merely possible but still not enough to meet the threshold of probability. The initial harm of the accident simply outweighed any kind of effect the faulty treatment might have had, to the point where it cannot be said on a probabilistic basis that the faulty treatment had any causal effect. It is worth repeating the traditional principle set out in Laferrière v. Lawson, 1991 CanLII 87 (SCC), [1991] 1 S.C.R. 541, at pp. 608-9, where I found that causation must be established on a balance of probabilities and that the loss of a mere chance cannot be a compensable harm. Accordingly, the chances of recuperation in this case were not significant enough on a balance of probabilities to establish that the faulty treatment caused the harm suffered.

F. Presumptions and Burden of Proof

107 The appellant strenuously argued for a presumption of causation and a reversal of the burden of proof in this case. He posited that since the respondent created a risk and that the harm subsequently occurred within the ambit of the risk created, there should be either a reversal of the burden of proof or a presumption of causation. He also argued that he was deprived of some important means of proof because more tests were not done, and therefore there should be a reversal of the burden of proof onto the respondent to show that his fault did not cause the damage.

108 There are two concepts here that must be disentangled. First, there is the evidentiary tool of presumptions. Second, there is an allocation of liability on defendants where there is an impossibility in proving causation between the specific fault and the damage. I will deal with these in turn.

109 The law of presumptions in the case at bar has its source in arts. 2846, 2847 and 2849 of the Civil Code of Québec (“C.C.Q.”), as well as arts. 1238 and 1242 of the Civil Code of Lower Canada. Since the law of presumptions has to do with proof, the new articles are applicable to this pre-C.C.Q. case by virtue of s. 9 of An Act respecting the implementation of the reform of the Civil Code. As mentioned above, there has been no change in the law on this matter in the transition from the old code to the new one, except that the C.C.Q. has codified the judicial specifications of the kind of presumptions that can be taken into account.

110 Presumptions of fact can establish both fault as well as causation. D. Jutras, “Expertise scientifique et causalité”, in Congrès annuel du Barreau du Québec (1992), 897, at fn. 24, has observed, [translation] “Presumptions of fact serve to establish, sometimes simultaneously, fault and causation.”

111 On the matter of presumptions and burden of proof, Baudouin and Deslauriers, supra, at pp. 358-59, say:

[translation] . . . in medical matters, presumptions of fact sometimes result in a reversal of the burden of proving either fault or a causal link between the action in question and the damage sustained.

J.-C. Royer, La preuve civile (2nd ed. 1995), at p. 518, is more specific in saying:

[translation] This constitutes a reversal of the evidentiary burden, since a presumption of fact does not alter the general principle of article 2803 of the Civil Code of Québec and, consequently, cannot reverse the burden of proof.

Royer says that in order to avoid the confusion of terminology regarding burdens of proof, authors and case law have preferred to use “unfavourable inference” against the defendant rather than a reversal of the burden of proof. One of the cases cited is Snell v. Farrell, 1990 CanLII 70 (SCC), [1990] 2 S.C.R. 311, at p. 330, which of course expresses greater comfort with use of the term “inference” rather than “reversal of burden”.

112 The C.C.Q. itself alludes to this inference or reversal of the evidentiary burden in art. 2847 where it says that a simple presumption may be rebutted by proof to the contrary. Jutras, supra, at fn. 24, accurately discusses the consequence of this reversal for the defendant: [translation] “The defendant may rebut the presumption by showing that the occurrence of the harm is just as consistent with a lack of fault as with the existence of fault.” Nevertheless, the legal burden of proof remains on the plaintiff.

113 As for the conditions of the applicability of presumptions in the civil law context, art. 2849 makes it clear that only serious, precise, and concordant presumptions are to be taken into consideration. Failing the satisfaction of these criteria, the plaintiff must actually establish the unknown fact rather than the trier of fact being permitted to draw an inference from the known fact to the unknown fact.

114 I am not clearly satisfied that the Court of Appeal made an error in its decision not to apply presumptions of causation in favour of the appellant. L’Heureux-Dubé J. in Quebec (Public Curator), supra, at para. 47, said that presumptions of fact were a method of proof among others. Although she was addressing a case where the lower court judge had made presumptions of fact, her comments regarding non-intervention are equally applicable where the lower court chooses not to make presumptions. The refusal to make presumptions is as much an evidentiary decision as is any other acceptance or non-acceptance of methods of proof.

115 There was evidentiary material for the appellant to make his case: the initial x-rays, the doctors’ notes, the nurses’ notes, his friend Jocelyn Richard’s observations, and the subsequent tests. Article 2849 C.C.Q. provides that only serious, precise, and concordant presumptions are to be taken into consideration. These criteria were not fulfilled in this case. The evidence pointed in different and sometimes opposite directions. Jutras, supra, at fn. 24, is correct in saying: [translation] “In cases where the expert evidence is disputed . . ., presumptions of fact are therefore of little use, since it will not be sufficient to assert that the fault is one of the possible causes.”

116 The Court of Appeal appropriately said that it is insufficient to show that the defendant created a risk of harm and that the harm subsequently occurred within the ambit of the risk created. To the extent that such a notion is a separate means of proof with a less stringent standard to satisfy, Snell, supra, and definitely Laferrière, supra, should have put an end to such attempts at circumventing the traditional rules of proof on the balance of probabilities. There may be a misapprehension of what I said in Laferrière, supra, at p. 609: “In some cases, where a fault presents a clear danger and where such a danger materializes, it may be reasonable to presume a causal link, unless there is a demonstration or indication to the contrary” (emphasis added). This is merely a reiteration of the traditional approach on presumptions, and does not create another means of proof in Quebec civil law in the establishment of the causal link. The Court of Appeal correctly interpreted this passage as pertaining to presumptions within the traditional rules of causation.

117 I have dealt with the law of presumptions and how it relates to the burden of proof. The second concept I wish to deal with is the impossibility of proving causation and the allocation of liability.

118 The appellant cited art. 1480 C.C.Q. in his discussion of establishing the causal link. This article imposes solidary liability on defendants where they have jointly committed a wrong causing an injury or where each has committed a separate fault, one of which may have caused the injury. For this article to operate, it is necessary for the plaintiff to show an impossibility to determine the causal connection between the damage suffered and the specific culprit. The classic scenario is one where hunters simultaneously fire their guns, injuring the plaintiff. This was in fact the situation in cases such as Labelle v. Charette, [1960] Que. Q.B. 770, and Massignani v. Veilleux, [1987] R.R.A. 541 (C.A.), as well as the common-law case of Cook v. Lewis, 1951 CanLII 26 (SCC), [1951] S.C.R. 830, where this Court ruled that if the defendants’ own wrongdoing prevents the plaintiff from making the necessary causal connection to the specific author of the wrong, liability is to be attributed collectively so as to avoid the injustice of leaving the victim with no recourse. (See also A. Mayrand, “L’énigme des fautes simultanées” (1958), 18 R. du B. 1, and Baudouin and Deslauriers, supra, at pp. 348 and 370-71, for the doctrinal discussion of the matter.) It is in cases such as these that there is a true reversal of the burden of proof, but not on the issue of causation between fault and damages, but rather on the particular issue of linking the damage to the particular author of the delict.

119 I agree with Morin J. in Stéfanik v. Hôpital Hôtel-Dieu de Lévis, 1997 CanLII 8479 (QC CS), [1997] R.J.Q. 1332, at pp. 1354-55, that art. 1480 C.C.Q. is about the modalities of liability rather than a question of proof, and thus the article is not applicable to situations arising before the advent of the Civil Code of Québec. Morin J. correctly stated though that this was not of great significance since art. 1480 merely enshrined the rule set out by the case law, some examples of which I have given above.

120 There is no reason to apply this rule in favour of the appellant. It should only be applied in cases where there is a true impossibility to determine the author of the delict. There was no such impossibility here: either it was the car accident, or the respondent, or a combination of both in certain proportions.

VII. Disposition

121 For the foregoing reasons, I would dismiss the appeal without costs.

Appeal dismissed.

Solicitor for the appellant: Frédéric St-Jean, Ste-Foy.

Solicitors for the respondent: McCarthy Tétrault, Montréal.