Maxwell v. R., [1979] 2 SCR 1072

Supreme Court of Canada

Maxwell v. R., [1979] 2 S.C.R. 1072

Date: 1979-06-14

Dwight Cecil Maxwell (Plaintiff) Appellant:

and

Her Majesty The Queen (Defendant) Respondent.

1978: October 30, 31; 1979: June 14.

Present: Laskin C.J. and Martland, Ritchie, Spence, Pigeon, Dickson, Beetz, Estey and Pratte JJ.

ON APPEAL FROM THE COURT OF APPEAL FOR ONTARIO.

Criminal law—Murder—Evidence—Judge’s charge to jury—Failure to charge jury as to improprieties of witnesses—Witnesses’ disregard of judge’s direction not to discuss evidence—Change of testimony by witnesses under oath—Adequacy of charge.

The victim in this murder was shot in a parking lot separating two Hamilton hotels at approximately 11:50 p.m. on September 21, 1974. At about the same time noise like a shot or a backfire attracted the attention of a passerby to the parking lot where he saw a man running to a brown or copper coloured van with lettering which he later identified as the name “Barber Appliance Services”. This witness on entering the parking lot found the body of the victim on the ground with what proved to be a fatal gun shot wound in the neck. The van had been some five feet from the body and a man was seen running towards it and entering the driver’s side while leaving the door ajar. Two other witnesses testified that they had seen the van and one was able to take down what he thought was the number of the licence plate. It seemed clear that the driver of the van at this time was the murderer. The evidence was inconsistent with any other rational explanation. The identified van was on the night in question in the custody of Penoffio, the firm’s Hamilton manager. Penoffio lived in a house at 87 Augusta Street, Hamilton, which was also occupied by a friend named Matthews and which the accused was visiting on the evening in question. The van was found next to this address shortly after the shooting and at that time the house was occupied by the accused, Penoffio and Matthews. Both Penoffio and Matthews were obviously potential suspects but neither told a story putting the other in the hotel parking lot at the fatal hour. The Crown’s case depended almost entirely on circumstantial evidence although in so far as the statements of the accused and Penoffio were self-serving they were dependent upon the jury’s assessment of their respective credibility. The accused was convicted of murder and his appeal was dismissed without recorded reasons by a unanimous Court of Appeal. The order

[Page 1073]

granting leave to appeal further limited consideration by this Court to two alleged errors in the judge’s charge, first in failing to caution the jury on the unreliability of the witness Matthews who changed his testimony after discussing it overnight with the witness Penoffio in breach of the trial judge’s direction forbidding such discussion and second in failing to caution the jury on the unreliability of Penoffio who, inter alia, had made previous contradictory statements under oath, and who generally by his evidence and conduct had shown himself to be an undesirable and untrustworthy character.

Held (Laskin C.J. and Spence, Dickson and Estey JJ. dissenting): The appeal should be dismissed.

Per Martland, Ritchie, Pigeon, Beetz and Pratte JJ.: There was some doubt as to whether Matthews had heard the order made when Penoffio was giving evidence forbidding discussions between witnesses. In view of this the trial judge cannot be said to have been in error in failing to instruct the jury as to the unreliability of Matthew’s evidence on account of Matthew’s having acted in breach of an order of which he may well have been unaware. Further the change in Matthews’ evidence had no direct bearing on the whereabouts of Maxwell at the time of the murder, its sole effect was to support the evidence of Penoffio as against that of the accused. In these circumstances a change in evidence from “I cannot remember” to saying “I have now been reminded” did not give rise to such an inconsistency as to require a special caution in the judge’s charge.

There is no doubt that Penoffio was unreliable; however the present case is distinguishable from Binet v. The Queen, [1954] S.C.R. 52, where the witness was an accomplice and a perjurer. Here there was no suggestion that Penoffio was an accomplice, it being on the contrary urged that he was a suspect. Furthermore as the only vital issue was here the identity of the driver of the van at the time of the murder, a matter on which Penoffio gave no evidence, his perjury was not concerned with this single material question and in the circumstances the trial judge was guilty of no error in failing to give more instruction than he did in respect to this witness.

[Page 1074]

Per Laskin C.J. and Spence, Dickson and Estey JJ. dissenting: In this case any evidence as to who committed the crime was fragile at best. Circumstantial entirely, there was in it ground for suspicion against Penoffio and Matthews as well as against the accused. Credibility was central to the case against any one of them and how the trial judge charged the jury on credibility in light of the contradictions in Penoffio’s evidence under oath and the disobedience of Penoffio and Matthews of the order against discussion of the evidence were crucial to the proper disposition of the appeal.

The charge to the jury consisted of a narrative review of the evidence leaving it to the jury to decide as to which submissions or theories to accept. The inconsistent statements of Penoffio under oath and the breach of the judge’s order to the witnesses were left to the jury as going to simple credibility. The charge was inadequate, both as regards Penoffio and Matthews there was non-direction amounting to misdirection. Not only were these witnesses who might have had an interest in exculpating themselves at the expense of the accused but also their interest was associated with such testimonial improperties as called for a clear caution against receiving and weighing their evidence by the standards applicable to an independent witness. The defence here was that it was not the accused but more likely Penoffio who killed the deceased. It was Penoffio who had motive, not the accused. The conflict in their evidence was striking and related to the van, a vital link in the murder. In more than one of Penoffio’s versions he would have had easy access to the van. The jury was entitled to find that Penoffio was a perjured witness and should have been warned in the terms approved in Binet v. The Queen.

There were three circumstances which certainly in their cumulative effect called for a strong caution by the trial judge to the jury against accepting the evidence upon which the Crown’s case was founded, first, the self-interest of the witnesses, second, the inconsistent, contradictory statements under oath on vital issues and third, the breach of the judge’s order; a caution especially necessary, absent independent evidence corroborative of the tainted witnesses.

[Binet v. The Queen, [1954] S.C.R. 52 distinguished; Lucas v. The Queen, [1963] 1 C.C.C. 1; Rustad v. The Queen, [1965] S.C.R. 555 rev’g 9 [1965] 1 C.C.C. 323 referred to:]

[Page 1075

APPEAL from a judgment of the Court of Appeal for Ontario dismissing without recorded reasons an appeal from a conviction at trial by O’Leary J. sitting with a jury on a charge of murder. Appeal dismissed, Laskin C.J. and Spence, Dickson and Estey JJ. dissenting.

C.C. Ruby, for the appellant.

D. Watt, for the respondent.

The judgment of Laskin C.J. and Spence, Dickson, and Estey JJ. was delivered by

THE CHIEF JUSTICE (dissenting)—This is an appeal, by leave of this Court, from a judgment of the Ontario Court of Appeal dismissing, without recorded reasons, an appeal by Maxwell from his conviction of murder after a trial before O’Leary J. and a jury. Leave to appeal was granted by this Court on the following composite question:

Did the Ontario Court of Appeal err in refusing to find that the trial Judge erred, in the circumstances of this case, in (1) failing to caution the jury on the unreliability of the witness Matthews who changed his testimony after discussing it overnight with the witness, Penoffio, in breach of the trial Judge’s direction forbidding any such discussion; and (2) in failing to caution the jury on the unreliability of the evidence of Penoffio who, inter alia, had made previous contradictory statements under oath?

The accused, thirty-two years of age at the material time, was then the operator of a small painting and construction business. He had served in the Canadian armed forces, worked as a prison guard, in security service and as a special constable with the traffic division of the Hamilton Police Department, and later as a house detective for a hotel before going into business for himself. He worked concurrently part-time for Barber Appliance Services and was at the time of the alleged offence doing some painting for one Stone-house and for one Penoffio and was given a key to the latter’s home in that connection. He was wont to sleep overnight in the homes where he was working but did have a residence of his own.

[Page 1076]

The deceased, victim of the killing, one Maurice Rodriguez, had been seen in one of the beverage rooms in the Elmar Hotel in Hamilton, Ontario, on the evening of September 21, 1974 and was also seen to leave by himself at about 11:30 p.m. He was shot and killed a short time later in a parking area between the Elmar and another hotel. There was evidence of a noise like a backfire or a shot being heard, coming from the parking area, and of a man seen running to a parked van which he drove away. Several witnesses described the van which had Barber Appliance Services written on the side and one witness noted the licence number as it was driven away. Shortly after midnight, about 12:05 or 12:10 a.m. in the morning of September 22, 1974, the van was located in the driveway of a house belonging to the witness Penoffio. The fact that the witness who took the licence number was mistaken as to one digit of the licence did not impede the speedy discovery of the van. The house, at 87 Augusta Avenue, was about three to five minutes’ driving time distant from the Elmar Hotel.

It is obvious, there being no evidence of any other possibility, that whoever drove the van away from the parking lot was the killer of Rodriguez.

The deceased had been a tenant of Penoffio at 87 Augusta for about a year, until Penoffio terminated the tenancy in April 1974 by notice to quit, allegedly because of the deceased’s misbehaviour which brought the police to the premises several times. There was resulting hostility between the deceased and Penoffio, who testified to harassing telephone calls by a caller, whom he believed to be the deceased and who hung up when the telephone was answered. There were other incidents, which need not be narrated, indicating a bad relationship between the deceased and Penoffio.

Penoffio, at the material time, was the Hamilton manager of an appliance service, Barber’s Appliance Services. The van seen leaving the scene of the murder belonged to this firm and carried the firm name on its side. Penoffio had picked up the

[Page 1077]

van on September 20, 1974 from the owner of the firm and left his own car with the owner. Of the three or four keys to the van, the owner had one, Penoffio had one and one was kept by an employee in Brantford where the firm also operated.

There was no evidence that the accused Maxwell was at the Elmar Hotel in the evening of September 21. The accused knew the deceased and, on the evidence, they were on friendly terms. There was no motive on Maxwell’s part to harm or kill the deceased. The accused had slept at the home of a witness, Stonehouse, on the night of September 20, being engaged in doing some painting for him. The next evening, September 21, Stonehouse and a friend of his invited the accused to join them for a drink at a tavern, the Running Pump. He declined, saying he was going to visit Penoffio. Stonehouse testified that he and his friend had offered to give the accused a lift and he was dropped at 87 Augusta about 10:25 p.m., saying as he left “I will see you cats later”. According to Stonehouse, the accused did not turn up at the Running Pump, but Stonehouse did see Penoffio there that evening.

The accused gave evidence as did Penoffio and also the then tenant of Penoffio, one Robert Matthews. It is the conflict in their evidence, a conflict on crucial matters, that goes to the heart of the two issues on which this Court gave leave.

I refer first to the evidence of the accused as to the course of events after he came to Penoffio’s house at 10:25 or 10:30 p.m. on September 21. He testified that he went there at Penoffio’s telephoned request to meet him about 10 p.m., the accused believing it was to join him in going out for a few drinks to a tavern, the Corktown, which they visited two or three times a week. The accused let himself into the house with a key which he had been given while he was painting on the premises. Penoffio was not at home—the accused even knocked at the bedroom door—so he helped himself to a bottle of beer and, when Penoffio did not return, assumed he had gone on

[Page 1078]

ahead to the Corktown. Finding a car key on the kitchen counter, he took the van and drove to the Corktown looking for Penoffio. He had a few draft beers and after fifteen or twenty minutes he drove back to 87 Augusta and parked the van in the driveway. Penoffio was not in the house and the accused took a pint of beer from the refrigerator and went into the living room where he turned on the television for the 11 o’clock news. He then listened to some music and drank some more beer. About this time he heard and caught a glimpse of someone leaving the house by the back door. He was unable to identify the person but it was not Penoffio. Shortly thereafter, about midnight, Penoffio did come in from the back door. A few minutes later the police arrived.

Penoffio testified that he awakened about 10:30 p.m. on the evening of September 21 and found the accused at his place with the tenant Matthews. He said they watched television for a time, and then he and the accused drove in the van to the Corktown, with Penoffio driving. The place was full and Penoffio decided not to stay, gave the key of the van to the accused and walked over to the Running Pump where he drank some beer and then returned to the Corktown. He had left the accused at the Corktown about 11 p.m. and it was about midnight when he returned. The accused was not there and, after having a beer, Penoffio walked home, arriving about 12:05 to 12:10 a.m. The van was in the driveway and the accused was in the kitchen having a bottle of beer. The police arrived shortly thereafter.

Penoffio testified at the trial that when he and the accused were driving to the Corktown the accused had asked for the loan of the van next day, Sunday, to enable him to move some furniture. Penoffio agreed. The van at that time contained a large number of vacuum cleaners. It was put to Penoffio on cross-examination that he had testified at the preliminary inquiry that the accused had asked to borrow the van on the following Tuesday to move a stove. The transcript, which the Court reporter said was accurate, showed that he did say

[Page 1079]

Tuesday at the preliminary hearing but he said at the trial that this was a mistake. Again, at the preliminary inquiry, Penoffio testified that after drinking at the Running Pump on the evening of September 21 (after he left the accused at the Corktown) he went directly home without going back to the Corktown. He said then, it being about a quarter to twelve, that

I walked up King Street to Catherine and up Catherine to my house. I noticed the truck in the driveway and I knew the lights in my house were on. So, I went to the back door and went into the kitchen.

At the trial he said his answers were untrue, that he did not walk up King Street to Catherine and up to Catherine to his house, but that he walked “up” Walnut from the Running Pump to the Corktown.

Walnut Street ran parallel to and was east of Catherine Street. Both were intersected by King Street and by Main Street which was below King Street. Caroline Street ran parallel to Catherine Street but at some distance to the west. The Elmar Hotel was at or near the corner of Caroline and Main and thus northwest of Penoffio’s house which was on Augusta at the corner of Catherine. The Running Pump was north of his house and the Corktown was south of it. His evidence at the preliminary inquiry meant that he would be walking south from the Running Pump close to midnight to get to his house. His evidence at trial meant that he would be walking north on Catherine Street from the Corktown to get home. A police officer had testified at the preliminary inquiry that while Penoffio’s house was under surveillance shortly after midnight, he had seen Penoffio walking north on Catherine Street to the house. Penoffio’s change of testimony at the trial thus made it conform to the police officer’s testimony at the preliminary hearing. He gave evidence that he had sought but not obtained from a police officer a copy of his evidence at the preliminary hearing while in custody during the first trial of the accused. He had been arrested for failure to appear at the first trial of the accused, which had resulted in a mistrial, and had been released after

[Page 1080]

two days on his promise to appear at the second trial.

It must be remembered, however, that the officer was maintaining a surveillance of Penoffio’s house after the murder was committed, and his testimony was that he saw Penoffio walking north to his house at about 12:10 a.m.

Again, in his evidence at trial Penoffio stated that he had never lent the van to Matthews. This was contrary to his evidence at the preliminary enquiry where he said the opposite. He did say at trial that the key to the van was usually kept on a ring on a nail in the kitchen of his house and that different persons had had access to it.

I come to Matthews’ testimony. He stated in evidence that he watched television until about 10:30 p.m. on September 21, after which he went to bed. According to him, the accused arrived at the house while he was watching television but they did not speak, although Matthews knew the accused. He said Penoffio was in his bedroom (a downstairs room) and did not join in watching television before he, Matthews, went off to bed. He said in chief that he could not remember whether he had seen Penoffio at the house before going to bed.

The trial judge had made an express order, of which both Penoffio and Matthews were aware, for the exclusion of other witnesses from the court room while evidence was being given by one of them and that witnesses who had completed their testimony were not to speak about the case during any recess with those who had not. Matthews’ examination in chief had concluded at the end of a day’s hearing and, despite the judge’s order, Penoffio spoke to Matthews during the night recess and prior to his cross-examination. In the result, admitting that Penoffio had spoken to him, Matthews said on cross‑examination that he had dinner with Penoffio the evening of September 21 at the house, that he talked with the accused when the latter arrived, that he did see Penoffio that evening before going to bed. It appears from the record that his change of testimony to say that he

[Page 1081]

saw Penoffio that night was prompted by Penoffio and not be any fresh remembrance.

When the police arrived at 87 Augusta and interrogated the accused, Penoffio and Matthews, all three were taken to the police station. Neither Penoffio nor Matthews was searched at the house or at the station for a key to the van but only asked whether they had one, and both men were released in the early hours of the morning.

The accused, when taken to the police station, was questioned there as to his whereabouts from 10 p.m. on the previous evening. Informed that the police were investigating the shooting of Rodriguez, the accused said he was more or less stunned and told them he knew nothing about it. Asked later if he had a key to the van, the accused gave it to them from a table where his personal effects were laid out. About 7 a.m. two police officers returned to interrogate him further and told him he was charged with murder. There were no police at the house between the time of the first search thereof at about 1 a.m. and a second search at about 7 a.m. On the second search the police found a shot gun with a single unfired shell in a paper bag in a cupboard underneath the basement stairs of the house. Alongside it was a box of similar shells. A spent shotgun shell was found in the back yard of the adjoining premises about fifteen feet from the back door of Penoffio’s house.

The shotgun, found in a dismantled condition, was allegedly the murder weapon but there were no identifying prints on it. Prints found on the box of cartridges did not match those of the accused and no prints were taken of Penoffio in that respect. The accused disclaimed any knowledge of the gun or of the cupboard under the stairs. Penoffio testified that he never kept a gun in the house, and to his knowledge Matthews did not have a shotgun in the house.

[Page 1082]

The only identifiable print of those found on the van was one of the deceased. Some blood was found but could not be identified as to grouping. There were spots of blood on the clothes of the accused, which he explained by reference to a previous altercation with others, or to cuts which occurred in the course of his work in painting around eavestroughs, but again they were not sufficient for identifiable grouping.

There was no identification of the driver when the van was driven away from the scene of the crime. Although a witness, Janice Brady, who lived next door to 87 Augusta, testified that she heard the van drive in to Penoffio’s driveway about 12.10 a.m. in the morning of September 22, she did not see either the van or the driver. She did say that she saw the accused and Penoffio together in front of Penoffio’s house about 10:20 p.m. on September 21. The accused’s evidence was that she was mistaken on this point and, equally, he contested the evidence of Penoffio and Matthews that he had been with them in Penoffio’s house that evening.

This is a case where any evidence as to who committed the crime was fragile at best. Circumstantial entirely, there was in it ground for suspicion against Penoffio and Matthews as well as against the accused. Credibility was very central to the case against any one of them, and how the trial judge charged on this in the light of the contradictions in Penoffio’s evidence under oath and of the disobedience by both him and Matthews of the trial judge’s order against discussion of the evidence are, in my opinion, crucial to the proper disposition of this appeal.

The charge to the jury consisted of a narrative review of the evidence, largely according to the submissions thereon made by counsel for the Crown and by counsel for the accused, ending with the statement “Now, members of the jury, that is the two sides as it has been presented to you. It is for you to decide which theories to accept and which to reject and the decision is entirely yours”. In dealing with Penoffio’s inconsistent statements under oath and with the breach of his order to

[Page 1083]

witnesses, the trial judge did not give any special caution to the jury but left the matters to them as going to simple credibility. I quote all that he said in this respect:

He [the accused] was asked whether Penoffio and Matthews had concocted their story and his answer was yes and they could also be confused and Janice Brady also is mistaken or confused when she says she saw Penoffio and Maxwell together that evening.

Anne Nutting, a court reporter, testified that her shorthand notes recorded the witness Penoffio as saying at the preliminary hearing that Maxwell wanted to borrow the truck for Tuesday whereas Penoffio said in the witness box here that at the preliminary hearing that he had said Sunday, not Tuesday.

Anne Nutting also testified that her notes recorded Penoffio as saying, again at the preliminary hearing, that he had loaned the vehicle to Matthews whereas he said from the witness box that he had not said that at the preliminary.

Now, it is for the jury to decide whether Penoffio was incorrectly heard by or incorrectly recorded by the reporter or whether he did in fact say Tuesday and not Sunday and that he did in fact say that he loaned the vehicle to Matthews at the preliminary hearing. It is for you to say who was in error, whether it is the reporter or whether it is Penoffio, and if you find that Penoffio is indeed in error as to whether he said those things at the preliminary, what effects, if any it is to have on his credibility. These are matters for you to weigh and you should consider, of course, whether people are making honest mistakes or deliberately misquoting, and if they are making honest mistakes, in any event, whether there is something wrong with their ability to recollect. These are matters for you to take into account; whether or not they are reliable.

Something that is worthy of some comment, though, is the fact that during the evidence of Penoffio I made an order excluding witnesses from the court room and that order goes on—it was read by the Clerk to advise those who have testified not to speak about the case to those who have not. Now in spite of that order, Penoffio, it appears, spoke to Robert Matthews during the night recess. In other words, while Matthews was still in the course of giving his evidence and the next day Matthews gave different testimony than he had the night before.

[Page 1084]

. . .

He gave evidence before the recess for the evening to the effect that he did not know if he saw Penoffio at 87 Augusta Street before he, Matthews, went to bed on the evening of September 21st, and he indicated that he went to bed about 10.30. The following day, he said, “after dinner I think Penoffio went to have a sleep and I saw him again about 10.00 or 10.30 for about 15 minutes”. So the change in his evidence was from not remembering whether he had seen Penoffio to saying that he saw him for about 15 minutes between 10.00 and 10.30 and he went on to say that Dwight Maxwell was there when he was talking to Penoffio in that fifteen minute period or, when he saw him for 15 minutes and then he admitted that he had spoken to Penoffio last night—which would be during the recess—about my evidence and Penoffio reminded me that I had seen him before I went to bed.

Now, of course, the purpose of an order excluding witnesses is to avoid the possibility of the evidence of one witness being contaminated, as it were, or improperly influenced by hearing what other witnesses have to say. Nevertheless, it is for you to say what effect the failure to abide by my order has on the credibility of either Matthews or Penoffio. As has already been submitted by counsel for the Crown, if they were planning to concoct a story it is conceivable that they would have concocted it long before the recess during which Penoffio spoke to Matthews, but there are arguments both ways in the matter and it does not follow because one or both of them failed to abide by the order, that you should automatically rule out the evidence, but it is a matter for you to consider and I brought that to your attention and you have heard from both counsel as to the views that you should take.

Obviously the order is made for a purpose and that is to prevent contamination from the receiving of suggestions and there is not any doubt that the evidence of Matthews was so affected by what Penoffio did. Whether he is to be believed is another matter and that is for you to decide.

II

In my opinion, the trial judge’s charge to the jury was inadequate, both as regards Penoffio and Matthews. There was non-direction amounting to misdirection in respect of these two witnesses, whose testimony was central to the Crown’s case against the accused. I say this not only because

[Page 1085]

they were witnesses who may be said to have had an interest in exculpating themselves at the expense of the accused, but because their interest was associated with testimonial improprieties so as to call for a clear caution against receiving and weighing their evidence by the same standards applicable to an independent witness.

I should note, before proceeding to a discussion of the relevant authorities, that the trial judge, although mentioning Penoffio’s inconsistent statements and his breach of the trial judge’s order as to out-of-court exchanges between witnesses, did not mention Penoffio’s failure to appear at the first trial of the accused which ended in a mistrial.

One further matter deserves attention at this point. The fact that counsel for the accused, in addressing the jury, may have urged them not to accept the evidence of one or more key Crown witnesses is no substitute for a necessary direction by the trial judge as to the caution with which such evidence should be viewed where, as here, the circumstances and evidence call for it. As Cartwright J. put it in Binet v. The Queen[1], at p. 54, (a case to which I will return later), “I do not think that the circumstance that counsel for the defence stressed the fact of the conflicting statements having been made in any way absolved the learned trial judge from the duty of dealing with them”.

Counsel drew the attention of this Court to a line of cases in England, e.g. R. v. Prater[2], and R. v. Russell[3], which speak of the desirability of giving the warning against uncorroborated evidence where a witness may be regarded as having some purpose of his own to serve, subject however to application of the proviso as to no miscarriage of justice if such a direction to the jury was

[Page 1086]

omitted. Lord Hailsham, in Director of Public Prosecutions v. Kilbourne[4] at p. 447, expressly approved of the desirability of such a direction but there appears to have been a recession in respect of its desirability, at least where the self interest of a Crown witness is patent: see R. v. Whitaker[5].

However, judgments in this country have been quite clear that the accomplice warning is not a requirement nor does it render a jury verdict vulnerable if not given in respect of a witness who may be said to have an interest of his own to serve but is not himself an accomplice: see R. v. Dutrisac[6]. The trial judge is not absolved, however, from the duty, measured by the particular circumstances before him, of drawing the jury’s attention to a witness who may be serving his own interests at the expense of an accused and of counselling caution in weighing his evidence.

The present case, as I have already noted, goes beyond the circumstances of a Crown witness affected by self-interest, involving as it does a Crown witness who has also made prior inconsistent statements under oath on a vital issue. As to such a witness, the proper direction should be as indicated in Binet v. The Queen, supra. The difference of opinion among the members of the Court in that case lay only in whether (as the majority thought) there should be a new trial or whether (as the minority of Cartwright J., as he then was, Rand J. concurring, thought) there should be an acquittal. Cartwright J. said this (at p. 54):

The learned trial judge warned the jury in terms to which no exception is taken of the danger of convicting on the uncorroborated evidence of an accomplice but he failed to give them any direction in regard to the fact that Giroux had on two previous occasions made statements on oath which were in direct conflict with the evidence which he had given at the trial on a vital point.

I respectfully agree with Barclay and Hyde JJ. that, in the circumstances of this case, the omission to direct

[Page 1087]

the jury in this regard was an error in law so serious as to require that the conviction be quashed. I do not find it necessary to refer to all the authorities which were discussed by counsel. I am in respectful agreement with Hyde J. that the applicable rule is correctly stated by Errol McDougall J. who gave the judgment of the majority in Rex v. Stack and Pytell [1947] 3 D.L.R. 747 at 762; 88 Can. C.C. 310 at 327 in the following words:

Where the testimony of a principal Crown witness is in direct conflict with a prior sworn statement made by him the trial Judge must caution the jury in the strongest terms with respect to the danger of accepting his evidence, and the failure to do so will necessitate a new trial, notwithstanding that the trial Judge properly instructed the jury with respect to the evidence of such witness in the event that they concluded that he was an accomplice.

This principle was followed in R. v. Dutrisac, supra, a judgment of the Ontario Court of Appeal and, earlier, by the same Court in R. v. Rosenberg[7].

Two other judgments of this Court deserve reference in respect of what was said in the Binet case. They are Lucas v. The Queen[8], and Rustad v. The Queen[9]. I should note first that the majority in Binet, speaking very briefly through Taschereau J., as he then was, said at p. 52, that they agreed with Cartwright J. “that the learned trial judge failed properly to instruct the jury on the great danger of accepting the evidence of an admittedly perjured accomplice on a vital issue”. However, it is clear from the reasons of Cartwright J. that the duty of a trial judge in respect of an accomplice who had made inconsistent statements under oath on a vital issue was not only to give the jury the usual warning as to corroboration, if they found that the witness was an accomplice, but also to give them a caution in the strongest terms of the danger of accepting his evidence. This was entirely apart from the witness’ position as a possible accomplice, and I take Taschereau J.’s statement of agreement with Cartwright J. as referring to the position of the witness whether or not he be an

[Page 1088]

accomplice. The two Ontario cases above-mentioned proceed on this basis.

In the Lucas case, involving a charge of murder, Kerwin C.J.C. considered Binet in the following passage of his reasons (at p. 10):

I reject the submission of counsel for the appellant that it was error in law on the part of the learned trial Judge not to charge in accordance with Binet v. The Queen, [1954] S.C.R. 52, to the effect that there is great danger in accepting the evidence of an admittedly perjured accomplice on a vital issue. Thomas was not an accomplice. There was a mass of evidence pointing to the accused which makes it impossible to find, as counsel submitted, that the evidence of Thomas was on a vital issue, and finally, there was much evidence in the record which did corroborate Thomas. The matter was left to the jury as one of weight and credibility and, in my opinion, correctly so.

It is enough to distinguish Lucas from the present case that in Lucas the witness, in respect of whose evidence the omission of a caution of danger in acting upon it was urged as an error of law, was found not to have given evidence on a vital issue. Cartwright J., who dissented in the Lucas case, recognized this difference from Binet but was prepared to carry Binet further as is evident from the following portion of his reasons (at pp. 20-21):

I agree with the submission of counsel for the appellant that it was the duty of the learned trial Judge to warn the jury of the danger of accepting the evidence of Thomas in view of his admitted perjury and of his bad record. Counsel for the appellant cited a number of authorities. I will refer only to the following:

In R. v. Stack & Pytell (1946), 88 Can. C.C. 320 at p. 337, [1947] 3 D.L.R. 747 at p. 762, Errol McDougall, J., giving the judgment of the majority of the Court of Queen’s Bench, said:

Where the testimony of a principal Crown witness is in direct conflict with a prior sworn statement made by him the trial Judge must caution the jury in the strongest terms with respect to the danger of accept-

[Page 1089]

ing his evidence, and the failure to do so will necessitate a new trial, notwithstanding that the trial Judge properly instructed the jury with respect to the evidence of such witness in the event that they concluded that he was an accomplice.

This passage was approved by Rand, J., and myself in Binet v. The Queen, [1954] S.C.R. 52 at p. 54, and I do not read the judgment of the majority delivered by Taschereau, J., as casting doubt upon it. The Court was unanimous in quashing the conviction; the difference of opinion was as to whether there should be a new trial or a directed verdict of acquittal. The Binet case is distinguishable from the case at bar on the grounds that there is nothing to indicate that Thomas was an accomplice in the murder of Crater and that his evidence was not direct evidence on the vital issue as to who was the murderer. Thomas’ evidence was however as to circumstances which, if it were believed, would weigh heavily with the jury.

In short, as is also evident from his reference to R. v. Ferguson[10], again a judgment of the Ontario Court of Appeal, Cartwright J. felt that the warning approved in Binet should also apply in respect of a disreputable witness. He noted too that Kerwin J., as he then was, had agreed in his reasons in Deacon v. The King[11], at p. 536 with a passage from the Ferguson case to the above effect.

I am not required in this case to consider the proposed extension by Cartwright J. of the kind of warning which Binet requires in the case of a witness who gives inconsistent statements under oath on a vital issue. Penoffio is not in the position of the witness in Lucas but is in the position of the witness in Binet, apart from the accomplice element.

Rustad v. The Queen was concerned, inter alia, with the correctness or acceptability of the trial judge’s charge as it concerned two witnesses who in statements to the police prior to trial, statements not made under oath, had failed to mention that the accused had told them that she had killed

[Page 1090]

the deceased. One of the witnesses, who testified under oath at an inquest, did not mention the accused’s admission at that time. The witnesses did give evidence as to the admission at the trial. The trial judge in a long and detailed charge (which I have read) did not caution the jury in the strong terms required under the Binet rule. Ritchie J., who delivered the judgment of this Court (Cartwright J. being among those who agreed with him) said this on the point (at p. 560):

Although it is true that Mr. Justice McInnes would have been justified in using stronger language to describe the weaknesses inherent in the evidence of both these witnesses, I am none the less of opinion that he said enough to indicate that in weighing their evidence the jury should give serious consideration to the inconsistencies in Mrs. Shannon’s statements and to the failure of both women to come forward with their stories at an earlier date. I think that the theory of the defence that these two witnesses were unworthy of belief was expressed in the judge’s charge with sufficient clarity to comply with the requirements indicated by this Court in Deacon v. The King, and in the other cases referred to in the reasons for judgment delivered by Sheppard J.A. on behalf of the majority of the Court of Appeal. I would not quash the conviction on this ground, but there are more serious omissions which require consideration.

In the result, the conviction of manslaughter was set aside and a new trial directed on that charge because of the trial judge’s failure to give proper direction to the jury as to the effect of the accused’s intoxication on the truth or falsehood of the alleged incriminating admissions to the two witnesses.

It may be that, in view of the order for a new trial and the grounds on which it was put, the admissions were wrapped up in the intoxication and hence the charge was defective in that composite context. Furthermore, the charge to the jury in the Rustad case was much more detailed and comprehensive than the charge in the present case and carried caution even though not explicitly stated. The same observation may be made about the judgment of the Ontario Court of Appeal in R. v. Agawa and Mallet[12]. Martin J.A. who spoke for the Court said this (at p. 394):

[Page 1091]

The learned trial Judge put the defence of each of the appellants fully and with exemplary fairness to the jury. In the course of his charge he reviewed the evidence given by each of the inmate witnesses, previously referred to, and in so doing, almost invariably drew the jury’s intention to the fact that the criminal record of the witness was filed as an exhibit and referred to the previous contradictory statements made by some of the witnesses at the preliminary hearing.

In the present case the inmate witnesses were in no way associated with the commission of the offence, and it was not incumbent upon the trial Judge to charge the jury that it was dangerous to act upon their uncorroborated evidence because of their bad character: R. v. Bouffard, [1964] 2 O.R. 111 at p. 127, [1964] 3 C.C.C. 14 at p. 30, 43 C.R. 124. Nor was such a direction required because of their possible motive to obtain clemency, nor, in all the circumstances, by reason of the fact that they had made contradictory statements on oath at the preliminary hearing: R. v. Dutrisac, [1971] 3 O.R. 412 at pp. 417-8, 4 C.C.C. (2d) 13 at pp. 19-20, 15 C.R.N.S. 15; Lucas v. The Queen, [1963] 1 C.C.C. 1 at p. 10, [1963] S.C.R. vi, 39 C.R. 101.

Although I do not doubt that there are cases where the trial Judge should, as a matter of prudence, direct the jury to view the evidence of a witness with caution, for example, because of his unsavoury character, motives of self-interest, or previous contradictory statements, here, in my view, considering the trial Judge’s charge in its entirety, the jury were adequately directed with respect to the matters that should be considered in assessing the credibility of the inmate witnesses.

When one turns to the reasons of the British Columbia Court of Appeal in the Rustad[13] case, delivered by Sheppard J., Norris J.A. concurring, and also by Davey J.A. to the same result (dismissal of the appeal from conviction of manslaughter), there is a difference in view between the two sets of reasons as to the effect of Binet, and I wish to refer to it.

Mr. Justice Sheppard treated the Binet case as if it concerned the proper instruction to be given to the jury only in the case of a perjured accomplice;

[Page 1092]

and, since the witness who had given inconsistent statements both prior to trial and also under oath was not an accomplice, the Binet warning was unnecessary. Again, he rejected case law in England to the effect that a witness who has given inconsistent statements may be regarded as negligible, pointing out that that view was rejected by this Court in Deacon v. The King[14]. The Deacon case is not germane here because it concerned a Crown witness who was declared adverse and consequently open to be examined by the Crown on previous inconsistent statements. Although a trial judge could properly draw attention to this on the issue of credibility, it afforded no ground for obliging him to direct the jury to discount the evidence of the witness.

I take no issue here with what the Deacon case held, but I do not think that Sheppard J.A. properly assessed the Binet case. Indeed, he appears in his reasons to have ignored, as not having any materiality, the inconsistency of statements made under oath and to have treated the case before him as if it concerned only prior statements by the witnesses made to police officers. He said this (at p. 338):

As the evidence of Mrs. Shannon and the effect thereon of her prior statements are matters of weight and credibility for the jury, the nondirection can be misdirection only if it failed to comply with Azoulay v. The Queen, [1952] 2 S.C.R. 495 at p. 497, where Taschereau, J. said:

The rule which has been laid down, and consistently followed is that in a jury trial the presiding Judge must, except in rare cases where it would be needless to do so, review the substantial parts of the evidence, and give the jury the theory of the defence, so that they may appreciate the value and effect of that evidence, and how the law is to be applied to the facts as they find them.

. . .

The previous statements of Mrs. Shannon are no part of the theory of defence but under the circumstances of

[Page 1093]

this case are at the most a detail of evidence or argument. This accused called no evidence and the learned trial Judge instructed the jury as to the onus being on the Crown, the presumption of innocence and as to those facts of which the Crown must assume the onus as necessary to constitute the offence charged and the included offence of manslaughter.

Davey J.A. was of the opinion that the Binet case did not lay down any general rule of law or of practice and, moreover, he did not see any disagreement between the majority and minority in that case on what Cartwright J. had said in quoting from the reasons of Errol McDougall J. in R. v. Stack and Pytell. He viewed Taschereau J.’s reasons for the majority to rest not upon the fact that the impugned Crown witness who gave perjured evidence was an accomplice but rather that the required direction in that case applied to all perjurers who gave vital evidence, whether accomplices or not. This, as I already noted, is also my assessment of the majority’s reasons in Binet.

Davey J.A. referred also to Cartwright J.’s reasons in Lucas, which I have quoted, and to the cases mentioned in those reasons and felt that the rationale upon which Binet was decided was reflected in them. He summed up as follows:

From these authorities it seems to me that the obligation to give such a direction arises not from a distinct rule of law or of practice, but from the obligation resting upon a trial Judge under Azoulay v. The Queen, [1952] 2 S.C.R. 495, and Kelsey v. The Queen, [1953] 1 S.C.R. 220, to review the substantial parts of the evidence, and to give the jury the theory of the defence, so that they may appreciate the value and effect of the evidence, and how the law is to be applied to the facts as they find them; and to present clearly to the jury the pivotal questions upon which the defence stands; and that, although it is not necessary that the trial Judge should review all the facts, it is not sufficient that the whole of the evidence be left to the jury in bulk for examination.

From that it follows that the need for such an instruction must depend upon the circumstances of the particular case and the nature of the defence. A trial Judge

[Page 1094]

ought to point to all substantial weaknesses in the Crown’s case. Where a principal ground of defence is the unreliability of the Crown’s evidence, and the vital evidence is that of a self‑confessed perjurer, whether an accomplice or not, I do not see how such a defence can be adequately put to the jury without warning the jury of the danger of convicting on such evidence, for the witness by his conduct has shown he has no respect for his oath, and he is just as likely to be lying at the trial as on the earlier occasion. Such was the case of Binet v. The Queen. The same need arises, although perhaps not in the same degree, in the case of a witness who has made previous contradictory statements not under oath upon a vital point, or whose record shows him to be untrustworthy. Where the evidence is not upon a vital point, or there is a substantial body of other evidence implicating the prisoner, or where there is an explanation for the previous untruthful statement, the nature of the instruction will vary according to the circumstances; it must be sufficient to enable the jury to appreciate the value and effect of the evidence in the light of the theory of the defence; it must present to the jury the pivotal questions upon which the defence stands. If judicial experience shows that it is dangerous to convict upon such testimony the jury ought to be so told. If the circumstances are such as to allow the defence and the evidence to be fairly put to the jury without giving the warning, the warning need not be given.

Where a plausible explanation is given at the trial for the previous contradictory statements, whether under oath or not, it may be sufficient to leave the defence to the jury along with the salient points touching the conflicting statements and the explanation without any warning and allow the jury to say whether it is satisfied with the Crown’s evidence at the trial and the explanation for the previous conflicting statements, if in the opinion of the trial Judge the jury might safely convict on that evidence. A jury ought not to be told there is danger in convicting on such evidence, if in fact there is none.

In my view, whether what was said in Binet to be a necessary direction arises from the nature of the circumstances and from the nature of the defence, as Davey J.A. would have it, or is required as a matter of law or practice, as in the two Ontario cases which applied Binet, the present

[Page 1095]

case, on its facts, exhibits a strong need for that direction.

The defence here was that it was not the accused but more likely Penoffio who killed the deceased. So far as motive was concerned, it was Penoffio who had one, not the accused. The conflict in their evidence was striking. The van was a vital link in the murder. Penoffio testified that he gave the van to the accused about an hour or less before the murder after they had gone out in it together. This was in contradiction to the accused’s evidence that he had taken the van from the house to go looking for Penoffio when he did not find him at home and had returned it to Penoffio’s driveway about an hour before the murder.

The accused, Penoffio and Matthews were all at 87 Augusta when the police came, and the van was then parked in the driveway. Penoffio gave unexplained contradictory evidence as to his whereabouts at the material time; one version, at the preliminary hearing, would have left him out on the street, without an alibi, at the time of the murder; the other, at the trial, established consistency with police evidence as to his whereabouts shortly after the murder but that is all. In either case, he would have had easy access to the van.

There was again an unexplained contradiction (the Court reporter had to be called when Penoffio denied what he said at the preliminary enquiry) as to his story of the accused’s request to borrow the van, first saying Tuesday and then, at the trial, Sunday. Moreover, Penoffio testified at the trial that he had only allowed the accused to borrow the van when he had said at the preliminary enquiry that Matthews used it. Once again the Court reporter was called to verify this when Penoffio first denied it in testifying at the trial.

These were all vital matters and the jury was entitled to find that Penoffio was a perjured witness and should have been warned in the terms

[Page 1096]

approved in Binet in such situations. Of course, the required warning need not follow any precise formula (I think this emerges from the reasons of this Court in the Rustad case), so long as the trial judge makes it quite clear to the jury that it is dangerous for them to act on the evidence of the inconsistent witness given on a vital issue, especially when, as here, it is the evidence of that witness alone or of that witness and an associate, Matthews, both possible suspects, that purports to ensnare the accused. In short, since Penoffio’s evidence was central to the Crown’s case, the trial judge would have been justified in stressing that there was very great danger in acting upon it, especially when, in addition to the inconsistencies in his evidence, Penoffio deliberately ignored the trial judge’s order prohibiting out-of-court exchanges with other witnesses and influenced Matthews to change his testimony.

Nothing approaching such a caution is found in the trial judge’s charge. Indeed, the trial judge failed to bring into account, when he talked of the contradictions in Penoffio’s evidence, the contradiction in his evidence as to the direction in which he was walking the night of the murder. Again, the trial judge failed to deal adequately with Matthews’ evidence on whether he actually saw or did not see Penoffio when the accused arrived at the house at about 10:30 p.m. on September 21.

The breach of the trial judge’s order as to communication between witnesses is a separate ground upon which a caution should have been given to the jury in acting upon evidence of a witness who so readily changed his testimony at the behest of a fellow witness to make it conform to the latter’s evidence. It was conceded by Crown counsel that the trial judge should have but did not tell the jury that they should act on a witness’s own recollection and not on what the witness was told by another. Indeed, evidence which is not that of the witness’s own independent recollection of

[Page 1097]

what he is testifying to would not likely be admissible: cf. R. v. Muise[15], at p. 497. The trial judge’s charge got close to this issue but, in my opinion, it did not sufficiently put the jury on guard against believing Matthews on the vital question of the whereabouts of the accused at the time of the murder.

Matthews’ first version that he could not remember seeing Penoffio in the house at 10:30 p.m. when he, Matthews, went to bed was consistent with the accused’s evidence on the point. Matthews’ change of testimony resulted in the corroborating of Penoffio and contradicting the accused.

I am not concerned here with any issue of the admissibility of Matthews’ evidence. In so far as he gave inconsistent statements under oath on a vital issue, the rule of the Binet case would apply equally to him as to Penoffio. In so far as he was involved in a breach of the trial judge’s order, barring communication between witnesses, the trial judge was not required to exclude his evidence (see the authorities cited in Dobberthien v. The Queen[16], but was obliged, when sitting with a jury, to tell them to receive the evidence with caution when it was changed by reason of breach of the trial judge’s order and related to a vital issue. If there had not been such a relation, the charge could be considered as passable; see Lucas v. The Queen[17], at p. 10.

I think it important, in concluding these reasons, that I stress two points. The first concerns an understanding of what is a vital issue where a suspect has given contradictory statements under oath in testifying for the Crown against the accused so as to call for a caution to the jury against accepting inculpating evidence given by that witness. The requirement of such a caution is not limited to cases where the witness has given direct evidence. Where, as here, the evidence

[Page 1098]

against the accused is circumstantial only (and, indeed, fragile), it is wrong to regard vital issues as being ultimate ones only, such as an issue of identity, and to absolve the trial judge of giving the caution merely because the Crown witness did not testify directly as to identity. The witness could not perforce do so when the entire case against the accused is built upon circumstantial evidence. Binet was a case where there was evidence of an eye witness, but this cannot mean that a vital issue must be defined only in terms of direct evidence so as to call in such cases only for a caution against accepting the evidence of a witness who has given inconsistent statements under oath.

Where substantially the only evidence against an accused is that of witnesses who are themselves suspect, or have interests of their own to serve, and it is circumstantial only, vital issues must be viewed in the light of these facts and hence must be taken to relate to evidence that establishes a crucial link in the chain of circumstances upon which the allegation of culpability of the accused rests. The Lucas case does not conflict with this view when regard is had to its quite different facts, especially to the fact that the case against the accused did not depend (as it does here) on the evidence of a questionable Crown witness.

The second point that deserves emphasis relates to the breach of the trial judge’s order against out of Court exchanges between witnesses who have not yet testified or have not yet concluded their testimony. It is true that the trial judge, when making the order, expressed doubt that all witnesses or potential witnesses had heard it. Penoffio was testifying when the order was made, and whether or not Matthews was aware of it is, in my view, irrelevant to the damaging effect of Matthews’ change of testimony when prompted by Penoffio during the night recess before he, Matthews, returned to the witness stand the next day. The deliberate breach of the order by Penoffio required more of the trial judge than a charge that the matter was one of credibility only, as if Pen-

[Page 1099]

offio and Matthews were impartial and uncontaminated witnesses.

Moreover, Matthews did not say that he was unaware of the trial judge’s order. He was asked this on cross-examination:

You had heard the instructions given by his Lordship that you should not speak to anyone with respect to this case after you left this Courtroom and before I cross‑examined you?

and his answer was:

It was kind of confusing afterward, I thought.

When this is coupled with the fact that his change of testimony resulted in his giving Penoffio’s recollection and not his own on an issue as to the whereabouts of the accused and of Penoffio prior to the murder, a vital issue in the chain of circumstances, the need for a strong caution to the jury with respect to the acceptance of Matthews’ evidence is to me unquestionable.

In fine, this case exhibits three circumstances which, if not separately, then certainly in their cumulative effect called for a strong caution by the trial judge to the jury against accepting the evidence upon which the Crown’s case was founded. There was, first, the self-interest of Penoffio and Matthews in exculpating themselves at the expense of the accused; second, the inconsistent, contradictory statements under oath on vital issues; and third, the breach of the trial judge’s explicit order against out of Court exchanges involving witnesses who had yet to testify or who had not concluded their testimony. The caution was especially necessary when there was no independent evidence corroborative of the testimony of the tainted witnesses.

In view of the serious defects in the charge to the jury, I would allow the appeal, set aside the conviction and direct a new trial.

[Page 1100]

The judgment of Martland, Ritchie, Pigeon, Beetz and Pratte JJ. was delivered by

RITCHIE J.—This is an appeal from a judgment of the Court of Appeal for the Province of Ontario dismissing the appellant’s appeal from his conviction at trial before Mr. Justice O’Leary sitting with a jury on a charge of having murdered one Maurice Rodriguez.

The Court of Appeal did not find it necessary to give any recorded reasons for its decision which was a unanimous one and which is now appealed to this Court on the limited grounds stipulated in the order granting leave to appeal.

I have had the privilege of reading the reasons for judgment prepared for delivery by the Chief Justice in which he has reviewed the evidence in such detail as to make it unnecessary for me to do more than recapitulate the salient facts in skeletal form.

The victim of this murder was shot while in a parking lot separating the Town Manor hotel from the Elmar hotel in the City of Hamilton at approximately 11:50 p.m. on the 21st of September, 1974. At about this time the sound of a loud noise, like a shot or backfire, attracted the attention of a passerby to the parking lot where he saw a man running to and entering a brown or copper-coloured van with lettering on its side which he later identified from a photograph as bearing the name of “Barber Appliance Services”. On entering the parking lot, this witness found the body of the victim lying on the ground with a gun shot wound in his neck which proved to be a fatal injury. The van had been parked about five feet from where the body was found and a man was seen running towards it and entering the driver’s side although leaving the door ajar. Two other witnesses testified that they had observed the van leaving the parking lot and one of them was able to take what he thought was the number of the licence plate. I agree with the Chief Justice that the driver of the van at this time was undoubtedly the murderer and I feel that the identity of this individual is the pivotal or vital question in the determination of this case. In this regard the evidence is indeed

[Page 1101]

inconsistent with any other rational conclusion than that the perpetrator of the crime was the man who ran from the body and entered the marked van.

I am also satisfied from the evidence that the van was as its label indicates the property of Barber Appliance Services, a firm with headquarters at Brantford and that on the evening in question, it was in the custody of one Penoffio, who was the Hamilton manager of that firm and who had left his own vehicle in the hands of the owner of the firm in exchange for the use of the van.

Penoffio lived in a house at 87 Augusta Street in Hamilton which was also occupied by a friend named Matthews and where the accused was visiting on the evening in question. It was in the parking area adjacent to Penoffio’s house that the brown van was found by the police shortly after the shooting, and at that time the house was occupied by the accused, Penoffio and one Matthews who appears to have stayed in the house all evening listening to the television, drinking beer and sleeping. Penoffio and the accused gave different versions of their activities during the evening and these are accurately summarized in the reasons for judgment of the Chief Justice. Both men were obviously potential suspects, but neither told a story putting the other in the Elmar hotel parking lot at the fatal hour.

The Crown’s case depends almost entirely on circumstantial evidence although in so far as the statements of the accused and Penoffio are self-serving, they are entirely dependent upon the jury’s assessment of their respective credibility.

The trial judge gave full and adequate instructions on the manner in which circumstantial evidence should be treated by the jury and it is not for a second Court of Appeal to enter upon the task of reassessing the credibility of witnesses. I am satisfied that this is a case where the jury’s verdict must stand unless it can be shown that

[Page 1102]

there was error on the part of the trial judge in the instructions which he gave in his charge.

The order granting leave to appeal to this Court limits our considerations to but two such alleged errors in the judge’s charge. It appears to me to be convenient to consider them separately.

The first alleged error was “in failing to caution the jury on the unreliability of the witness Matthews who changed his testimony after discussing it overnight with the witness, Penoffio, in breach of the trial judge’s direction forbidding any such discussion”.

In the course of the trial, the learned judge made an order directing that all witnesses should be excluded from the court room when they were not testifying and directing further that there should be no communication between one witness and another who had not completed his testimony.

It will be seen that the first question upon which leave to appeal was granted is concerned with the failure of the trial judge to caution the jury as to the unreliability of the witness Matthews arising in part from his breach of the judge’s direction forbidding discussions between witnesses and in part from the subsequent change of his evidence.

In my respectful opinion, this question can only be answered in the affirmative on the ground that the witness Matthews knew or ought to have known that the judge had given the direction therein referred to and if it can be made to appear that the change in his testimony reflected directly on the guilt or innocence of the accused, i.e. on the question of his whereabouts at approximately 11:50 p.m. on the evening in question.

The trial judge’s order containing a direction referred to in this question was made while Penoffio was still giving evidence, but his own statement makes it more than doubtful whether the witness Matthews was within earshot when that order was made. In this regard, the learned judge said at the conclusion of Matthews’ evidence:

[Page 1103]

The words that passed between this witness and a prior witness would be in contravention of the order made excluding witnesses, the general order, but unfortunately I do not know whether the witnesses heard that original order. They were already absent from the court room and I do not know whether there is any loud speaker that relays that order to the witnesses. It is perhaps a failing of communication.

Counsel for the defence then observed that “Penoffio knew he shouldn’t be speaking to anybody that still has to give evidence” and the trial judge then observed:

All the witnesses, we presume, know that, but I wonder has a message gotten through to them. There is the order excluding witnesses and that contains the further admonition not to discuss the evidence, but it might be wise if a written copy of the order could be placed in the witness room so the witnesses will be aware that there is a direct order that they are not to discuss the evidence with someone who is still testifying or still has not testified, so I am going to ask the Clerk to make sure that a copy of the order that was made be placed in the witness room because I know most of the witnesses, if not all of them, were absent from the court room when the order was made. (The italics are my own.)

In view of this circumstance, I do not think that the trial judge can be said to have been in error in failing to instruct the jury as to the unreliability of Matthews’ evidence on account of his having acted in breach of an order, the existence of which may well have been unknown to him.

The change made in Matthews’ evidence after his discussion with Penoffio is accurately summarized in the reasons for judgment by the Chief Justice where he says:

Matthews’ first version that he could not remember seeing Penoffio in the house at 10.30 p.m. when he, Matthews, went to bed was consistent with the accused’s evidence on the point. Matthews’ change of testimony resulted in the corroborating of Penoffio and contradicting the accused.

In my opinion, this change had no direct bearing on the whereabouts of Maxwell at the time when the murder was committed, its sole effect being to

[Page 1104]

support the evidence given by Penoffio as against that of the accused.

For all the above reasons and with the greatest respect for those who may entertain a different view, I am of opinion that the first question posed in the order granting leave should be answered in the negative and that there was no need for a further caution as to Matthews’ testimony which was adequately summarized by the trial judge when he said:

So the change in his evidence was from not remembering whether he had seen Penoffio to saying that he saw him for about 15 minutes between 10:00 and 10:30 and he went on to say that Dwight Maxwell was there when he was talking to Penoffio in that fifteen minute period or, when he saw him for 15 minutes...

In so far as the changed version of Matthews’ evidence placed Maxwell and Penoffio in and around 87 Augusta Street at between 10:00 and 10:30 p.m., it is corroborated by the evidence of Janice Brady who saw the two men together outside the house at about 10:20 p.m. I cannot think that a change from saying “I cannot remember” to saying “I have now been reminded” gives rise to such an inconsistency as to require a special caution by the trial judge in addressing the jury.

The second alleged error specified in the order granting leave to appeal is, however, concerned with the evidence of Penoffio and raises the question of whether the learned trial judge was in error in “failing to caution the jury on the unreliability of the evidence of Penoffio who, inter alia, had made previous contradictory statements under oath”.

There is no doubt that Penoffio’s evidence and his conduct showed him to be an undesirable and untrustworthy character. In the first place, Penoffio undoubtedly heard the order containing the judge’s direction not to communicate with witnesses and he deliberately disobeyed. Furthermore, he gave contradictory evidence under oath in the manner outlined in the reasons for judgment of the Chief Justice, and lastly, he showed his distaste for this whole proceeding by absenting himself from the first trial of Maxwell making it necessary to

[Page 1105]

issue a warrant for his arrest before he appeared at the second trial.

All these matters undoubtedly indicate that Penoffio was unreliable but this is not a case like Binet v. The Queen[18] where a witness had sworn an affidavit before trial to the effect that the accused was in no way involved in the crime of assault with which he was charged and then appeared at the trial and having been sworn gave evidence that the accused was a direct participant. The evidence at trial was not given until after the witness had himself been convicted of the offence with which both had been charged and tried separately. There can be no doubt that in the Binet case the witness was an accomplice and a perjurer and it was on this ground that this Court found “that the learned trial judge failed properly to instruct the jury on the great danger of accepting the evidence of an admittedly perjured accomplice on a vital issue”. The distinction between that case and the present is that there was no suggestion that Penoffio was an accomplice, it being on the contrary urged that he was a suspect. Furthermore, there is here, in my opinion, only one “vital issue” which concerns the identity of the driver of the van at the time of the murder and upon this question Penoffio gave no evidence whatever, and while it can be said that he perjured himself, his perjury was not concerned with the single question which lies at the root of this whole case.

The circumstances here bear a close similarity to those which Chief Justice Kerwin was dealing with in the case of Lucas v. The Queen[19] as appears from the following passage which has already been quoted in the reasons for judgment of the present Chief Justice:

I reject the submission of counsel for the appellant that it was error in law on the part of the learned trial Judge not to charge in accordance with Binet v. The Queen, [1954] S.C.R. 52, to the effect that there is great danger in accepting the evidence of an admittedly per-

[Page 1106]

jured accomplice on a vital issue. Thomas was not an accomplice. There was a mass of evidence pointing to the accused which makes it impossible to find, as counsel submitted, that the evidence of Thomas was on a vital issue, and finally, there was much evidence in the record which did corroborate Thomas. The matter was left to the jury as one of weight and credibility and, in my opinion, correctly so.

The case of Rustad v. The Queen[20] is to the same effect. It is contended that these cases are distinguishable from the present one on the ground that in the Lucas case it was found that the question of evidence did not relate to a vital issue, but it will be apparent from what I have already said that I consider it to be plain that the identity of the man who entered the van in the hotel parking lot is the only vital issue in this case and, as I am satisfied that the evidence of Penoffio did not relate either directly or indirectly to this question, I am of opinion that the learned trial judge was guilty of no error in failing to give more instruction than he did in respect to that witness.

As I have indicated, this case must turn in great measure on circumstantial evidence and in this regard it appears to me that one of the most striking circumstances is that the accused himself stated that he had picked up the key of the van when waiting for Penoffio in his house at about 10:00 p.m., whereas Penoffio states that he gave the key to the accused when they were together in a tavern a short time later. Either story would mean that the accused was in possession of the key enabling him to drive the van some time before the murder took place, and there is no evidence that the key was in anybody else’s possession until the police arrived at Augusta Street shortly after the killing when it was found to be in the accused’s pocket. The accused’s evidence was that he did not leave the house from slightly before 11:00 p.m. until after the police arrived. If this were true, it would mean that he was not the driver of the van when it left the parking lot at 11:50 p.m. and therefore not the man who shot Rodriguez, but it

[Page 1107]

is obvious that the jury did not believe him and, as I have said, I am not prepared to substitute my view of this witness’ evidence for that of the jury.

For all these reasons I would dismiss this appeal.

Appeal dismissed, LASKIN C.J. and SPENCE, DICKSON and ESTEY JJ. dissenting.

Solicitors for the appellant: Ruby & Edwardh, Toronto.

Solicitor for the respondent: The Attorney General for Ontario, Tforonto.