I. INTRODUCTION

A standard jury instruction, CALCRIM No. 201, tells jurors not to use the internet in connection with the case on which they are sitting. But we live in an age in which information is as available as air. Jurors can look up aspects of a case when they get outside of the courtroom with just a few taps on their mobile phones. So we may have to give some thought to bolstering the instruction. In the case at hand the judge did everything usually done, including giving the CALCRIM instruction, and still had to deal with a difficult juror misconduct issue. The case involved a prosecution against the CEO of an investment firm for defrauding an investor named Kim. The jurors wondered about the role of the firm’s accountant in the fraud, and in fact the CEO’s defense was that it was the accountant, not him, who was guilty. During deliberations, one juror, the foreperson, looked up the accountant on the internet, and thought she had discovered a legal proceeding showing the accountant and the defendant CEO had defrauded an investor named Kim and that the accountant had been “convicted” of it, but somehow the case got waylaid because of some timing issue. Needless to say, this was information that did not come out at the trial. The internet reference was almost certainly this court’s opinion in Sarsenstone Corp. v. Jewelinski (June 28, 2012, G044543) [nonpub. opn.]. That was a civil case that came up on demurrer, so its statement of facts assumed the allegations of the complaint were true, and those allegations were certainly not kind to the defendant. The juror’s misconduct in finding that information came to light after the jury had returned its verdict. The trial judge conducted a series of interviews of all the jurors – culminating in interviewing the errant foreperson – and concluded there was no actual prejudice from the juror’s misconduct, so he denied the defense’s new trial motion based on that misconduct. But what the juror herself said she had found – the extraneous information to which she exposed herself – was inherently and substantially likely to

2 have prejudiced her. The juror said she learned the defendant had committed fraud against Kim, and it was Kim’s financial losses that were the very basis of the case. In cases of juror misconduct by receiving extraneous material, prejudice is presumed. If the material, “judged objectively, is inherently and substantially likely to have influenced the juror,” the prejudice cannot be rebutted. It is a per se reversal. (In re Carpenter (1995) 9 Cal.4th 634, 653.) So we must reverse the judgment of conviction and remand the matter for a new trial.

II. FACTS

In 2005, defendant Gregory Fernandez was the CEO and owner of a Yorba Linda based investment firm known as Old Canal Financial. At its height, Old Canal had about 140 employees. Old Canal’s basic investment strategy was to buy up portfolios of “underperforming” mortgages, and work magic on them by somehow cajoling the borrowers to begin to make some payments. When successful, the newly-generated income stream would cause the value of the portfolios to rise and Old Canal would then sell them. In 2009, a felony complaint was filed against Fernandez for defrauding a single wealthy investor, Jennifer Kim, by charging her large commissions on two investment pools without telling her. The alleged fraud happened the mid-2000’s, during a roughly three-week period in the autumn of 2005 when the economy was otherwise going great guns. Even so, Old Canal was in financial trouble and allegedly used new money from investors like Kim to pay its own operating expenses and line Fernandez’s pockets. Ultimately Fernandez was brought to trial in the spring of 2013 on four counts. Count one was for garden-variety grand theft (Pen. Code, § 487, subd. (a)), count two was for fraudulently selling unqualified securities (Corp. Code, § 25401), count three was for selling unqualified securities without an exemption (Corp. Code, § 25110) and count four was for scheming to defraud in connection with the sale of

3 securities (Corp. Code, § 25541). Fernandez’s defense consisted in part of blaming Old Canal’s financial officer and accountant, John Jewelinski, for the loss of Kim’s money. For example, in his opening statement, defense counsel argued Fernandez had “no control over the money that came into Old Canal,” and asserted that Jewelinski was stealing from Old Canal. Fernandez’s counsel made a point of telling jurors it was actually Fernandez who reported Jewelinski to local police, which caused the police to contact the dissatisfied investor Kim, and it was Kim who caused the police to turn on Fernandez. Fernandez also took the stand himself during the trial. He testified that he didn’t handle the money that came into Old Canal, Jewelinski did, and that Jewelinski specifically handled the money that came in from Kim. Fernandez also said Jewelinski had taken the opportunity, while Fernandez was on vacation, to forge Fernandez’s signature, and when Fernandez got back he learned that Jewelinski embezzled $2 million from the firm. The jury hung on counts one (grand theft) and four (scheming to defraud), but returned guilty verdicts on counts two (fraudulently selling unqualified securities) and three (selling unqualified securities without an exemption). During deliberations, there was some discussion among the jurors as to whether Fernandez and Jewelinski were “in it together” or whether Jewelinski was the lone culprit. The foreperson, Juror No. 11, a high school biology teacher who had just come off a year and a half as a research scientist, (her later words to the trial judge), decided to Google “Jewelinski” because there had been speculation by the jurors about Jewelinski’s role, if any, in swindling Kim. Juror No. 11 did not mention her extracurricular research until after the jury had been discharged, and she was sharing an elevator with at least two other members of the jury, Juror Nos. 10 and 5. The mention of the Googling to Juror No. 10 caused Juror No. 10 to write a letter to the trial judge. We reproduce the letter in the margin.1 At that

1 Here is the entire letter verbatim:

4 point the trial judge decided to call in all the jurors seriatim (though Juror No. 6 could only be reached telephonically), finishing up with Juror No. 11. Juror No. 10’s interview is noteworthy because it confirms what her letter said. When asked what Juror No. 11 had told her she had found on the internet, Juror No. 10 replied: “That Mr. Fernandez and Mr. Jewelinski went on it [sic] together, and that they had taken the money from Ms. Kim. That Mr. Jewelinski had been tried, and I believe had been convicted, but it was overturned having something to do with the time frame of when he was, when he was arrested. And that was all I kind of got.” (Italics added.) One by one, the trial judge interviewed all the jurors, ending with Juror No. 11. By the time he got to Juror No. 11, the judge had made up his mind not to hold Juror No. 11 in contempt because the interviews were clear that Juror No. 11 had kept all her information to herself. In her interview, Juror No. 11 admitted she had indeed Googled Jewelinski. She specifically said she used the search engine Google and there were “two things that came up.” Here is the colloquy with the court about what she found: “The Court: That was going to be my next question. What information, if any, did you find?

“This letter is to inform you of possible Jury Misconduct in the above referenced trial. I am Jury [sic] #10 and left the court room yesterday feeling confident that the 11 jury members and I in the above referenced case did the best job that we were able to do and had came [sic] back with correct verdicts. [¶] After we were dismissed, several of us entered the elevator and someone said, I can’t wait to go home and search this case. Another one stated that he already was on his phone searching. Then Jury [sic] #11 stated that she was so frustrated on Friday evening (the day we started deliberation) that she had gone home and searched the case but it had not influenced her decision in any way. We all got out on the 3rd floor to check in with Jury Services. When we returned to the elevator one jury [sic] asked her what she had found out. At this point I was fairly stunned and not sure of what I should do so I was not intently listening to her answer. The main points that Jury [sic] #11 stated were that Jewelinski and Fernandez where [sic] in on it together, they committed fraud on Kim, Jewelinski was tried and convicted (I think) but it was overturned due to a statute issue and that they were tried together at one point but not sure what happened. That was all I really got. We got out of the elevator, I said goodbye and left the group. [¶] I cannot say that her views influenced any of the jury members, but I also cannot say that they did not. She was as firm in her believes [sic] that Fraud was committed as I was in mine that Fraud was not committed. ‘Was Fraud Committed’ was the issue that we could not agree upon. We were truly divided 6 and 6 with too strong of feelings on both sides to overcome. [¶] I have thought about this all night and felt that this was true misconduct by a fellow juror and I was obligated to inform you of the facts as I recall them.” (Italics added.)

5 “Juror 11: I had found that Mr. Jewelinski had been sued over something that had nothing to do with Old Canal Financial. “The Court: Was that a – to your belief was that a criminal case or civil case? “Juror 11: I don’t even remember the gist of it. It was with Sarenstone.” Juror No. 11 also admitted to checking out Fernandez’s connection to Jewelinski on “LinkedIn” a kind of professional Facebook. She found a flow chart showing they were connected on LinkedIn. The trial judge did not inquire as to the precise nature of the suit that Juror No. 11 had found on Google, and there is no indication that Fernandez’s retained counsel urged him to explore the matter. Nor did the trial judge explore why Juror No. 10 characterized Juror No. 11’s own words as amounting to finding information that Fernandez had committed fraud on Kim. Fernandez brought a new trial motion on the heels of the juror interviews, based on the misconduct of Juror No. 11. As against the new trial motion, the prosecutor argued there had been no actual prejudice because Juror No. 11 never shared her research with anybody else. The trial judge denied the new trial motion, reasoning that the “effect of the misconduct [was] minimal.” He pointed out that Fernandez himself had emphasized his relationship with Jewelinski, a relationship that was “significantly stronger relationship with Mr. Jewelinski than what Juror # 11 discovered during her research[.]” Thus he concluded: “That the two of them could be in on it together is an interpretation of the evidence and is not based on any research. It is the Court’s conclusion that it is not substantially likely the outside information affected the verdict and that it is not substantially likely that Juror # 11 was actually biased against the defendant.”

6

III. DISCUSSION

If a trial court grants a new trial motion based on juror misconduct, that decision gets the benefit of the usual deferential standard of review which grants of new trial motions generally receive. (See People v. Ault (2004) 33 Cal.4th 1250, 1266.) But the rules are tighter for denials of new trial motions based on jury misconduct brought to light after the verdict and jury discharge, i.e., past the point of correctable return. In that situation, the operative test was laid down in In re Carpenter, supra, 9 Cal.4th 634, 653, and later restated in In re Lucas (2004) 33 Cal.4th 682, 696-697. The test is this: If there is reception of extraneous matter by a juror, then there is a presumption of prejudice that the prosecution must rebut. To rebut the presumption, the prosecution must overcome both of two, independent hurdles: First, the extraneous matter cannot be of such a damning nature that it objectively and inherently would influence the juror exposed to it, and, second, whatever the nature of the material, it must not have actually prejudiced that juror.2 The reason there are two facets to the test is to simplify the trial court’s inquiry and to promote certainty that improper evidence is not received. The court must first look at whether the evidence is of a type that would be inherently and substantially prejudicial. If it is, the court does not get to the second inquiry about actual prejudice in the case. The idea is that we do not want judges to have to be mind-readers to resolve these issues. If the evidence is of a type that experience has taught us to be

2 Here is the exact language from Lucas: “We have explained that the following standards govern our review of claims that jurors have been exposed to extraneous evidence: ‘To summarize, when misconduct involves the receipt of information from extraneous sources, the effect of such receipt is judged by a review of the entire record, and may be found to be nonprejudicial. The verdict will be set aside only if there appears a substantial likelihood of juror bias. Such bias can appear in two different ways. First, we will find bias if the extraneous material, judged objectively, is inherently and substantially likely to have influenced the juror. [Citations.] Second, we look to the nature of the misconduct and the surrounding circumstances to determine whether it is substantially likely the juror was actually biased against the defendant. [Citation.] The judgment must be set aside if the court finds prejudice under either test.’ (In re Carpenter [, supra,] 9 Cal.4th [at p.] 653; see also People v. Danks (2003) 32 Cal.4th 269, 302–303.)” (In re Lucas, supra, 33 Cal.4th at pp. 696-697.)

7 inherently prejudicial in a substantial way, the judge need not engage in the formidable task of divining its impact on the psyche of a juror in the particular case. Here the judge took the more pragmatic and intuitive approach and cut straight to the issue of whether Juror No. 11 was in fact prejudiced, and did not address the problem of whether the extraneous information to which she was exposed was objectively and inherently likely to have influenced her. The Attorney General’s office makes a roughly parallel mistake, by simply ignoring those aspects of the record showing that Juror No. 11 learned Fernandez committed fraud against Kim, and, on top of that, Jewelinski had already been convicted of it for his part.3 According to the Attorney General’s office, Juror No. 11 only learned (a) that Fernandez and Jewelinski were connected on LinkedIn and (b) Jewelinski had been sued over something that had nothing to do with “Sorenstone/Sarsenstone.” Thus, the office argues, as the trial judge himself found, the information learned by Juror No. 11 regarding the Fernandez-Jewelinski link was merely cumulative of a relationship that was already well established by the evidence – after all, Jewelinski was the accountant of a firm owned by Fernandez, and Fernandez himself had said Jewelinski was his childhood friend. What is omitted from the Attorney General’s response is that Juror No. 11 herself talked about more than just the Fernandez-Jewelinski relationship. She used the words fraud and “convicted” (or “guilty”) to describe to Juror No. 10 what she learned. The fraud statement went directly to Fernandez’s relationship to the victim Kim, not just his relationship with his sidekick Jewelinski. And the “convicted” statement went indirectly to Fernandez’s relationship with Kim, because – as several jurors mentioned in

3 Our syntax recognizes the hearsay nature of the information – what Juror No. 10 said about what Juror No. 11 said. But in the nature of things that’s the only way such juror misconduct will be revealed. The core problem is one of methodology, and no doubt has bedeviled courts in other cases. In this case, Juror No. 11 had enough warning to selectively forget in her interview precisely what she had seen about the defendant Fernandez on the internet.

8 their interviews – the jury did indeed wonder whether Fernandez and Jewelinski were “in it” together or whether Fernandez was to be believed in asserting it was all Jewelinski’s doing. What Juror No. 10 said was that Juror No. 11 reported they “went in together.” As our high court explained in People v. Nesler (1997) 16 Cal.4th 561, 582, the question of prejudice from juror misconduct “is a mixed question of law and fact subject to an appellate court’s independent determination.” Moreover, reviewing courts must look to the “entire record to resolve the issue, keeping in mind that the trial court has found the relevant historical facts and resolved the conflicting evidence, but that the question of prejudice is for our independent determination.” (Id. at p. 583.) People v. Holloway (1990) 50 Cal.3d 1098 (Holloway) is on point and leaves us no room to do other than reverse Fernandez’s conviction. Holloway involved the murders of two sisters in Sacramento. During the guilt phase (the second day of trial in fact), a juror read a newspaper article stating the defendant was on parole after having served time for assaulting a Sacramento woman with a hammer. As in the case before us, the juror kept the information to himself. (Holloway, supra, 50 Cal.3d at pp. 1331 & 1334.) Also as in the case before us, the information came out after the verdict had been returned. The case came to the Supreme Court after the trial court denied the defendant’s new trial motion. The Holloway court quickly noted that it had been established since 1886 that jurors are not to receive information about a case from newspapers. (See People v. McCoy (1886) 71 Cal. 395, 397.) The high court noted the misconduct created a presumption of prejudice and then segued into whether the prejudice could be rebutted. It was in that context that the Holloway court introduced what would become the “objectively and inherently likely to prejudice” element into juror misconduct analysis, stating that: “‘In effect, the court must examine the extrajudicial material and then judge whether it is inherently likely to have influenced the juror.’” (Holloway, supra, at p.

9 1109, quoting 2 ABA Standards for Criminal Justice (2d ed. 1980) std. 8-3.7, Commentary, p. 8.58.) The Holloway court then held the particular newspaper article which the single juror had read so prejudicial the presumption of prejudice had not been rebutted. (Holloway, supra, 50 Cal.3d at pp. 1111-1112.) In doing so, the high court cited with approval two prior appellate opinions, People v. Andrews (1983) 149 Cal.App.3d 358 and People v. Thomas (1975) 47 Cal.App.3d 178, which had also reversed convictions because of the “extremely prejudicial” content of newspaper articles read by jurors in those cases. (See Holloway, supra, 50 Cal.3d at pp. 1110-1111.) Here, as in Holloway, it makes no difference that the errant juror kept her research to herself during jury deliberations. As the high court wrote in Holloway: “Defendant was entitled to be tried by 12, not 11, impartial and unprejudiced jurors. ‘Because a defendant charged with crime has a right to the unanimous verdict of 12 impartial jurors [citation], it is settled that a conviction cannot stand if even a single juror has been improperly influenced.’ [Citations.]” (Holloway, supra, 50 Cal.3d at p. 1112.) Here, Juror No. 11 herself used words in the elevator to another juror that indicated the site she visited showed both Fernandez and Jewelinski defrauded the victim Kim, and Jewelinski was convicted of it. The conviction was overturned on the technicality of a time limit but that would not change the “facts” related by Juror No. 11. The ineluctable conclusion is that whatever Juror No. 11 found, it was objectively and inherently likely to have influenced her – especially reinforced by the fact Juror No. 11 indicated to Juror No. 10 that Jewelinski had gotten off on a technicality (the running of a statute of limitations).4

4 While common sense would dictate to a virtual certainty that what Juror No. 11 discovered was the unpublished Sarsenstone opinion, we do not need to take the step of actually basing our decision on that fact. Whatever Juror No. 11 found was so bad that Juror No. 11 characterized its contents as Fernandez having defrauded Kim and Jewelinski’s having been convicted of that very fraud. Accordingly we deny Fernandez’s appellate-level request to take judicial notice of the Sarsenstone opinion, since it was not put in front of the judge at the trial level.

10 We are also instructed by the two appellate decisions that Holloway cited with approval. In Andrews, the information contained in the newspaper article didn’t directly link the defendant to the crime, it only linked his wife to the crime; the defendant’s link was to other felony charges. Here, Juror No. 11 told Juror No. 10 that both Fernandez and Jewelinski had “defrauded” Kim, i.e., Fernandez was guilty of the very charges the jury was considering. Likewise, in Thomas the newspaper story did not link the defendant directly to the crime being considered, only a codefendant. Here the link was direct to defendant Fernandez. Juror No. 11 said that both Fernandez and Jewelinski had defrauded Kim. This information was so inherently prejudicial and so substantial in deciding whether Fernandez had committed fraud and securities violations in this case that it precludes resort to the second prong of the Lucas test.

IV. CONCLUSION

The judgment is reversed, and the cause remanded for a new trial in accordance with the views expressed herein.

BEDSWORTH, ACTING P. J.

WE CONCUR:

FYBEL, J.

THOMPSON, J.

11