Health Genetic Center Corp. o/a Health Genetics Center v. New Scientist Magazine, 2018 ONSC 7224

COURT FILE NO.: CV-11-00420193-0000

DATE: 20181207

ONTARIO

SUPERIOR COURT OF JUSTICE

BETWEEN:

HEALTH GENETIC CENTER CORP. o/a HEALTH GENETICS CENTER and DR. YURI MELEKHOVETS

Plaintiffs

– and –

NEW SCIENTIST MAGAZINE, PETER ALDHOUS, and REED BUSINESS INFORMATION LTD.

Defendants

Ioulia (Julia) Melekhovets and Mark Donald, for the Plaintiffs

Sandra Barton and Erin Farrell, for the Defendants

HEARD: March 19, 20, 21, 22, 23, 26, 27, 28 and 29, 2018, April 3, 4, 5 and 9, 2018

LEDERER J.

[1] The plaintiffs, Health Genetic Center Corp. and Dr. Yuri Melekhovets, claim that an article written by the defendant Peter Aldhous defamed them. The article appeared in New Scientist Magazine, which is owned and published by the defendant, Reed Business Information Ltd.

BACKGROUND

[2] In the present day, women who are pregnant and their partner or partners may wish to know or confirm the identity of the father. Science has developed tests which can assist in answering this question. The tests require obtaining samples of the DNA of the mother, the prospective father and, when the inquiry is made before birth, the fetus. For the parents, this is easy. A swab of the cheek will do. For the fetus it is more problematic. Conventionally, this was done using one of two procedures. The first is amniocentesis, the retrieval of amniotic fluid by inserting a needle through the mother’s abdominal wall. The second is CVS, the retrieval of placental tissue by inserting a device through the cervix. While these methods support a reliable means for conducting prenatal paternity testing, they are invasive and carry a risk, albeit a small one, of miscarriage.

[3] Clearly, it would be better if means could be found to obtain fetal DNA in a non-invasive fashion.

[4] With proper DNA samples in hand it is possible to compare the DNA of the fetus (the child) with that of the mother and the alleged biological father. DNA is inherited equally from each parent. Half of the fetal DNA comes from the mother and the other half from the father. By comparing the DNA profiles of the mother and child, it is possible to determine which half of the fetal DNA was likely inherited from the mother. On this basis one can conclude that the remaining parts of the child’s DNA came from the father. With appropriate genetic markers identified and sufficient numbers of them available, a comparison of the fetal DNA to that of the prospective father can be made and the possible father excluded as the parent or found, with a high degree of statistical certainty, to be the biological father.

[5] The issues in this case are centred on the belief of the plaintiff, Dr. Yuri Melekhovets, that he had found the means and developed a procedure of obtaining fetal DNA from the blood of the mother, that is, without the risk of intruding directly on the fetus. Yuri Melekhovets believed that through this means DNA of the quantity and characteristics needed could be retrieved and duplicated such that a proper analysis directed to identifying the father could be completed. Yuri Melekhovets believes the test he used to be benign in its effect and reliable in its result.

[6] In or around the year 2001 the plaintiff began marketing the test, some directly, some through agents. It was a retail program. The sales were to, and payment was made by, the women and men who were affected. Problems appeared. Some of the individuals who had purchased and relied on the test began to question the results. In some instances more conventional tests were undertaken, which, rather than confirm those results were diametrically different. In short, those said to be the fathers were not and others who the test had excluded were. In some of the instances referred to in the evidence, this had dramatic and negative impact on the lives of those involved.

[7] The defendant, Peter Aldhous, became aware of the concerns; interested his editors in the issue; conducted, what the defendants refer to as an 18 month investigation; and wrote the article which was published in the magazine. Looked at in its overall context (as opposed to the individual words that the plaintiffs say are demonstrative of the defamation) the article is critical of the test and the plaintiff. It is titled: “The danger of unreliable paternity tests”. It reviews the experiences of some of the purchasers; refers to issues with the test that the defendants believed had been discovered; mentioned the plaintiff, Yuri Melekhovets by name and, among other things, notes:

The test is run on a sample of the woman’s blood and cheek swabs from the possible fathers. Our investigation suggests that the results are unreliable – with potential devastating consequences...

[8] The plaintiffs say that the article is defamatory of them, has caused them damage and that the defences that can apply to the allegations being made have no application that would save the defendants from liability.

[9] The breakthrough which marks the beginning of the events that led to this action became apparent in 1996 and 1997. In those years two academic papers were produced by Y.M. Dennis Lo and various of his colleagues. The first (1996) was entitled: “Two-Way Cell Traffic between Mother and Fetus: Biologic and Clinical Implications”[^1] and the second: “Presence of fetal DNA in maternal plasma and serum”.[^2] The summary of the second of these papers is instructive as to the relevance of its content:

… We investigated the equivalent condition in pregnancy – that is, whether fetal DNA is present in maternal plasma and serum.

[10] Under the heading “Discussion” it begins:

Our results show that fetal DNA is present in maternal plasma and serum. Use of maternal plasma or serum for the detection of fetal DNA for non-invasive prenatal diagnosis may therefore be possible…

[11] This attracted the attention of Yuri Melekhovets. In the affidavit which was his evidence-in-chief at trial, he deposed:

I based my research on numerous articles which were already published regarding the extraction of fetal DNA from the mother.

[12] After noting the two articles by Y.M. Dennis Lo, he went on:

Both of these articles were very influential in the development of H.G.C.’s prenatal paternity test done on maternal blood.[^3]

[13] He was not alone. Others pursued the idea of prenatal, non- invasive, fetal DNA retrieval. During the course of being cross-examined the defendant, Peter Aldhous, alluded to this:

I guess the other thing that is in the article, and I don’t know whether it will be called a conclusion, is there is a sentence in there something to the effect that talking about that there is material in the article that relates to the difficulty researchers have had of developing tests of this sort…[^4]

[14] Dr. Gordon Lauc was and may still be a Professor of Molecular Biology and Biochemistry in the Faculty of Biochemistry at the University of Zagreb. He is a specialist in working with difficult and mixed DNA samples. He has published work on DNA in maternal blood. Gordon Lauc attempted to extract fetal DNA, for both male and female fetuses, from 20 pregnant women.[^5] He did not consider the results achieved to be “sufficient for reliable paternity testing”.[^6] Gordon Lauc advised Peter Aldhous that other experts had tried to study fetal DNA in maternal blood for 20 years but had been unable to reliably discern the fetal material from maternal DNA and to amplify it for study or commercial application.[^7]

[15] Dr. Karl Reich operates a paternity testing company, Independent Forensics. He met with Peter Aldhous. An affidavit deposed to by Karl Reich records the substance of the meeting as he recalled it. Karl Reich explained that there was a great deal of interest in non-invasive prenatal paternity testing since the other methods of determining paternity prenatally were invasive and carried a small risk of unfavourable medical results, such as a miscarriage. Karl Reich had been personally involved in trying to develop a non-invasive prenatal paternity test. It would have been of limited use since it relied on comparing the Y chromosome of the fetus against the Y chromosome of the alleged father. Since only male fetuses carry the Y chromosome only half the commercial market would be available. Independent Forensics also looked at the process of developing a non-invasive prenatal test that identified circulating fetal cells using immunological markers. The test proved too difficult to design and would have taken more resources than it could afford. Independent Forensics decided not to continue with either approach.[^8]

[16] Dr. Ali Al-Salih is the operator and Laboratory Director of DNA Reference Laboratory. Its website showed that it offered a non-invasive prenatal paternity test based on maternal blood. Ali Al-Salih advised Peter Aldhous that DNA Reference Laboratory had stopped offering these tests approximately 2-3 years earlier because, in his view, the test had important limitations, and he was concerned about the reliability of its results. These limitations included the inability to distinguish maternal and fetal DNA contamination from prior pregnancies, and general unreliability.[^9]

[17] Dr. Denise Syndercombe Court provided an affidavit and appeared in court. Her qualifications are impressive. She is an academic who has published over 100 peer-reviewed research publications, runs an accredited laboratory undertaking relationship tests and does specialist DNA casework for organizations in both the United Kingdom and Europe:

I am a recognized expert in forensic science, including DNA profiling, which is used to compare DNA characteristics in related and unrelated individuals, and am consulted in civil and criminal cases on a regular basis.[^10]

[18] She was qualified as an expert in “forensic genetics, with particular expertise in relationship and paternity testing, DNA typing analysis and extraction, working with STR’s and SNP’s and sequencing, the use of PCR, the use of restriction enzymes and dealing with mixed integrated DNA samples.”[^11]

[19] In or about 2005 a woman presented herself at a genetic testing laboratory. She was seen by Denise Syndercombe Court. She was pregnant. She was a white woman married to a white man. She had slept with a man of a different race. The woman needed to know who the father was before the baby was born. She had had a non-invasive prenatal paternity test that was conducted on a maternal blood sample. She had purchased the test online. The test results she had received indicated that the man she had the affair with could not be excluded as the true biological father.

[20] I note that, as it was explained to the court during the course of the trial, in paternity testing, the conclusion that an alleged father “cannot be excluded” indicates that there is a very high degree of statistical certainty that the alleged father is, in fact, the true biological father. The evidence suggested that, in such circumstances, the probability of paternity is 99.9% as compared to an untested, random man of the North American population.

[21] Denise Syndercombe Court was unable to recall the name of the laboratory that had conducted the test but did remember that it was Canadian. During the course of his investigation Peter Aldhous spoke to a number of companies that advertised as providing non-invasive prenatal paternity testing. He learned that most, if not all, did not provide the tests but instead acted as brokers who purchased the tests from the plaintiffs. There was no evidence, nor was any alternative suggested, that the company that undertook the test was other than the plaintiff, Health Genetic Center Corp.

[22] Upon her review of the report made with respect to the test that had been undertaken, Denise Syndercombe Court found that the genetic loci that had been tested yielded a very low “combined paternity index”, or “probability of paternity”, and that the test was not a reliable basis on which to draw conclusions about paternity.[^12] She undertook a follow-up test and reached a different result as to the paternity of the child.[^13]

[23] During October 2008, Denise Syndercombe Court was contacted by another woman who had received results from a test done by the plaintiffs. Denise Syndercombe Court reviewed the results. She advised the woman that the test did not provide proof that was sufficient to demonstrate that the man it identified as the father was the father. A follow-up test using the conventional procedure was arranged. It demonstrated that, in fact, the man who had been identified was not the father.[^14]

[24] The conversations she had with these two women concerned Denise Syndercombe Court. She had read about the presence of circulating fetal DNA in maternal blood. She did not know of any reliable forensic or commercial uses that stemmed from this research. She was unaware that any well-regarded laboratory was offering these kinds of tests. Others, including the laboratory she was associated with, had attempted to apply the research but she understood from her own work that the research had encountered obstacles.[^15]

[25] For the two women, the non-invasive prenatal paternity tests had produced unreliable results. There could be other women making life-altering decisions (to have an abortion or not) on the basis of tests where the results could be questionable. In October 2009 Denise Syndercombe Court, with two colleagues, published an article about these cases and the erroneous results. The paper was presented at a conference. The significance of this is that it gave “peers [an opportunity] to question the scientist presenting the information at the time, [as] data would have been presented for them to consider”. The article was peer reviewed and was published in a peer reviewed academic journal, “Forensic Science International”.[^16]

[26] In early 2009 Denise Syndercombe Court had occasion to speak with Peter Aldhous. While their discussion was with respect to an unrelated issue, she told him about the circumstances concerning the two women. Denise Syndercombe Court expressed her view that the science of prenatal testing on maternal blood was “underdeveloped” and that the “results of [the two tests done by the plaintiff] were very likely to be inaccurate, and that the method should not have been used to determine paternity”. It was this conversation that piqued the interest of Peter Aldhous and was the catalyst for the investigation that he undertook, and the article that he wrote, which is the subject of this law suit.[^17]

[27] To this point in these reasons, it should be clear that the idea of a non-invasive prenatal DNA test based on maternal blood was a novel proposition, subject to a great deal of scientific interest and, at least in 2009, eight years after the plaintiff began marketing it, remained problematic. In October 2009, it was the subject of a peer reviewed article, the one written by Denise Syndercombe Court and her colleagues, published in Forensic Science International.

[28] I turn now to the plaintiff, Yuri Melekhovets. Counsel for the defendants was at some pains to demonstrate that Yuri Melkhovets did not have the background, training or expertise that would suggest he would be at the forefront of the advancement of the science necessary to develop a safe, replicable and reliable test to identify paternity using maternal blood as the source for fetal DNA. “After extensive post-secondary studies and receipt of [his] Phd in Molecular Biology and Genetics from Moscow State University [he] was invited to come to Canada and work as a research scientist at the University of Guelph.” Three years later he accepted a position at the University of Toronto (HIV Research Laboratory, Department of Microbiology and Medical Genetics). He worked on anti-HIV-1 gene therapy. Thereafter, he entered the private sphere. Since he was, in his own words “a highly respected molecular biologist” he was offered the position as the head of a DNA diagnostic laboratory. While there he implemented tests for the detection of different genes in humans and animals.[^18]

[29] In the affidavit deposed to by Yuri Melekhovets which was the foundation for his direct evidence at trial, he says that his research “led [him] to pioneer a method of non-invasive prenatal paternity testing by extracting fetal DNA from a mother’s blood sample”. He notes that he began offering this test through the corporate plaintiff in 2002.[^19]

[30] It is hard to understand, in any refined way, how the background and experience of Yuri Melekhovets could have provided him with the knowledge to develop the test in question, through all the detail, steps and stages required without the assistance and expertise of others. Certainly, there are aspects of the tests that were not part of his immediate experience.

[31] I point out a particular demonstration of this concern.

[32] It is the Y chromosome that separates men from women. Males have them; females do not. If the fetal DNA includes a Y chromosome the child will be a boy; if not the child will be a girl. Y chromosomes can only be passed from father to son. They are passed down as single units, not in pairs. Absent mutation, which is very rare, fathers and their sons share the same genetic coding on their Y chromosomes. This can go on from generation to generation, without change to the DNA sequence. Over time, this has resulted in large populations of males sharing much of the same genetic information on their Y chromosomes. Mutations do appear. These natural occurrences take place at random intervals over many generations. Geneticists have used population studies to identify patterns of Y chromosome variation. These patterns have been mapped out on an “evolutionary” or “phylogenetic” tree which illustrates how Y chromosomes have evolved over time. It should be obvious that for the purpose of paternity testing, where the fetus is male, understanding this history and the patterns represented on the phylogenetic tree would be important, if not critical. An individual male can only be a member of one branch. Each branch represents a “haplogroup”, so a male can only be a member of one haplogroup. Each haplogroup can be traced to a specific ethnic origin.[^20] Science has accepted the study of the ancestry of the Y chromosome as portrayed on the phylogenetic tree:

There is only one accepted model of Y chromosome inheritance.[^21]

…geneticists who study the Y chromosome have developed a single model that they use to trace the ancestry of the Y chromosome.[^22]

…all results should likewise compare favourably to known scientific standards such as Y-chromosome haplotypes…[^23]

The Y chromosome is uniquely positioned for evolutionary studies because of its unique biology. The history of a Y chromosome is recorded in its SNPs. Because of this history some SNP combinations are biologically impossible.[^24]

[33] Yuri Melekhovets agreed that he has never studied the ancestry of the Y chromosome and yet he denies the science established and accepted by those who have:

Q. And there are geneticists who specialize in studying the Y chromosome?

A. Very likely, yes.

Q. And particularly there are geneticists who specialize in studying the development or evolution of the Y chromosome?

A. Yes.

Q. And you have never done that, sir? You have never studied the evolution of the Y chromosome?

A. No, I never studied evolution of Y chromosome, no.

Q. And what the geneticists who study the Y chromosome have done is they have created what they have called a phylogenetic tree?

A. Yes, one of many.

Q. I understand that is your view, sir.

A. Yes.[^25]

[34] The refusal of Yuri Melekhovets to acknowledge this accepted science has manifested itself in the testing he has done. Dr. Peter Underhill was accepted as an expert in the Y chromosome including its ancestry and interpreting DNA data on the Y chromosome as well as population genetics related to the Y chromosome.[^26] He was asked to review work undertaken by Yuri Melekhovets. In respect of one of the cases he reviewed, he noted:

With respect to the [redacted] report from the plaintiffs’ lab, I could very clearly see that the SNP pattern reported was highly questionable. At that time, I had worked with many of the SNPs in the report so I was immediately able to see that the results were inconsistent. The alleged father and male fetus, according to the plaintiffs’ result, would put both the fetus and alleged father in three haplogroups at the same time. This means that the individuals were being placed on three branches of the phylogenic tree, and thus in multiple haplogroups, simultaneously.

In any event, it is not possible for one male to belong to all of these three different haplogroups.

he concluded:

Providing a paternity conclusion to a customer based on such obviously erroneous data indicates a lack of understanding of the significance of Y chromosome ancestry. It is surprising that a geneticist who presents himself as an expert in paternity testing: (a) does not understand Y chromosome ancestry and associated haplogroups; and (b) would not immediately appreciate the significance of having typed such a rare SNP.[^27]

[35] Denise Syndercombe-Court did a similar review of another case and came to a similar conclusion:

Five of the 11 SNPs I tested, for each of the fetus and alleged father, showed a result that differed from the plaintiffs’ result. The results of my testing indicated that Mr. [redacted] could not be excluded as the biological father. However, the results provided no statistical evidence on which conclusions about paternity could be drawn because all of the SNPs identified in the plaintiffs’ testing report were located on the Y chromosome. This specific combination of SNPs was incapable of distinguishing between Mr. [redacted], and his father or any other paternal relation. I also observed that, contrary to the profile generated by the plaintiffs, the fetal and alleged father profiles I generated, were consistent with the established ancestry of the Y chromosome.[^28]

[36] The failure of Yuri Melekhovets to work within accepted science, and to differ from those who were familiar with areas he was not, tends to confirm the expressed concerns for his expertise and training. Did he have the knowledge to do what he claims he had done?

[37] Dr. Bruce Budowle was retained, by the defendants, as an expert to provide evidence to the court in this action. Like Denise Syndercombe Court his credentials are impressive. In 1979 he received his PhD in Genetics from the Virginia Polytechnic Institute. His research interests are listed as Forensic Science, Genetic Marker Systems, Technique Development, Molecular Biology, Population Genetics, Human Genetics, Microbial Forensics and Pharmacogenomics. His curriculum vitae lists 658 publications on these topics. He is the Director of the Center for Human Identification in the Department of Molecular Biology, Immunology and Genetics at the University of North Texas Health Science Center. He has spent between 35 to 40 years studying, teachinggenetics and DNA analysis.[^29] From 1983 to 2009 he worked with the FBI. While there he held a variety of positions. He focused primarily on DNA-based techniques. In his work he has used “all of the techniques the Plaintiffs use in their non-invasive…prenatal paternity test.”[^30]

[38] Bruce Budowle reviewed the curriculum vitae of Yuri Melekhovets, the documents in the case, the examinations for discovery, the trial transcripts that preceded his appearance, and the relevant documents. He interviewed Yuri Melekhovets and visited the plaintiffs’ laboratory. From this review, Bruce Budowle expressed concern as to whether Yuri Melekhovets had the requisite training and understanding of the underlying science, methods and technology needed to tackle the task of producing a prenatal paternity test where the fetal DNA is sourced from maternal blood. At trial, Bruce Budowle testified that Yuri Melekhovets:

… underestimated the challenges and complexities of this kind of testing and did not appreciate what has to go in to get reliable results from these kinds of tests.[^31]

[39] I accept this conclusion.

[40] In his affidavit Yuri Melekhovets deposed that:

I have devoted my entire professional life to the advancement of science.[^32]

[41] It would have been better if Yuri Melekhovets had acted in a fashion consistent with this assertion. He has refused to follow the understood and accepted requirements of developing science and the methodology associated with this work. I will return to this theme as these reasons proceed.

DEFAMATION

[42] What about defamation and the law that governs it? Defamation is a civil wrong or a tort. It is founded on the idea that we, as individuals, are part of a society made up of reasonable people:

A statement is thus legally defamatory if it tends to lower the reputation of the plaintiff in his or her community in the estimation of reasonable persons.[^33]

[Emphasis added]

[43] Thus, for a publication to be “defamatory”, the court must be satisfied of three things:

(1) that the impugned words were defamatory, in the sense that they would tend to lower the plaintiff’s reputation in the eyes of a reasonable person;

(2) that the words, in fact, referred to the plaintiff; and

(3) that the words were published.[^34]

[44] A finding that a publication is defamatory is not the end of such a law suit. Rather it is the beginning:

If the plaintiff proves the required elements, the onus then shifts to the defendant to advance a defence in order to escape liability.[^35]

[45] In this case the three requirements required to demonstrate defamation are present.

[46] I begin with the obvious. The article refers to the plaintiffs, both by name (the second of the three requirements):

...the tests are run by a lab in Toronto, Canada operated by a company called the Health Genetic Center. The lab’s director is Yuri Melekhovets, who trained as a geneticist in Moscow, Russia.

[47] Clearly, the article was published (the third requirement). Words are considered to be published, if they were communicated to at least one person other than the plaintiff.[^36] The article appeared both in the print version of New Scientist Magazine, and on its website.

[48] In considering whether the words would tend to lower the estimation in which the plaintiff is held by the reasonable person (the first requirement), it is not enough to look at individual words or phrases, determine that in their ordinary meaning they are or are not defamatory and, on this basis, conclude that the test is or not met. In the context of this case it is possible for a medical test to be referred to as “dangerous”, “unreliable” and even “prone to error” without the “publication” in which they are found to be defamatory. These words could appear in a circumstance where the generally accepted science demonstrates that, even with the risks implied, the test referred to is the best that has yet been devised. To find the meaning behind the alleged libel, to determine whether there is any “defamatory sting”, the court is to consider the publication as a whole and not dwell on any isolated phrase or word.[^37] In every defamation action, the trier of fact must determine the defamatory sting from both the plain meaning of the words complained of and from what the ordinary, reasonable person would infer from them in the context in which those words were published.[^38] In this case, read as whole, the article points to the supposed failings of the test as undertaken by the plaintiffs. One does not have to go further than the introductions. As already noted, the online version is titled, “The danger of unreliable paternity tests.” The lead caption, separated and expressed in italics, goes on: “Can prenatal blood tests identify a fetus’s father? The result could sway a decision to abort but our investigation suggests it isn’t always accurate.” The print version is headlined: “Pregnant, Desperate and Confused”. The lead-in caption on the title page states: “When a paternity test could determine whether a fetus gets aborted there should be no room for error, warns Peter Aldhous.” The article leads with the story of one of the two cases that were the subject of the paper that had been published by Denise Syndercombe Court. It, then, provides conclusions:

The test is run on a sample of the woman’s blood and cheek swabs from possible fathers. Our investigation suggests that the results are unreliable-with potentially devastating consequences. “Paternity testing can have profound effects on people’s lives and when there is an unborn child involved, may lead to a termination,” says Denise Syndercombe Court of Barts and The London School of Medicine and Dentistry, who ran the follow-up test for Katharyn.

[49] The article turns to the test, its evolution, the lab, names Yuri Melekhovets, reviews efforts to communicate with him, introduces concerns and returns to the circumstances of individuals who have purchased the paternity test. Read as a whole the article meets the legal test. It would lower the reputation of the plaintiff in the estimation of reasonable persons reading it.

JUSTIFICATION

[50] This takes these reasons to the issue that lies at the core of the case. The defences and whether they can be relied on by the defendants. I start with justification:

Defamatory words are presumed to be false. The onus is on the defendant to displace the presumption of falsity by establishing the truth of the defamatory words as a matter of fact.[^39]

[51] No matter how damaging or disparaging it may be to the plaintiffs, the truth can never be actionable. A plaintiff has no right to shield his or her character or reputation from an imputation that is not false.[^40] The defence of justification or “Truth” is a complete defence. If the facts which comprise the defamatory material are true, a plaintiff cannot succeed…[^41].

[52] Truth is not absolute. Where an allegedly defamatory statement of fact is “substantially true” the defence of justification will apply:[^42]

What is required to be proven is not the truth of each and every word or the literal truth of the statement, but rather the truth of the substance of the allegation or the sting of the charge...[^43]

Where the gist or sting of the charge is proven to be true, minor inaccuracies do not defeat the defence of justification… Conversely, if the overall impression of the publication is false, the defence fails even if some or even all of the literal words are proven to be true. Half-truths can be just as damaging as outright falsehoods, and their effect may be even more severe because they can be more difficult to explain…[^44]

[53] Is there evidence that stands as proof of the “truth” of the statements in the article or that those statements are “substantially true”?

[54] In 2000 Yuri Melekhovets set out to do what no other scientist or commercial laboratory had accomplished: to develop a prenatal paternity test in which the fetal DNA would be extracted from maternal blood. Within a year he claimed to have succeeded. He did this at a time when others with more background in the applicable science were attempting to do the same and withdrawing without success. No other scientist or commercial laboratory would accomplish the desired result until 2010, ten years later. In and of itself, this does not demonstrate that the test developed by Yuri Melekhovets was dangerous or unreliable. What it does is underscore the need to follow the precepts of science designed and in place to ensure the efficacy and safety of new ideas and programs before they are commercialized and made available to the public. The proclamation of Yuri Melekhovets that above all else he was a scientist recognizes this imperative. What he did denies it.

[55] Fundamental to this is the idea of peer review. Science does not hide its discoveries. Those who make them, make their findings known through presentations, papers and other publications so that those with training, background and interest in what has been done (their peers) can assess them, question them and replicate them to confirm the value of whatever advance is being asserted. It is significant that the concerns of Denise Syndercombe Court were presented to people interested in the science and published in a peer reviewed journal. This is not what Yuri Melekhovets did. No paper was published. No presentation was made. No debate or discussion within the community of those interested and trained took place. Instead, Yuri Melekhovets moved to commercialize his findings by selling the test to the public.

[56] It is not that concerns were not expressed. Michelle Beckwith is the Paternity and Bioinformatics Supervisor at PTC Laboratories. She was tendered, accepted and recognized as an expert in forensic DNA analysis, mixture analysis and statistics as they relate to DNA profiling used for relationship testing.[^45] Michelle Beckwith had reason, on three occasions, to contact Yuri Melekhovets as to the results of tests he had undertaken.

[57] One occurred following a test report dated January 17, 2001. It was not a paternity test but one examining whether, for estate purposes, one woman was the aunt of another. Michelle Beckwith had reason to be concerned with the validity of the results. The report concluded that the women were “biologically very closely related.” Michelle Beckwith considered that, on the data presented there was only a 51% chance that the second woman was the aunt of the first. Such a result is considered to be inconclusive.[^46] Michelle Beckwith telephoned Yuri Melekhovets. He told her that her statistics were not relevant. He indicated that he was not able to develop a proper DNA profile of the aunt because there was a specimen collection error:

Meaning that my statistical analysis wasn't relevant. And what mattered was the samples that were submitted to him and that none of us could prove where those samples came from. We had no knowledge of where those samples came from.[^47]

[58] Evidently, Yuri Melekhovets was prepared to assume that the error was not in the test but in the samples he had received and on which the test was undertaken. There was no suggestion that any effort was made to confirm this suggested problem.

[59] The second of the tests was a prenatal test on maternal blood that was performed during February 2001. Michelle Beckwith saw errors in the work. She telephoned Yuri Melehkovets. She sought an explanation as to what she understood to be errors in his approach to calculating the paternity index in his test report:

…I do remember however, while I was on the phone, I managed to figure out that what was reported on this case wasn't a calculated index. It was the actual manufacturer's average expected index on a case and that's what was reported here.[^48]

[Emphasis added]

[60] Yuri Melekhovets was dismissive of these concerns:

He really felt that what I was saying was completely irrelevant and I had no idea what I was talking about.[^49]

[61] The third call concerned a woman who was six months pregnant. If the father was not her husband she was prepared to have a late term abortion. The report of Yuri Melekhovets indicated that the other man could not be excluded, that is to say there was a statistical likelihood that he was the father.[^50] A conventional test was undertaken. It excluded the other man as the father. PTC Laboratories identified typing errors, methodology errors, and assumption errors in the report that had been prepared by the plaintiffs. The only information Yuri Melekhovets passed on with respect to the test was that he had conducted it by obtaining the child’s sample through “fetal cell separation” from the mother’s blood.[^51] As he had in the first case, Yuri Melekhovets questioned whether the samples he had been sent, and on which the test was performed, were from the persons that were identified:

Again, there was a lot of focus on no one can tell where the samples came from and that you know, there's no way to know that what he got sent is what he was supposed to have been sent or what we received.[^52]

[62] The inquiries made by Michelle Beckwith are not, strictly speaking, peer review, at least in its formal sense. They were concerns and questions brought forward by a peer. These questions should have raised concerns about the reliability and, given the obvious impact on the mother, the fetus and even the psychology of the men involved, the risk carried by the test. The response of blaming, without any stated reason, the clients who provided the samples and questioning, without any foundation, the expertise of the peer demonstrates an attitude inconsistent with the requirements of good science.

[63] There was some suggestion the PTC Laboratories was a competitor of the plaintiff and that this was the motive for the actions taken by Joseph Gorman (its general counsel) and Michelle Beckwith in questioning the test and providing evidence at the trial. There was nothing in the evidence of either of these witnesses that would imply, much less confirm, this idea. To the contrary, it would seem that PTC Laboratories had and was prepared to refer clients to Yuri Melekhovets[^53]. The telephone calls and his response ended that sort of relationship:

As a result of this encounter [the second of the three mentioned], PTC believed that it could not responsibly refer any clients to Dr. Melekhovets or Genetest in the future.[^54]

And because of that, and because he told our scientist that he was correct, we felt we could never refer another client to him.[^55]

[64] Any scientific experiment, test or program that is run may produce a given result. This does not prove anything. The end may be random, that is “occur by chance”. It may not demonstrate anything that is regular or reliable about the process or the interaction of the elements involved. Any experiment, test or program has to be “validated”. Validation is the process that:

(a) assesses the ability of procedures to obtain reliable results under defined conditions,

(b) rigorously defines the conditions that are required to obtain the results,

(c) determines the limitations of the procedures,

(d) identifies aspects of the analysis that must be monitored and controlled; and

(e) forms the basis for the interpretation of guidelines to convey the significance of the findings.[^56]

[65] Validation is essential in the development of any diagnostic methods, including genetic relationship testing methods, such as prenatal paternity testing.[^57] It is the way the reliability of the test is proved. Any diagnostic method that is used by a scientist for a stated purpose, but has not been properly validated should be and is suspect:[^58]

…Results generated from any analyses gathered during the validation process (which includes sample collection, transfer, analytical method, and interpretation, all of which must be supported by proper training) must be fully evaluated so that the proper protocol eventually can be implemented. The limitations of the diagnostic method should be defined so that a proper degree of confidence can be associated with findings…

Validation tests must cover the range of samples that generally will be encountered when the diagnostic method is being used in practice. The variation in performance (i.e. data generated) must be assessed and must be used to define how the results derived from the diagnostic method are to be interpreted.[^59]

[66] Yuri Melekhovets says that he undertook two validation studies: one in the year 2000 and a second in 2005. The second study was prompted by a significant change in the methodology of the test. In its original form the test relied on genetic markers known as “short tandem repeats” (STRs). Beginning in 2006, following the second validation study, the test used “single nucleotide polymorphisms” (SNPs). In the paternity testing the child’s DNA sequence, as found at a specific genetic marker is compared to the father’s DNA at that same marker, to determine whether they match.[^60] How this comparison is made depends on the marker being used:

• When using STRs the geneticist will compare the number of repeats the child has at the targeted STR to the number of repeats the father has at that same location. Repeats refers to the sequence of the four nucleotide bases that make up DNA (adenine, cytosine, guanine and thymine or A, C, G and T for short). An STR with the pattern “AATG – AATG – AATG” is a “3” because the sequence repeats 3 times.

• When using SNPs the comparison is fundamentally different. It is not a question of looking for a repeating sequence. Rather it is a direct comparison to see whether the child’s DNA at a given SNP matches the alleged father’s DNA at that same SNP.

[67] It was the change from a test utilizing STRs to one utilizing SNPs that made the second validation study necessary. Bruce Budowle is the only person, other than Yuri Melekhovets, to have ever reviewed the two validation studies. They were not given to the experts who were introduced to this proceeding by the plaintiffs. Not unreasonably counsel for the defendants submitted that any opinion they may offer should be weighed against that omission.

[68] Yuri Melekhovets has acknowledged that the purpose of the 2000 study was to prove that he could generate a DNA profile for a fetus from a sample of maternal blood.[^61] To that end, he claims to have proven that the test was sensitive enough to:

• detect DNA in maternal blood,

• type fetal DNA in maternal blood,

• would not generate false positives, and

• would not detect fetal DNA in a sample if there was not fetal DNA in the sample.[^62]

[69] The studies said to make these findings do not, at least by the work that was produced, demonstrate that a proper validation study had been completed. A letter subsequently written on October 26, 2010 (prior to the article being published) by counsel representing the plaintiffs to the defendants indicated that whatever work was done during the year 2000 in furtherance of validating the test, it could not have confirmed the ability of the plaintiffs to extract fetal DNA from maternal blood:

…From the outset, this test was quite primitive; but each year, our client made significant improvements to the protocols and finally developed one that was capable of detecting and genotyping fetal DNA extracted from maternal blood samples.[^63]

[Emphasis added]

[70] For himself, Yuri Melekhovets acknowledged that it became apparent that STRs were not sensitive enough to detect the presence or absence of fetal DNA, at least in the earlier stages of pregnancy. The court and Yuri Melehkovets had the following exchange:

THE COURT: And there is a whole lot of stuff in all these various affidavits about what the difference is and why that one is different than the other, but for the purposes of your affidavit, the difference, as I understand it, is that you realized that short tandem repeats are not as reliable early in the pregnancy as SNPs can be?

THE WITNESS: Not -- I don't say not reliable, but not sensitive enough to detect presence or absence of fetal DNA.

THE COURT: All right.

THE WITNESS: It is sensitivity of reaction.

THE COURT: So I take it that and what I took from that, and this is just me, is what I took from that is that as the fetus matures, that sensitivity increases?

THE WITNESS: Exactly.

THE COURT: All right, and therefore –

THE WITNESS: Because when you do test at 10 weeks and you do test at 25 weeks, amount of fetal DNA will be completely different.

THE COURT: Okay, and this is not a problem with SNPs; therefore, you have moved to them in your prenatal testing?

THE WITNESS: Yes.[^64]

[71] In preparation for the change from STRs to SNPs, Yuri Melekhovets undertook the second of his validation studies. He testified that he was satisfied with the study, and found that SNP typing proved to be more sensitive in the detection of fetal DNA in maternal blood samples when compared to STR typing. It appears that this was his singular concern with this study and its outcome:

This was a great study because it showed a promising correlation between STR typing and SNP typing on known human samples. SNP typing proved much higher sensitivity in the detection of fetal DNA in maternal blood samples when compared to STR typing.[^65]

[72] It bears noting that Yuri Melekhovets holds this view even though at the time he had no prior experience working with SNPs. He worked with Vita-Tech Laboratory from 1996-2000. There he worked with STRs but not SNPs.[^66] There is nothing that suggests he worked with SNPs at either of his earlier places of employment: the University of Guelph (1991-1994) or the University of Toronto (1994-1996).

[73] Bruce Budowle does not agree with the conclusions of Yuri Melekhovets. To be clear, the only peer review undertaken with respect to the test developed by Yuri Melekhovets does not accept that the validation was proper or the test reliable. The report prepared by Bruce Budowle is replete with the criticisms that support these conclusions. What follows are examples:

…[A]s a matter of scientific reliability it was incumbent upon the Plaintiffs to demonstrate that this particular method of paternity testing was valid. The Plaintiffs claim to have run a “validation test” (i.e., a study on the performance and limitations of their methods) and claim that the “validation test” proved that their novel test can be used to generate a reliable fetal DNA profile from maternal blood. Neither claim is true. As will be explained in detail below, the “validation test” run by the Plaintiffs would not satisfy any accepted standards for validation testing in the scientific community, and this specific testing methodology is flawed.

It is my considered opinion that: (i) there is insufficient scientific evidence to support the Plaintiffs’ claim that their prenatal paternity test on maternal blood is either reliable or accurate; and (ii) the Plaintiffs’ test is unreliable.[^67]

The Plaintiffs have not conducted fundamental studies that are necessary to validate their test. Without these studies, the Plaintiffs, and for that matter any interested party, cannot assess whether the Plaintiffs’ test performs as they tout it to perform, or the limitations of the plaintiffs’ prenatal paternity test on maternal blood. Because these validations tests were not performed, the sensitivity of the tests, the ratios for making allele cells, the false positive (obtaining a DNA signal when there is no DNA or an allele) and false negatives (not obtaining a signal when there is a DNA or an allele within the sensitivity of detection of the assay) rates, the strength of an inclusion are unknown. The overall reliability, validity and effectiveness of the Plaintiffs’ test and thus inferences about paternity (either inclusion or exclusion) cannot be relied upon. [^68]

There is no documentation that the Plaintiffs have determined the amount and range of amounts of fetal DNA encountered in a pregnant mother’s plasma sample, in the plasma of a previously pregnant mother who currently is not pregnant, and in samples that could not have fetal DNA (such as a male donor sample) to serve as a negative control… Therefore, validation studies should be performed to know the limits of the effects of stochastic variation that reasonably can be encountered in the assay and indicators should be enacted to guide the user on when there is high risk of unreliable results. There is no such information provided in the Plaintiffs documentation.[^69]

Therefore, when the Plaintiffs’ attempt to compare the differences in the amount of fetal:mother DNA in the plasma, as they indicate at step 4 of the test… they employ must be very sensitive and quantitatively accurate to determine the genetic types of the fetus. Any quantitative assay as required for typing results in the Plaintiffs’ assay must be highly precise. The Plaintiffs have not performed such validation studies and have no guidelines on how to interpret the quantitative difference in allele cells.[^70]

… In fact no one – and given the Plaintiffs lack of documentation not even the Plaintiffs – know what portion is actually or likely recovered using the Plaintiffs’ test. This uncertainty is because the Plaintiffs have never run any validation tests or studies on the efficiency of recovery of DNA with their assay from general control samples or from plasma samples. In other words, the Plaintiffs have no data on the amount of fetal DNA they can recover with their method.[^71]

While it may be difficult to quantify the amount of fetal DNA in a mixture in plasma from a pregnant mother, several steps could have been taken to determine the limitations of testing low levels of fetal DNA. First a dilution series of several known samples from ideal quantities of DNA down to essentially no DNA should have been run. Basically, an evaluation (part of the validation process) of this nature would start with controlled known quantities of DNA. The PCR test should be run to see whether the results accurately reflect the amounts of DNA put into the analysis and more importantly record how varied the results can be. DNA then should be further diluted to lower quantities and the PCR test should be run again. This process should be repeated until the PCR is no longer able to detect and measure DNA… These studies should cover the range of DNA amounts that could possibly present in a mixture of maternal and female DNA in maternal blood. This type of validation testing is essential and routine.[^72]

Since the Plaintiffs’ test is purported to be based on a quantitative evaluation of heterozygote alleles, it is absolutely required to determine the variation of heterozygote alleles under various conditions that may be encountered with their test. This inherent variation has to be quantified with the specific assay conditions employed. There are no data of the degree of variation of the ratio of alleles of heterozygote types generated for known samples (single source samples), let alone mixtures, using the Plaintiffs assay. Thus, one cannot assert that the mother’s cellular fraction heterozygote balance is exactly the same every timetyped, and there is no support that the heterozygote balance from the mother’s cellular fraction DNA is exactly the same as in her plasma sample every time typed.[^73]

There is no documentation provided regarding the degree of stochastic [random] variation and any guidance for interpretation of heterozygote balance.[^74]

Restriction digestion is requisite according to the Plaintiffs for removing the mother’s homozygous DNA type from a plasma sample. It is not clear why this part of procedure is needed. If the Plaintiffs’ claim that an imbalance in heterozygote alleles can be used to determine paternal contribution, then the sensitivity of detection should be sufficient to detect the paternal contribution even in the presence of the mother’s DNA (when she is a homozygote at a marker). Thus, it is not clear why restriction digestion is needed to detect a paternal obligate allele different than that of a homozygous mother. Nonetheless, if the Plaintiffs are using restriction digestion, then the Plaintiffs have not conducted any studies to ensure reliability of this aspect of the assay.[^75]

Recommended conditions should be followed when using restriction enzymes; however, that does not mean that the intended conditions within the laboratory in fact approximate those recommended by the manufacturer. Favorable conditions can only be demonstrated by validation studies and running control samples that are digested during the test. No data have been provided that such validation studies have been performed.[^76]

The Plaintiffs’ claim to have conducted validation studies, which they produced at Plaintiffs Answers to Undertakings Tabs 18A, B and C. I have reviewed what the Plaintiffs have provided as their entire validation studies. In my opinion, the studies conducted by the Plaintiffs do not constitute proper validation studies, and they do not support the Plaintiffs claim that their prenatal paternity test on maternal blood is effective and reliable. In fact, the Plaintiffs validation studies suggested that the Plaintiffs do not appreciate the process and documentation required to conduct proper validation studies.[^77]

[Emphasis throughout these quotations added]

[74] I return to the idea that Yuri Melekhovets, for all his claiming to be a dedicated scientist did not behave like one. He recognized that a validation study is a way to prove that a test does what it is designed to do,[^78] and to identify limits of the test and conditions that are required in order to obtain reliable results.[^79] He agreed that a validation study is used to prove that a test works as claimed,[^80] and can be used to go back and investigate the reasons for flaws in the test.[^81] And yet, he did not act on this understanding. Indeed, he was dismissive of the need for peer review of his validation studies:

I don’t need to convince other scientists how good this test [is].[^82]

[75] To my mind this demonstrates the problem with the approach taken by Yuri Melekhovets. While he professes a dedication to science, he has ignored its standards and failed to adhere to what a proper scientific methodology requires. His motivation was revealed in the words that precede the above quotation. In making this statement he was not concerned with science; he was concerned with business, the commercialization of the test:

THE COURT: But, you see, the reason I ask this is I thought that is what it was, but in science, speaking in the grander sense, pure validation studies is the sort of thing as I understand it that you would subject to peer review, which would in effect confirm the validation. This was – – none of this was peer reviewed.

THE WITNESS: To be honest with you, I little bit disagree, and I will explain to you why.

You do validation studies in the lab for the lab for your testing. For example, when you go to let’s say any dealership and buy new car, you don’t ask dealership for validation study on this car. It is how this car performs. If it is driving well, same with testing.

But you do validation study to convince yourself that you have in your hands is actually works. It is number one.

Number two, what you said about publications, I publish lots of papers, okay. And in science, in University, let’s say it is different because publications is grants, grants is students, students, again student grants, it is just cycle.

In business it is a little bit different. If I will publish validation studies, people ask me for protocols. I cannot say, sorry, I cannot publish protocols because it will be a little bit strange.

And also as I said, I would like to repeat, everything what done in business is done in business, like before anybody offers something you need to be sure whatever you are offering is a good product, let’s say this way.[^83]

[76] That the commercial value of the test was the centre of his concern was confirmed in the lead up to the publication of the article. Representatives of the magazine contacted Yuri Melekhovets. He retained counsel. In an email to that lawyer, dated November11, 2010, Rowan Hooper, at the time the News Editor at New Scientist (at the time he swore the affidavit relied on as his direct evidence at trial, the Managing Editor)[^84] requested:

• a list of the markers Yuri Melekhovets was using. In particular the markers that were deployed in the initial panel for both male and female fetuses and to list separately any subsequent markers that maybe added to the analysis in the event of a failure to distinguish between alleged fathers.

and

• the research data that had been used in the laboratory protocols, referred to in the email as “published data.”[^85]

[77] In response the lawyer noted that the nature of the questions asked entered into the realm of the plaintiffs’ protocols and the manner and method of their creation. To divulge this information the plaintiffs required that the defendants enter into a non-disclosure agreement because the information involved represented “trade secrets”.[^86] In the context of business this is understandable; in the context of science it is a further denial of the assistance that review by others can provide.

[78] As it is, I accept the evidence of Bruce Budowle. No proper validation study was undertaken with respect to the test, relying on either STRs or SNPs. As a practical matter, other than the general assertions of Yuri Melekhovets, there is little, if any, evidence that contradicts it.

[79] There are other indications of the failure of Yuri Melekhovets to live up to the requirements and standards of good science.

[80] There was some concern, raised through the trial, as to whether the validation studies as delivered to and reviewed by Bruce Budowle were complete. During the course of being examined for discovery Yuri Melekhovets was asked whether there was any part to his validation study that had not been produced. He said that he had given the defendants everything he had. Counsel for the defendants was careful to explain the importance of this:

What I do not want is to suddenly be told that there were other studies done or other documentation as part of this study, and that he just doesn’t have it anymore. And so, if that’s the case I want to know that now, and that’s the type of information, just so we’re clear, that when a judge sees this transcript they’re going to want to know that Dr. Melekhovets can tell us everything he did for his validation study. And he will also be able to tell us if any of the documents that were produced as a result are missing. And if he can’t, then certainly we can assume that if it’s missing it’s not important.

[81] Counsel for the plaintiffs responded:

Ms. Barton, at this point I’m willing for my client to answer any questions regarding the documents that we’ve produced, but there are no more documents that were produced in terms of validation study, so you can be assured of that, and that’s on the record.[^87]

[82] At trial, Yuri Melekhovets agreed that a sensitivity study was an important part of a validation study, and claimed for the first time that he did perform a test for sensitivity as part of his validation work. There is no test of this kind included in the material which was produced in the litigation.[^88] Yuri Melekhovets also claimed that he had conducted a study on the ratio of alleles (a form of genetic marker occurring at a locus on a chromosome) as part of his validation study, even though no such study was included in the validation study documents produced in the litigation.[^89] Yuri Melekhovets claimed, at trial, that he selected the STRs for use in the validation study by checking the frequency of the selected STRs in the human population, and by checking how sensitive these markers were in detecting fetal DNA in maternal blood.[^90] He admitted that the validation study that was produced for the litigation did not contain this work.[^91]

[83] In these circumstances, I am not prepared to give any weight to the assertion that there were some parts of the validation studies that were undertaken but not maintained by the plaintiffs or produced in this action. It is apparent that proper validation is of central importance in confirming that a new procedure or test has been demonstrated to be efficacious and reliable. For the purposes of these reasons if they are not here; they do not exist.

[84] There is another requirement that should be in place before a new test is put on the market. The innovator (the creator of the test) must prepare a protocol that outlines the steps and procedures that must be followed in order to run the test:

Any time a new scientific procedure or application is developed for use, it is important that the procedure be tested to ensure that: (i) a protocol is developed that, if followed, would allow someone versed in the science to repeat the procedure and obtain the desired result [the test must be replicable] and (ii) the strengths and limitations of the procedure are known and well-described [the process must be clear and understandable].[^92]

[85] There are industry-wide standards regarding the type of information that must be included in all such protocols. Specifically, protocols should cover all aspects of the process, starting from the moment the sample is first brought into the lab through to the final interpretation of the results obtained from that sample.[^93] Yuri Melekhovets agreed that a protocol should describe all of the steps that must be taken to run a test successfully, and describe any limitations to the test that are discovered during the validation study. He agreed that the point of a protocol is that someone who is trained in the science should be able to repeat the test if they follow the protocol.[^94] The plaintiff produced his protocol for this action.[^95]

[86] Bruce Budowle testified that he is regularly asked “by industry and by government authorities to inspect labs, verify testing protocols and assess reliability of testing methods…” He was asked to and did so here.[^96] In his view most of the elements required in a protocol were not included in, or formed part of, the protocol relied on by Yuri Melekhovets:

… There also is no indication that the Plaintiffs run a restriction digestion positive control sample during their testing to monitor if the enzyme is performing as expected. Indeed, there is no statement or listing in their protocol of how a positive control is run and what is the expected DNA profile of the control sample. Indeed, the Plaintiffs have not demonstrated if such reliable restriction digestion conditions are met…[^97]

As a result, the Plaintiffs also could not have developed any controls that could run each alongside their test, to track whether any such contamination has occurred. Indeed, no such controls are called for or described in the Plaintiffs’ protocol. Controls (known samples with known DNA type subjected to the same methods used for paternity testing samples) are essential to monitor performance.[^98]

Replicate studies show the range of results that can occur. Replicate studies define the degree of variation in the results obtained under controlled conditions. These data are imperative because the Plaintiffs rely on quantitative data to interpret results (note that the Plaintiffs have provided no quantitative data at all in protocols or validation studies or quantitative interpretation guidelines in their protocols). The plaintiffs have not performed this very basic study (that Lo et al reported) and thus do not know the precision of their assay.[^99]

The plaintiffs use the “AimStepTM Professional Early Detection hCG Pregnancy Test” to test whether the mother is pregnant. If the test is positive and only if positive, the DNA typing process proceeds. I note, however, that the AimStepTM Professional Early Detection hCG Pregnancy Test was designed to be conducted on urine samples and not on blood samples. Instead of running this test on the putative mother’s urine, the Plaintiffs run this test on the putative mother’s blood. The Plaintiffs’ protocol does not provide any data to support the conclusion that this test can be used to determine pregnancy, using a blood sample.[^100]

The Plaintiffs’ protocol does not describe what interpretation would be applied if no DNA remains after digestion (which would be indicated by a negative result)… Relying on a negative result is problematic because a negative result could be due to a failure or complication of the assay; and importantly it is not described in the Plaintiffs’ protocol. Indeed, there are no interpretation guidelines in the protocols provided by the Plaintiffs. Therefore, there is no description on how to interpret results. Word of mouth is not a quality way to perform analyses and convey to employees and interested parties processes – a laboratory must document how to interpret data.[^101]

The Plaintiffs have no quantitative criteria regarding what constitutes a sufficient difference in a ratio to infer a paternal contribution. No such criteria are described in the Plaintiffs’ protocol – Dr. Melekhovets testified that it is “kind of in his head”.[^102]

The Plaintiffs need this information to determine how much of a difference the signal generated in the cellular fraction from that generated in the plasma fraction, is needed to infer that the additional signal is attributable to the presence of fetal DNA in the plasma fraction. The Plaintiffs have not conducted any validation studies on this, nor is this issue addressed in the Plaintiffs’ protocol.[^103]

The Plaintiffs’ protocol also does not explain how the Plaintiffs interpret the paternal contribution when there is no difference in the ratios of the heterozygote alleles between the two fractions. It is also not described whether there is any variation (a common feature of PCR assays) in the ratio of heterozygote signals if the same sample (non-mixture) was run multiple times. These issues can impact reliability, can change the strength of an inclusion of an alleged father (i.e., false positives and false negatives, and the statistical calculation). There is no guidance by the Plaintiffs regarding their use of their protocol.[^104]

As part of a quality system, in every assay, positive and negative controls are run. These controls should be documented in the applicable protocol. Every protocol should specify the controls used. Information on controls was not provided in the Plaintiffs’ protocol, even though the Plaintiffs claim that negative controls are used.[^105]

[Emphasis throughout these quotations added]

[87] No evidence was presented that runs counter to the observations made by Bruce Budowle respecting and demonstrating that the protocol relied on by the plaintiffs was not in keeping with what is expected and was insufficient for the purpose. The protocol is a document that outside auditors or peers will use to review a new diagnostic method, to assess its potential validity. This is because peer review of a detailed protocol would allow others to assess and attempt to obtain the same results by following the same protocol. The failure to produce a proper protocol is a further demonstration that the requirements of proper scientific methodology were not met in the work that supposedly supported the usefulness of the test produced and sold by the plaintiffs.

[88] The concerns this caused those who reviewed this work were confirmed by what has been learned about at least some aspects of the test. From 2001 to approximately 2006 the plaintiffs used STRs in the test. In time, Yuri Melekhovets came to understand that those genetic markers were not sensitive enough for use in his test early in the pregnancy. For pregnancies at 25 weeks and up, STRs could be used but not earlier. At 10 weeks SNP typing was “very good”.[^106] From this foundation it is apparent that any testing done during the years 2001-2006, where the pregnancy was at less than 25 weeks, was problematic. Bruce Budowle characterized the use of STRs to type free-circulating fetal DNA as “troubling”:

The rather scant protocol that Plaintiffs provide[d] for STRs… are similar in nature to that of SNPs (i.e., no quantitation of DNA, excessive number of PCR cycles and complexities of mixture interpretation) it is highly likely that unreliable results would be obtained and many results would be overblown and un-interpretable.[^107]

[Emphasis added]

[89] Yuri Melekhovets has acknowledged that the test depends on there being enough fetal DNA in the maternal blood to generate the requisite profile. His study did not determine the range of amounts of fetal DNA in the plasma fraction[^108] and did not determine how much was extracted. Simply put, absent this knowledge the efficiency of the test in recovering DNA is unknown.[^109] Bruce Budowle used the papers written by Y.M. Dennis Lo and his colleagues to determine the maximum amount of fetal DNA the plaintiffs might possibly extract from a maternal blood sample. The conservative nature of this work is demonstrated by the assumption that 100% of the fetal DNA found in the blood sample could be extracted. Bruce Budowle refers to this as “a technically impossible feat”.[^110] Even using this assumption, he concluded that the amounts of DNA present were so small that:

Such small amounts of DNA pushed the limits of any reliable DNA testing and are subject to undefined variation, which the Plaintiffs have not addressed.[^111]

[90] He summarizes his concern, as follows:

As stated above, for the purpose of this calculation I have overestimated the amount of DNA that actually can be recovered with the Plaintiffs’ test. In fact no one – and given the Plaintiffs’ lack of documentation not even the Plaintiffs – knows what the portion is actually or likely recovered from the Plaintiffs’ test. This uncertainty is because the Plaintiffs have never run any validation tests or studies on the efficiency of recovery of DNA with their assay from general control samples or from plasma samples. In other words, the Plaintiffs have no data on the amount of fetal DNA they can recover with their method. Even assuming that Plaintiffs’ test could achieve the impossible – i.e. 100% recovery of fetal DNA from plasma – it is highly unlikely that the amount of DNA would produce reliable and reproducible DNA testing, even for forensic genetic applications.[^112]

[91] In any case, the quantity of DNA obtained from maternal blood is insufficient to complete the analysis. This is not a new or unexpected problem. The same problem occurs in the forensic context where DNA is retrieved at a crime scene but in insufficient quantities to undertake the analysis that follows. To address this limitation, Polymerasse Chain Reaction technology (PCR) was developed. PCR is used to amplify (make copies of) a small section of DNA.[^113] This is done by taking small quantities of DNA and copying them over and over in a test tube. (The process is also referred to as molecular Xeroxing.[^114]) PCR is now a common technology used to make multiple copies of a piece of DNA.

[92] The number of cycles used to copy the targeted DNA will depend on the quantity of DNA first extracted for analysis. The smaller the quantity of DNA, the greater the number of cycles required to generate enough DNA for analysis. The manufacturer of the PCR kit used by the plaintiffs recommends that there not be more than 30-35 cycles.[^115] This recommendation would be based on validation studies performed by the manufacturer to determine the maximum number of cycles that could be used to reliably and accurately copy the DNA of interest.[^116] If the user is going to exceed the number of cycles recommended by the manufacturer, he, she or it should validate the reliability of the copying by conducting an independent or separate validation study.[^117] The plaintiffs exceeded the recommended number of cycles. Consistent with the evidence of Bruce Budowle that only very low amounts of fetal DNA can be expected to be retrieved from maternal blood, many more cycles were required. The plaintiffs utilize 65 cycles.[^118] Yuri Melekhovets was asked whether he had validated the PCR program as he had modified it, including the reliance on 65 cycles. He acknowledged such studies had not been produced in the record.[^119] In its absence, I give no credence to its existence much less any implication that somehow the PCR program was able to function properly to amplify (produce) the required quantity of fetal DNA. I accept the evidence of Bruce Budowle expressing concern for the number of cycles relied on:

The Plaintiffs employ a 65 cycle real time PCR assay, although there are no criteria or such data to support the reliability of using such an increased number of cycles. The Plaintiffs’ protocol does not provide any guidance on how to measure or interpret results in this scenario. Using this many cycles makes it impossible to quantitatively compare results. The results generated are no longer related to the amount of DNA in the sample; therefore it is impossible to perform the comparisons claimed by the Plaintiffs.[^120]

[Emphasis added]

[93] Other witnesses commented on the high number of cycles the plaintiff used. They each expressed concern. Dr. Robert Allen explained that running 60 or 65 PCR cycles, instead of the recommended 40, could result in the reactants in the PCR reaction being completely consumed before the copying process is completed, thereby compromising the ability of the PCR to reliably copy DNA during each cycle or at all.[^121] Robert Allen has never seen a PCR test that required more than 40 cycles, and testified that most labs try to keep the number of PCR cycles as low as possible.[^122] Dr. Monte Miller agreed that PCR can become less reliable when you run more than 40 cycles.[^123] Dr. Dean Stetler agreed that the plaintiffs use approximately double the number of PCR cycles normally used in human identity testing, forensic or parentage cases and that errors in DNA typing can occur.[^124] Both Monte Miller and Dean Stetler met with Bruce Budowle in advance of the trial to discuss the issues to be raised and to determine their areas of agreement and disagreement. These conversations touched on the concern for the number of cycles the plaintiff had employed using PCR to amplify the available DNA. For Dean Stetler this was his area of expertise;

Dr. Stetler I now want to talk about your area of particular expertise which is PCR and here particularly PCR cycles?[^125]

[94] He agreed that:

The amount of DNA tested by the Plaintiff's non-invasive prenatal paternity test equates slightly less than 1 cell to slightly less than 10 cells of DNA.

Such small amounts push the limits of any DNA testing and are subject to stochastic effects.

These uncontrolled effects of analyzing such low levels of DNA are exacerbated because the Plaintiffs use about twice the number of PCR cycles compared with any human identity testing used in forensics or parentage cases.[^126]

[95] For his part Monte Miller was unprepared to agree with this proposition but noted that the issue of whether sufficient DNA was utilized by the plaintiffs is best measured by their analytical results.[^127] To do this you need the “analytical results”. There is no suggestion they were provided. This is another example of the problems created by the failure of the plaintiffs to allow the test and its methodology to be evaluated and validated. Monte Miller has not seen any data.[^128]

[96] The failure to respond to the uncertainty associated with running an increased number of cycles of PCR brings real and substantive risk. In any testing of any sample of any substance there is a danger of contamination. In the context of the plaintiffs’ test this refers to the appearance, in the samples being used, of DNA that does not come from the maternal blood:

The Plaintiffs’ non-invasive prenatal paternity test is at high risk of contamination.

Contamination in this context means the appearance of DNA that does not derive from the sample taken.[^129]

[97] The increased number of cycles used in the amplification of the DNA samples enhances the risk of contamination:

The risk of contamination is a serious concern because the Plaintiffs’ non-invasive prenatal paternity test employs 65 cycles in their PCR.[^130]

[98] All of Monte Miller, Dean Stetler and Bruce Budowle agree that running the number of cycles of PCR as high as 65 raised the prospect of contaminating the DNA that is the subject of the test[^131]:

…PCR performed at so many cycles is far more susceptible to contamination and thus false positives can result…. There are well-documented examples of contamination from the manufacturer of laboratory consumables (such as tubes).[^132]

[Emphasis added]

[99] I note an example where the risk of DNA being present but from a source other than the woman being tested may have materialized. Celeste Biever works as a “Beat editor” at New Scientist magazine. Using a pseudonym (Sarah Johnson) the defendants delivered a sample of her blood to the plaintiffs, asserting that it was the blood of a woman who was 16 weeks pregnant. Celeste Biever was not and had never been pregnant. Accompanying the blood were samples of the DNA of Peter Aldhous and Rowan Hooper. These samples were represented as coming from prospective fathers of the non-existent fetus.[^133] The plaintiffs applied the test and reported that Peter Aldhous could not be excluded as the father of the non-existent fetus. How could this happen? Bruce Budowle offered the opinion that a result offering a DNA profile for a non-existent fetus demonstrated that the pregnancy test stage of the plaintiffs’ test was unreliable.[^134] One explanation for an error at this first stage of the test could be the possibility that the blood sample was “contaminated” meaning that it contained DNA from some other source. Dean Stetler thought the test worked properly even though it treated Celeste Biever as pregnant when she was not because he thought there was another source of DNA involved.[^135] This could happen if the blood was kept or delivered in a tube or vial that had not been properly cleaned or where those handing the sample or the equipment had not taken the appropriate care.[^136] The fact is that, in this case, there was no record indicating whether a pregnancy test was performed, leaving the possibility that this fact was simply accepted without inquiry. Dean Stetler thought that the woman who was not pregnant might have had DNA in her blood from her own mother.[^137] This is not strictly speaking contamination but it is DNA from another and unexpected source. Dean Stetler acknowledged that this possibility raised another prospective challenge. A pregnant woman could have three sources of DNA in her blood: the mother, the fetus and the grandmother of the fetus. Dean Stetler confirmed that he would have anticipated that the validation studies would have addressed such other extraneous sources of DNA.[^138] But they did not. Bruce Budowle concluded that unless the plaintiffs used DNA-free consumables or implemented procedures to clean contaminating DNA found in consumables (i.e. test tubes), the high number of PCR cycles used could produce false positive results arising from the presence of DNA from some other or extraneous source.

[100] For the most part, humans are genetically identical. Approximately 99.9% of DNA is the same for all humans. This is why people have similar characteristics (two eyes, two ears, a nose and so on).[^139] For the purpose of relationship testing, geneticists are interested in identifying DNA information that makes an individual unique. Some DNA matches occur because many people within a specific population have the same DNA profile for a specific genetic marker. These are not helpful in finding the distinctions necessary for proper relationship testing. Other matches are less common within a given population and are, therefore, considered to be informative for this purpose. The latter can be used to distinguish between prospective fathers within that population. For example, the genetic marker related to brown eye colour would not be informative in China, where all men have brown eyes. On the other hand, 80% to 90% of Swedes have light coloured eyes. Thus, in China the dark brown eye colour test results would not exclude any male, while in Sweden a good portion of the population could be excluded.[^140] The ability, or a lack of ability, of a given DNA marker to make a useful distinction is understood to be a measure of its “discriminatory power”. The discriminatory power is reflected in:

• the variability of the genetic marker within that given population,

• the number of markers that support the distinction (many people within a given population may have very similar DNA profiles even though they are not related, it follows that several markers must be typed in order to increase the chances of excluding alleged fathers; the more markers are available, the more the number of possible people who fit the profile will be reduced), and

• the genetic markers must be inherited independently from each other (genetic markers are not independent where the characteristic of one is linked to the characteristic of the other, if, in the population of concern, all men who have brown eyes also have black hair they are not independent markers).

[101] In order to establish the discriminatory power of any given set of genetic markers within a particular population it is necessary to conduct population studies and collect population data necessary to understand the potential for any given characteristic within that population. The plaintiffs have done none of this. Bruce Budowle quotes Yuri Melekhovets as having said that population data “has nothing to do with the probability of paternity”[^141] Bruce Budowle says “this statement is simply wrong.” He considers this claim to be “disconcerting because it demonstrates a lack of understanding of basic principles of genetics and paternity testing.”[^142]

[102] This is a further indication of the failure of the plaintiffs to adhere to the fundamental understandings of the applicable science. In the absence of understanding the applicable population it is impossible to appreciate the discriminatory power of any particular genetic marker or the number or independence of markers necessary and required to arrive at the sought-after determination of paternity.

[103] The discriminatory power of a paternity test is the measure of its effectiveness. Paternity tests are expected to be sufficiently strong such that if a man being tested for paternity is not the true biological father, the test will exclude him. Hence the test is directed by the determination of, on one hand, whether a man can be positively “excluded” as the biological father but on the other, rather than a definitive statement that he is the father, one that says he cannot be excluded from being the father. Unlike “exclusion”, the “inability to exclude” cannot be viewed as absolute. Even so the conclusion that a prospective father “cannot be excluded” is accepted as being another way of saying that there is a high degree of statistical certainty that the subject of the test is, in fact, the true biological father. The fact that a paternity test has concluded that an alleged father cannot be excluded is supposed to be a scientifically significant and meaningful finding.

[104] In paternity testing, the “Probability of Paternity” (PoP) is the statistic used to measure the test’s discriminatory power. When a prospective father cannot be excluded the probability of paternity indicates what the statistical certainty is that the man is the child’s true biological father. According to industry standards, paternity tests are expected, at minimum, to have a probability of paternity of 99% in the United States, and 99.999% in Europe.[^143] A paternity test with a lower probability of paternity is not considered to be reliable because such a test will allow too many men who are not the true biological father to fall into the basket of “cannot be excluded”. A test that has a low probability of paternity would render what is, essentially, a meaningless result.[^144]

[105] The plaintiffs claim their test is 99% accurate. No studies were ever performed to substantiate the claim. The Health Genetic Center Customer Service Guide (the customer service manual used by the employees of the plaintiff corporation) directed HGC staff to advise clients that “the test is 99% accurate when all alleged fathers are submitted”. This is based on a comparison between the follow-up test and prenatal test.[^145] However, there is no evidence that the plaintiff has conducted any follow-up paternity tests for clients. Moreover, it does not take much to see that requiring that DNA from all prospective fathers be submitted is a significant qualifier. It was not suggested that any of the examples of tests undertaken by the plaintiff and referenced during the trial, in fact, were subject to testing that included all the prospective fathers or that purchasers were advised that the accuracy offered depended on this. Nor was it suggested that any other available test for paternity relied on this requirement to assert its accuracy.

[106] It is not possible to know what the probability of paternity is for the test created by, and utilized by, the plaintiffs because they have never done the requisite studies.[^146] There is no way of knowing what the discriminatory power (the probability of paternity) is. Bruce Budowle concluded:

Moreover, the Plaintiffs do not know what the strength of their test actually is. The strength of the association in any particular case is unknown as the Plaintiffs do not report a Paternity Index.[^147]

[107] Do these facts stand to support the application of the defence of justification? Has the “defamatory sting” found in the article been shown to be “substantially true”? More specifically has it been demonstrated that the assertions that the test is “dangerous”, “unreliable” and its results “unreliable” as well as “inaccurate” are substantially true. In their closing submissions counsel for the plaintiffs had little to say about the background or context for the claim. In short, with respect to the defence of justification, it is their position that the onus of proving justification lies with the defendants. It is for the defendants to prove that the test was unreliable and inaccurate. As the plaintiffs see it, the defendants have not met this onus and, on this basis, argue that the defence must fail.

[108] In taking this position the plaintiffs rely on Bernstein v. Poon.^148 In that case the defendant, in a book and subsequent interviews, was critical of a diet program marketed by the plaintiffs. As here, the plaintiffs alleged that they had been defamed. The court agreed: calling the program a “starvation diet” was defamatory as was the expressed concern that it promoted an increase to the patient’s metabolic rate and by this means added to the weight loss. The defendant argued that he was “criticizing the science underlying the claim that patients on the Bernstein Diet can lose five pounds of fat per week, rather than accusing the plaintiffs of misleading dieters”.[^149] The court did not accept that referring to the underlying science was a true representation of the concern the defendant was expressing:

…Whether he admits it or not, he is telling his readers that the Bernstein Diet claims to increase a dieter’s metabolic rate, that the use of injections forms part of this claim and that as this is not possible, dieters are being misled…[^150]

[109] Whatever was meant by “underlying science” in that case, it is unrelated in character to what is happening in this one. In considering the defence of justification, the issue there was whether the plaintiff was, by its claims, misleading the patients as to the results of the diet. The question was whether the diet could perform as advertised. What was involved was a disagreement between the defendant and the plaintiff. The defendant produced no expert evidence to support his concerns. His subjective belief as to the truth of his words was not enough.[^151] The suggestion that the promise of an “increased metabolic rate” was misleading was dealt by the recognition that those words were never used by the plaintiffs.[^152]

[110] While insisting that the words he used that were found to be defamatory were true, the defendant argued that they should be regarded as “comment” rather than as statements of fact.[^153] The court concluded that this was the correct characterization of the words.[^154] This being so the defence of justification was rendered redundant.[^155]

[111] The concerns expressed were the subject of debate among professionals concerned with losing weight.[^156] The comments made by the defendant were considered in the context of the defence of “fair comment.” Hence the redundancy of any consideration of “justification” as a defence. Having found that the discussions of diets and nutrition engaged the public interest and having characterized the statements found to be defamatory as “comment” the court went on to consider whether they represented the honest belief of the defendant (could any man honestly express that opinion on the proved facts[^157]). The court found that, with two exceptions, they did.[^158] The defendant’s comments regarding the injections and those he made on a video disputing the veracity of the plaintiff’s description of the method and purpose of the clinics took him beyond “honest belief” and outside the defence of “fair comment”.

[112] The situation in the case I am to decide is different. This is not a case where professionals in a broad and well-occupied field have differing opinions based on established facts. In this case “underlying science” refers to the foundational requirement of the test. Whether the plaintiff, standing alone, had done what no one else could do: extract fetal DNA from the blood of the mother of a quality and quantity sufficient to allow for a determination of the paternity of the child.

[113] The issue was not the efficacy of a product (a diet program); it was the validity of the science that underlay the test. This is not a matter of “comment” or “opinion”. It is fact (not comment) that would be demonstrated by the exercise of a proper and complete study as required by established and accepted scientific methodology. The submissions made on behalf of the plaintiffs suggest that they do not see the distinction between this and Bernstein v. Poon:

The defendants’ key scientific experts – Dr. Bruce Budowle, Dr. Denise Syndercombe-Court and Dr. Rick Staub – all acknowledged – directly or indirectly – that they have no practical experience in conducting prenatal paternity tests on maternal blood.

[114] I pause to interject that, at the time this work was being undertaken no one had this experience. I point out and confirm that the experts called on behalf of the defendants were more than just experienced in working with DNA and familiar with problems in extracting fetal DNA from maternal blood. I repeat that, in particular, the qualifications of Bruce Budowle and Denise Syndercombe Court were singularly impressive.

[115] The submissions made on behalf of the plaintiffs continue:

The admission is startling, particularly considering Dr. Budowle’s emphasis on the importance of validation studies as the gold standard for investigation leading to scientific acceptance.

[Emphasis added]

[116] Counsel for the plaintiffs understands the evidence of Bruce Budowle differently than do I. So far as I recall and as a search of the record suggests, he did not say that validation studies were the “gold standard”. Rather, he and others, taken individually and collectively, underscored that validation studies are an essential part of a true and proper scientific approach to a new advance such as the extraction of fetal DNA from maternal blood.

[117] The submissions made by counsel for the plaintiffs go on:

Yet none of the Defendants came close to even approximating this level of investigation when preparing the article.

As Dr. Aldhous admitted at trial, that sort of in-depth scientific discussion and review is the provenance of Nature and Science magazine – New Scientist’s main competitors, and his former employers. In other words, it is open to New Scientist to emulate Science or Nature, they simply chose not to.

[118] In other words, according to counsel for the plaintiffs, it was the responsibility of the magazine to undertake the validation study (or something akin to it) not properly done or completed by the plaintiffs. To my mind, this entirely misses the point. In failing to complete proper validation studies, prepare and publish a proper protocol, to proceed without the requisite population studies, without an understanding of the discriminatory power of the test (the probability of parenting) and, without them commercializing the test; Yuri Melekhovets moved from science to business in the absence of a meaningful understanding of whether the product had positive value within the accepted guidelines (99% probability of paternity in North America; 99.999% accuracy in Europe). The substance of the submission is that in order to question the test, the magazine had to do what the plaintiff did not, complete the science. The idea is that the innovator can proceed without adhering to the requirements of science and then defend itself from criticism by suing for defamation saying that the defendant, in making the statements said to be defamatory should have carried out the necessary studies.

[119] The submissions made on behalf of the plaintiffs object to this understanding:

To be clear: [this] is not to say that all science magazines are obliged to embark upon peer-reviewed validation studies. But, when a science magazine chooses to weigh in on [a] complex scientific debate and make empirical findings as to its outcome; where it ventures to make empirical factual conclusions to its readership; and when it seeks to impress upon its readers that it has, in fact, discovered the “good guys” and the “bad guys” in a scientific debate – it must acknowledge the limits of its findings, or risk the consequences.[^159]

[120] In saying that a science magazine does not have to do the studies the innovator has not but should have and then suggesting that without them the magazine cannot draw overarching conclusions about the outcome of “a complex scientific debate” the plaintiffs fail to grasp to what has happened. First, fully examined, there is no debate about the validity of the test. The work required to confirm that it did as was claimed was not done. There is no evidence other than the unsubstantiated claims of Yuri Melekhovets that the test works: that he could retrieve, amplify and reliably use DNA from the maternal blood. The experts brought to the trial by the plaintiffs (Dean Stetler and Monte Miller) were not given what Yuri Melekhovets held out to be validation studies to review. Bruce Budowle found them to be deficient for the purpose. They did not validate the legitimacy of the test. No one suggested that a proper protocol was available. The evidence is overwhelming that the science necessary to confirm the assertion that the test was reliable and accurate was not done.

[121] The implications of failure to complete the science necessary to confirm the validity of the test leads inexorably to the conclusion that the test cannot be reliable or accurate. This is confirmed by the failure to understand the fundamental requisites of such a test for example:

• the actual discriminatory power of the the DNA markers being relied on accompanied by the failure to have a proper understanding of the genetic characteristics of the populations to which the prospective fathers belonged,

• the assurance that the DNA produced through amplification using many more cycles that recommended by the manufacturer are of a sufficient quantity and quality to properly be the subject of the test, and

• the fact that the test apparently can find that fathers and sons can belong to three haplogroups when accepted science says that is impossible

all of which lead to the conclusion that the test will be and is unreliable. As I understood them, neither Bruce Budowle nor Denise Syndercombe Court would have any trouble with the logic of this proposition. They were clear in their assertion of this conclusion. In any case it would be the natural inference from what they said.

[122] This is enough to demonstrate that the statements referred to and found to be defamatory are “substantially true”. On this basis the defence of justification applies. Even if I am wrong in this, there is more. There is evidence which directly demonstrates that the test is unreliable and inaccurate.

[123] Michelle Beckwith expressed her concern on three separate occasions. Her evidence indicated that in each case the result was wrong. The response was that in two cases the samples must not have been improper and that in the other the expertise of the person raising the concern was questionable. There was no suggestion that Yuri Melekhovets made any inquiry into the sampling that had been done or had any knowledge of the background and experience of Michelle Beckwith. In the absence of any demonstrated substance to these responses I accept the evidence of Michelle Beckwith that these three incidents represent three occasions on which the results of the test were not reliable. Denise Syndercombe Court re-tested two of the cases dealt with by the laboratory of the plaintiffs. Again, in both cases the results of test undertaken by the plaintiffs were wrong.

[124] It should not be lost that these errors hurt people, change and complicate their lives.

[125] Milan Jeknich learned that his girlfriend, Kimberly Robbins, was pregnant. They were unsure whether he was the father.[^160] Kimberly Robbins did not want to undergo an invasive test, one that used amniocentesis.[^161] An internet search identified the plaintiff corporation as offering a test that was 99.9% accurate. The test report concluded that it was considered as proven that Milan Jeknich was the biological father.[^162] The probability of paternity was said to be 99.9% as compared to the random man of the North American population.[^163] This was not the result that either Milan Jeknich or Kimberly Robbins expected.[^164]Milan Jeknich changed his life. He and Kimberely “repaired” their relationship and moved in together. He attended with Kimberly at her medical appointments and became involved with “life changing decisions about the baby’s health while in utero”. These decisions were emotional and difficult — decisions Milan Jeknich would not have been able to make if he did not believe he was the father.[^165] Once the child was born, Milan took paternity leave, adjusted his work schedule in order to care for the child. and put his desire to attend law school on hold.[^166] His friends and family understood that Milan Jeknich had a son.[^167] Further research raised concerns. A follow-up test was done using another laboratory. Using conventional testing, it found that Milan Jeknich was not the child’s father.[^168] He spoke to Yuri Melekhovets who, Milan Jeknich says, assured him the test was scientific.[^169] For his part Yuri Melekhovets denies the conversation took place.[^170] If I were required to, I would accept the evidence of Milan Jeknich. His recollection was clear and the denial consistent with Yuri Melekhovets’s failure to recognize any of the problems that came to light. As it is, the presence or absence of this call does not impact on these reasons. A second follow-up test was performed by a third laboratory. It confirmed the results of the first follow-up test. Milan Jeknich was not the father. It concluded that Milan Jeknich had a 0% probability of paternity whereas a previous boyfriend of Kimberly Robbins had a 99.96% probability of paternity.[^171] It should go without saying, but it was said, that this had a devastating impact on the lives of Milan Jeknich, Kimberly Robbins and the child’s biological father.[^172]

[126] Kimberly Robbins and Milan Jeknich retained legal counsel. The credit card payment for the original test was not processed by the plaintiff corporation. It was processed by HealthGene Corporation, Ontario.Demand letters were sent. Both companies responded. These responses confirm the consistent failure of Yuri Melekhovets to act as the scientist he claims to be. The corporate plaintiff excused itself by saying that the test was performed by an independent third party.[^173] The alleged third party, HealthGene Corporation, explained that it was a veterinary diagnostic laboratory that did not offer any prenatal paternity testing.[^174] Yuri Melekhovets was asked about this:

Q. You knew very well when you wrote this letter that HGC was involved in paternity testing?

A. HGC was in paternity testing, yes.

Q. You also knew that credit card statements that were made to HGC were run through HealthGene?

A. Yes.

Q. And instead of investigating to determine whether or not you had made an error on your test, you sent them a letter telling them that HealthGene was a veterinary diagnostic laboratory?

A. Yes.

Q. That was your response?

A. Yes, response.

Q. So you think that is appropriate, sir --

A. It is appropriate response if lawyer cannot find out HealthGene is veterinary lab and find out who is actually do testing, what kind of lawyer is it? This is our response. You contact wrong company. Contact correct company, you will get your response on your request.

Q. Okay. And that is how you viewed your responsibility as the President and CEO and shareholder of both these companies?

A. No comments.

Q. Pardon me, sir?

A. No comments.[^175]

[127] During the trial, despite previous admissions made during the examinations for discovery, and despite direct evidence at trial from both Kimberly Robbins and Milan Jeknich, Yuri Melekhovets refused to admit that the report in question originated from the corporate plaintiff.[^176]

[128] On May 23, 2003 Milan Jeknich and Kimberly Robbins each obtained default judgment in the Maricopa County Superior Court, in the State of Arizona against HealthGene Corporation, Genetest Laboratories in the amount of $500,000 U.S.[^177]

[129] When Celeste Biever sent a sample of her blood to the plaintiffs, she was not pregnant. Peter Aldhous sent a sample of his DNA as did Rowan Hooper. Neither of them was the father of a child being carried by Celeste Biever. Despite this, the test concluded that Celeste Biever was pregnant, when she was not and that Peter Aldhous was the father when he could not have been.

[130] When this arose during the trial, I expressed concern about this evidence having been gathered relying on a statement of fact knowing that it was not true. At the time and in their submissions the defendants worked to justify this approach. The submissions made outline that a so-called blind test is scientifically justified. It is worthwhile to know how the test will respond when the sample provided contains no fetal material. Could it produce a false positive result? Such a test could provide information about the errors in DNA typing that had been revealed in some of the plaintiffs’ reports. This does not address the concern. Those involved in submitting the samples of concern were not acting as scientists but as journalists. To my mind there are ethical considerations that remain from purporting to act out of a broad public concern when they set out, in effect, to trick someone. The witnesses for the magazine suggested they were surprised by the result. It was not what they expected. Maybe so but if they did not recognize the possibility a different result why do it? Be that as it may, the fact remains the test considered Celeste Biever to be pregnant when she was not and Peter Aldhous to be the father when he could not be. The test was wrong.

[131] There were fourteen cases in which follow-up tests were conducted by accredited laboratories where these subsequent tests demonstrated that the results reported by the plaintiffs were wrong.[^178] I say wrong, as opposed to different, because the follow-up tests either occurred after the child had been born or utilizing accepted means of extracting fetal DNA (for example: amniocentesis).

[132] The fourteen cases demonstrate that the test is unreliable, inaccurate and prone to error. This serves to confirm that the defence of justification has been made out. On this basis alone I can, and I do, dismiss the action. Even with this, there are other defences that were relied on, so I go on.

QUALIFIED PRIVILEGE

[133] Prior to the release of Grant v. Torstar^179 in 2009, where statements of fact (as opposed to comments) were at issue only two defences were generally available. Justification was one; qualified privilege the other.[^180] Qualified privilege has been explained as:

An occasion is privileged if a statement is fairly made by a person in the discharge of some public or private duty, or for the purpose of pursuing or protecting some private interest, provided it is made to a person who has some corresponding interest in receiving it. The duty may be either legal, social or moral. The test is whether persons of ordinary intelligence and moral principle, or the great majority of right-minded persons, would have considered it a duty to communicate the information to those to whom it was published.[^181]

[134] The privilege does not attach to the statement itself.[^182] At common law an occasion of qualified privilege exists if the defendant had either a duty to publish or a legitimate interest in publishing the defamation and the person or persons to whom the defamation was published had a legitimate interest in or duty to receive it.[^183] Employment references, business and credit reports, and complaints to police, regulatory bodies or public authorities are classic examples of occasions of qualified privilege.[^184] The reciprocity between the party publishing the statement and those receiving it is an essential characteristic of the circumstances that permit reliance on the defence. Generally, it was understood that the privilege did not extend to situations involving publication to the world at large.[^185]

[135] In Myers v. Canadian Broadcasting Corp. (the trial decision, see fn. 182) the trial judge recognized that the availability of the defence of qualified privilege “may be expanding”.[^186] She referred to Grenier v. Southam Inc.[^187] where the Court of Appeal upheld a finding that qualified privilege can attach to a communication by the media published to the world at large if it is published in the context of a social or moral duty to raise the underlying issue. In that case, the privilege applied to a newspaper article which indicated how persons could get help when a member of their family had become obsessed with gospel preachers. The judge made note of the case of Silva v. Toronto Star Newspapers Ltd.[^188] There, the privilege was held to attach to a newspaper article which reported alleged abuses in a subsidized home for the elderly. The court found that the defendants had a moral duty to report their allegations.

[136] This expansion continues to be recognized but the inherent caution remains:

In recent decades courts have begun to moderate the strictures of qualified privilege, albeit in an ad hoc and incremental way. When a strong duty and interest seemed to warrant it, they have on occasion applied the privilege to publications to the world at large.

Despite these forays, the threshold for privilege remains high and the criteria for reciprocal duty and interest required to establish it unclear. It remains uncertain when, if ever, a media outlet can avail itself of the defence of qualified privilege. [^189]

[137] In Leenen v. Canadian Broadcasting Corporation^190 the court concluded that an essential element of the defence of qualified privilege is whether the communication was made in good faith. An honest belief in the truth of the material broadcast is not sufficient to constitute good faith.[^191] It is important that those publishing or broadcasting defamatory statements do so after a proper and thorough investigation of the facts.[^192]

[138] This is a case where the defence of qualified privilege applies. The hidden risk to those who have relied on this test and those who may yet come to make use of it and the harm that may be caused to those who learn too late that the results they have received may be and are wrong demonstrate both the social and moral duty to raise the issue and the interest of the world at large to receive it. This is made clear by the number of tests performed and the unhappy experience of Milan Jeknich and Kimberley Robbins. The staff at the magazine came to this through a conversation and expression of concern by an expert in the field (Denise Syndercombe Court) who subsequently published a peer reviewed article raising concern for the efficacy of the test. The 18 month investigation undertaken by those at New Scientist magazine substantiated the nature of the concern and the seriousness of the risk unknowingly taken by those who relied on the test. The plaintiffs were given an opportunity to respond to the concerns being raised prior to the publication of the article. In the letter which initiated the communication from the magazine, the plaintiffs were given four days to respond.[^193] The implication was that in the absence of a response within that timeframe the article might still be published. Rowan Hooper acknowledged that there was a tactical component to this:

Well, we want to create a sense of urgency in order to get a response from someone.[^194]

[139] The letter was dated October 18, 2010, the response was said to be required by October 22, 2010. As it transpired there was an exchange of correspondence between counsel for the plaintiffs and the magazine. The article was not published until December 1, 2010.[^195]

[140] These facts; the initial introduction to the issue, the investigation and the opportunity to respond demonstrate the presence of the requisite good faith.

[141] The plaintiffs’ response is the submission that application of the defence of qualified privilege, in circumstances where the publication is to the world at large, depends on the need for the publication being urgent. In doing so, the plaintiffs rely on the trial decision in Myers v. Canadian Broadcasting Corporation. In that case the CBC aired an episode of the program called the fifth estate. The particular program was titled “The Heart of the Matter”. It considered questions about the safety of a drug and the general process by which the Health Protection Branch of Health Canada approved drugs for patient use. The broadcast included clips from interviews including one with the plaintiff, a Toronto cardiologist. He sued. The trial judge found that he had been defamed. She found that the defence of qualified privilege was not available as the occasion was not such as to impose any duty to communicate the defamatory information to the public. The CBC had communicated a particular interpretation or theory about the possible problems with the regulation and possible harmful effect of the drug. It was not “fresh news”. It was part of an ongoing medical debate. There was no urgent need to ventilate the subject matter of the program.

[142] I repeat that in the case before the court there is no “ongoing medical debate”. This case relies on established facts and the inferences that inevitably arise from those facts. The plaintiffs have marketed a test that has not been validated (did it extract the necessary fetal DNA from the blood of the mother?). There is no proper protocol that would allow it to have been confirmed through replication by others (did the process of amplification produce DNA of sufficient quality relying, as it did, on many more cycles than recommended by the manufacturers of the PCR process?). Issues that require determination before the test could be reliably utilized have not been resolved (the establishment of the discriminatory power and probability of paternity). Foundational work was not complete (studies to identify the genetic characteristics of the populations of interest were not undertaken). The evidence of the experts called including but not limited to Bruce Budowle, Denise Syndercombe Court, Dean Stetler, Monte Miller and Dr. Rick Staub lead directly to the conclusion that the test could not be relied on. This was confirmed by the cases where conventional tests showed the plaintiffs’ test to have been incorrect in the result it reported.

[143] Tatiana Volossiouk is the Laboratory Manager at Health Genetic Center. She is the primary scientist who performs the plaintiffs’ non-invasive prenatal paternity test done on maternal blood. She has performed approximately three thousand of these tests.[^196] It is impossible to know how many of these tests are inaccurate and wrong. For one thing the plaintiffs do not keep the necessary records. Those affected may never know. They may find out years in the future at a time where significant stress is already apparent: when a blood transfusion, genetic testing, organ donation or some other extraordinary medical intervention is required.

[144] In the absence on an ongoing debate and the presence of the risk demonstrated by the errors made, there was an urgency to the publication of the concerns expressed in the article. It is better that those affected learn of the possible concern now, by design, and not later, by happenstance.

[145] The plaintiffs submit there can be no urgency. This is said to be demonstrated by the 18 months it took to complete the investigation before publication. In Myers v. Canadian Broadcasting Corp., the trial judge noted:

There was no urgent need to ventilate this subject matter. The CBC spent four and a half months researching the issue, and, having regard to all the circumstances a reasonable person would have done the same.[^197]

[146] The differences between that case and this one negates the supposed impact of this quotation. Consistent with the proposition that there is no ongoing debate is the acceptance that in this case the fundamental problem is the failure of the plaintiffs to do what should have been done to demonstrate the efficacy of the test before it was marketed and sold to the public. When Denise Syndercombe Court spoke to Peter Aldhous she raised a concern. She was not in a position to prove a point. The investigation had to be undertaken to establish if there was a risk. Only with the investigation completed did the urgency become apparent.

[147] I find that, were it necessary, the defence of qualified privilege is made out and would serve to exculpate the defendants from any liability otherwise accruing from the defamatory nature of the article.

FAIR COMMENT

[148] The defendants rely on the defence of fair comment.

[149] The following tests apply:

(a) The comment must be based on a matter of public interest.

(b) The comment must be based on fact.

(c) The comment, though it can include inferences of fact, must be recognizable as comment.

(d) The comment must satisfy the following objective test: could any person honestly express that opinion on proved facts?

(e) Even though the comment satisfies the objective test the defence can be defeated if the plaintiff proves that the defendant was actuated by express malice.[^198]

[150] Any concern as to the applicability of the defence requires consideration of whether the words of concern are comment and whether, as comment, they are supported by facts. “The comment must be an expression of opinion on a known set of facts, and the audience must be in a position to assess or evaluate the comment.”[^199] There are sound policy reasons for this:

But it is obvious that to entitle any publication to the benefit of this defence it must be clear to those who read it what the facts are and what comments are made upon them. And for two reasons. Because it is impossible to know whether the comments are fair unless we know what the facts are; and because the public must have an opportunity of judging the value of the comments.[^200]

What is important is that the facts be sufficiently stated or otherwise be known to the listeners that listeners are able to make up their own minds on the merits of Mair’s editorial comment.[^201]

[151] In this context, what is comment? The Supreme Court of Canada adopted a view expressed by the Court of Appeal of New Brunswick:

…“comment” includes a “deduction, inference, conclusion, criticism, judgment, remark or observation which is generally incapable of proof”.[^202]

[152] And went on to cite a well-known text in furtherance of a more refined understanding:

Brown’s The Law of Defamation in Canada (2nd ed. (loose-leaf)) cites ample authority for the proposition that words that may appear to be statements of fact may, in pith and substance, be properly construed as comment. This is particularly so in an editorial context where loose, figurative or hyperbolic language is used (Brown, vol. 4, at p. 27-317) in the context of political debate, commentary, media campaigns and public discourse.

[153] To my mind the words of concern to the effect that the test was “dangerous”, “unreliable”, “inaccurate”, “prone to error”, “contradictory”, “awry” and beset by “genetic errors” are comment. They stand as all of, or any of, “deduction, inference, conclusion, criticism, judgment, remark or observation” based on the 18 month investigation undertaken prior to the preparation and publication of the article in New Scientist Magazine. Over the course of carrying out this work Peter Aldhous spoke to 23 experts in various fields of DNA and relationship analysis and testing.[^203] In the affidavit that stood as his direct evidence Peter Aldhous said that these experts “confirmed” that results obtained by the plaintiffs’ test, or the methodology likely used by the plaintiffs in administering the test, were not reliable and not based on valid scientific principles.[^204] This is his view. It does not change the character of the words of concern, used in the article as conclusions, inferences, deductions, observations, judgments or criticisms. This is not, as counsel for the plaintiffs would have it, “a pitched battle between two groups: the plaintiffs and the defendants”.[^205] There is no evidence to oppose the concerns expressed. The evidence of Yuri Melekhovets stands as nothing more than unsupported assertions. The statements of concern are conclusions, deductions, inferences and so on that arise clearly and directly from the evidence that the test was not validated, there is no proper protocol on which efforts to replicate its findings could be based, critical understandings like the discriminatory power (probability of paternity) have not been determined and necessary information such as the genetic characteristics of the relevant populations has not been identified. The logic the words of concern represent is, in the circumstances, so clear and powerful that for the purpose of the defence of justification they satisfy the requirement of substantial truth but they are not facts because they cannot be proved as facts. The only party who could have undertaken a proper validation of the test, prepared a proper protocol so the others could see if the results were replicable and done what was necessary to determine the discriminatory power was the plaintiffs. If they had, it would be established through the exercise of a proper scientific methodology that the test was or was not inaccurate, unreliable and prone to error. The plaintiffs have not kept proper records. Even with them it would not be possible to undertake a systematic retesting of any significant number of the 3000 tests Tatiana Volossiouk says she has performed.

[154] To me this situation is not unlike the classic explanation of circumstantial evidence in a criminal case. If a person enters the building with a wet umbrella, we accept that it is raining. There is no other logical conclusion. As unlikely as it may be, we may not know about the faulty water fountain that sprayed as the person walked by.

[155] The statements of concern are comment; they are not facts. This, on its own, does not determine whether the defence of fair comment can apply. The question that remains is whether the facts on which the comments are founded are available such that the reader was free to make his or her own judgment as to the validity of the comments. Without this, the policy rationale would not be met. The requirements of the defence would not be satisfied.

[156] To support “comment” it is required that the factual foundation is properly disclosed, sufficiently indicated or so notorious as to already be understood by the audience:[^206]

What is important is that the facts be sufficiently stated or otherwise be known to the listeners that listeners are able to make up their own minds on the merits of [the] comment.[^207]

[157] It is plain that the facts which support the statement that the test of the plaintiffs is unreliable, inaccurate and prone to error are not notorious. To the contrary, it required the consideration and input of experts in a highly specialized field to identify the concerns that have led to these conclusions. Would it be clear to the reader of the article?

[158] Mainstream Canada v. Staniford[^208] provides some assistance in understanding the parameters of the factual foundation required to support a defence of fair comment. The plaintiff, Mainstream Canada was a large and significant producer of farmed salmon. The defendant, Don Staniford was a committed opponent to this form of fish production. In furtherance of his continuing campaign, a press release was published on the Internet. It announced “a smoking hot international campaign against Big Aquaculture”. The press release included mock cigarette packages which, like actual cigarette packaging, contained warnings: “Salmon Farming Kills”, “Salmon Farming Is Toxic” and “Salmon Farming Seriously Damages Health”.[^209] Mainstream Canada asserted that it was the subject of these statements and visual images. In their natural and ordinary meaning, they would be understood to mean that its business and products killed people. Mainstream Canada sued Don Staniford claiming it had been defamed. He relied on the defence of fair comment, arguing that what was published were not statements of fact but comment or opinion on matters of public interest.

[159] In the trial decision the judge reviewed her findings as to what stood as “facts” that “went to the pith and substance of the opinions expressed by Don Staniford.”[^210] She concluded that “it would take a determined reader to locate in the publications the facts on which Mr. Staniford’s comments are being made”. Nonetheless, the judge concluded that the facts were sufficiently stated, or otherwise known to readers, such that they could make up their own minds about the merits of what was said by Don Staniford.[^211] She dismissed the action.

[160] The British Columbia Court of Appeal took a different view. It reversed the trial judgment. The facts upon which the comments were based were not sufficiently identified:

…the judge made reference to a “determined reader”. If she was suggesting that it was sufficient for the facts upon which the defamatory comments were based to be known by determined readers who conducted research on the GAAIA website and beyond, then I respectfully disagree with her. This would mean that there would be a category of readers (i.e., the non‑determined readers) that was not in a position to assess or evaluate Mr. Staniford’s comments. In my opinion, the defence of fair comment is not intended to be available when only a portion of the audience was aware of the facts upon which a comment was made or could have been aware of those facts with relative ease without being required to use their own initiative to search out those facts.[^212]

[161] The factual foundation supporting comment is to be readily available to all the readers of the publication with “relative ease” and without the need for any individual initiative. The Court of Appeal went on:

It is not sufficient for the defence of fair comment for facts upon which the comments were made to be contained on website pages that were not alleged to contain defamatory comments or in hyperlinked documents unless those other pages or hyperlinked documents were identified by a clear reference to contain such facts.[^213]

[162] The location of any supporting facts must be clearly identified. The Court of Appeal continued:

Some of the facts upon which Mr. Staniford based his comments were neither contained in the press release nor notorious …and, while there was a general reference to the website in the press release, there was nothing in it to inform the readers that the comments were based on facts to be found on the website, or setting out where on the website they could be found.[^214]

[163] General reference to the factual foundation and where it can be found is not sufficient.

[164] What “factual foundation” for the defamatory comment is there in the article? It includes the defamatory comments but does it include sufficient supporting facts?

[165] A review is necessary.

[166] The article indicates that the magazine undertook an investigation which is referred to in such a way that it is clear the article relied on its findings. Statements regarding the conduct of the investigation and the participation of identified experts are statements of fact. It is factual that the experts referred to in the article made statements indicating conclusions repeated in the article. Of the 23 experts noted in the trial record, four are mentioned in the article: Yuri Melekhovets, Denise Syndercombe Court, Peter Underhill and Gordon Lauc. Denise Syndercombe Court is referred to in the context of the effect paternity testing can have on the lives of those involved and as having conducted some follow-up testing. Peter Underhill is mentioned in respect of tests reliant on the Y chromosome. Gordon Lauc is referenced in regard to his own efforts to undertake non-invasive paternity testing.

[167] The fact of contradictory results having been found is made clear: generally (“our investigation suggests that the results are unreliable”), specifically (reviewing the case of Kathryn and the case of Kimberly Robbins and Milan Jeknich) and somewhere in between (another company, Paternity Testing Corporation, the firm Michelle Beckwith is associated with, has run follow-up tests and found conclusions about paternity which were “inaccurate”). The circumstances surrounding the non-existing fetus is also reviewed in the article. It is a fact that a woman who was not pregnant submitted a sample, was identified as being pregnant and a man identified as the father. The article goes on to indicate that Denise Syndercombe Court was asked to test samples from the three adults who had supplied samples in respect of the nonexistent fetus. The results showed that the plaintiffs’ laboratory had incorrectly recorded “one SNP variant from the woman and one of the alleged ‘fathers’”.

[168] The magazine reviewed particular concerns with Y chromosomes. It identified two circumstances where father and fetus were identified as belonging to three branches of the evolutionary tree and that experts had advised that this was “biologically implausible.” Denise Syndercombe Court was given DNA samples from each of the four individuals. In the article she is said to have found that, while the fathers still matched the fetuses, the profiles differed from those provided “by the Canadian lab” for five SNPs in one case and seven in the other. Unlike the report from the plaintiffs, the profiles prepared by Denise Syndercombe Court were consistent with the evolutionary tree for Y chromosomes. In one case the fetus and the alleged father belonged to the most common haplogroup for men of European descent. The article advised that, based on the profile prepared by Denise Syndercombe Court, referred to as the “true profile for the tested SNPs”, millions of men would match the fetus. The article records that conventional paternity testing, based on more informative genetic markers, would have shown that in each of the two cases the man tested was not the father. In both cases the plaintiffs’ test had said that the two men could not be excluded as the father and that it was very unlikely that a random man would share the same SNP profile as the fetus. The article reports that Yuri Melekhovets was approached on the basis that there was an error concerning the testing done in respect of the Y chromosome. They did not fit with the evolutionary tree. His representative said that this was not relevant because the lab did not perform ancestry testing. This is where Peter Underhill is referenced. He is quoted as saying that: “They should have been able to use the Y chromosome tree to see that inconsistency”.

[169] The article also provided information concerning the difficulty in extracting maternal blood and using it as the source for fetal DNA in paternity testing. It acknowledged that geneticists, at the time the article was written, had some success developing tests on maternal blood to detect fetal genes not also carried by the mother. It was possible to determine fetal sex by detecting sequences on the Y chromosome. “But these methods have not easily translated to paternity testing.” Gordon Lauc is referred to having considerable experience but while being able to work with maternal blood and confirming male fetuses by looking at STRs on the Y chromosome, for female fetuses he was unable to reliably extract fetal DNA. The article establishes the problem with DNA being hard to isolate. One danger is the concern for a “false positive” finding. Testing this concern was the reason for submitting the sample from the woman who was not pregnant.

[170] The article records facts necessary to support comments the test was “dangerous”, “unreliable”, “prone to error”, “contradictory”, “awry” and beset by “genetic errors” being the words and phrases of concern to the plaintiffs.

[171] This takes me back to the five characteristics that govern the defence of “fair comment”. The plaintiffs conceded that the matters at issue are of public interest. It follows from these reasons that the comments made are based on fact, are recognizable as comment and that a person could honestly express those opinions on the facts as established. I will return to the question of malice later in these reasons. For the moment I say only, absent malice, the defence of “fair comment” is satisfied and available to be relied on by the defendants.

RESPONSIBLE COMMUNICATION

[172] There is one remaining defence that requires some, albeit brief, comment. There is a tension between two significant values. The protection of reputation through the law of defamation and the freedom of expression which is a fundamental right under the Charter of Rights Freedoms.[^215] Any law that restricts comment, as the law of defamation does, risks trenching on the freedom of expression. In recent years the courts have responded to the concern that some rebalancing between these competing values protected on the one hand by law of defamation and the other by the right to freedom of expression was necessary. The seminal case is Grant v. Torstar Corp. Its opening paragraphs make the point:

Freedom of expression is guaranteed by s. 2(b) of the Canadian Charter of Rights and Freedoms. It is essential to the functioning of our democracy, to seeking the truth in diverse fields of inquiry, and to our capacity for self-expression and individual realization.

But freedom of expression is not absolute. One limitation on free expression is the law of defamation, which protects a person’s reputation from unjustified assault. The law of defamation does not forbid people from expressing themselves. It merely provides that if a person defames another, that person may be required to pay damages to the other for the harm caused to the other’s reputation. However, if the defences available to a publisher are too narrowly defined, the result may be “libel chill”, undermining freedom of expression and of the press.

Two conflicting values are at stake — on the one hand freedom of expression and on the other the protection of reputation. While freedom of expression is a fundamental freedom protected by s. 2(b) of the Charter, courts have long recognized that protection of reputation is also worthy of legal recognition. The challenge of courts has been to strike an appropriate balance between them in articulating the common law of defamation. In this case, we are asked to consider, once again, whether this balance requires further adjustment.[^216]

[173] The case established, for Canadian law, the defence of responsible communication:

For the reasons that follow, I conclude that the common law should be modified to recognize a defence of responsible communication on matters of public interest.[^217]

[174] In respect of the applicability of this defence the relevant factors that aid in determining whether the defamatory communication was a matter of public interest and was responsibly made are:

• the case is one of public importance[^218] (the public interest in the matter is high in that the issue touches on difficult and sensitive life decisions);

• the issues raised are serious[^219] (their impact on the lives of those involved is significant);

• the urgency is profound[^220] (it is not marked by the time it took to complete the investigation but by the contingent concern for its impact on those who had been the subject of the test and were unaware of the possible inaccuracy of results that continue to inform their lives; their need to know required the defendant to publish when it did);

• the sources of the information that supports the statements, comments and conclusions are beyond any practical or demonstrated reproach[^221] (for the most part the qualifications of the experts were not questioned, nor could they be, some were particularly impressive; there was nothing in the evidence of the two witnesses from PTC Laboratories (Michelle Beckwith and Joseph Gorman) to suggest they were motivated, in their evidence, as competitors of the plaintiffs. To the contrary the evidence was that they had sent clients to the plaintiff but stopped doing so when they did not get satisfactory answers to the questioned that were raised);

• the plaintiffs were given an opportunity to respond[^222] (it is true that in the original request for information it was indicated that a response was expected within four days; the fact remains there was an exchange of correspondence between the magazine and a lawyer acting on behalf of the plaintiffs; the article was not published for some weeks thereafter; the plaintiff, Yuri Melekhovets, was referred to in the article; it was the position of the plaintiffs that Y chromosome ancestry was not relevant because the plaintiffs were not engaged in ancestry testing, it cannot be doubted that the Y chromosome is important in paternity testing, it is genetically what distinguishes men from women; the article also reports that Yuri Melekhovets did not accept that his lab had made errors, he suggested that the mistakes were the fault of others, this was his consistent response, in the absence of any substantive demonstration of errors by others this objection becomes abject; the depth of this is confirmed by the response (also reported in the article) that these others did not follow the direction of the plaintiff that all prospective fathers should be tested, there is no evidence that the plaintiffs themselves, ever, even once, acted on this direction; finally the plaintiffs cannot complain about a failure to review their position when they failed to provide substantive information about the test; the article reports that the plaintiffs refused to give details of the procedure but suggested it was based on “excellent research data” published by other laboratories; this is all reported in the article); and

• the submission made on behalf of the plaintiffs suggest there was no need to publish the defamatory comments[^223] other than to embarrass the plaintiffs (this ignores the point, the comments to the effect that the test was unreliable, inaccurate and prone to error are warnings for those who had relied on the test to consider, accept, ignore or reject).

[175] I reject the assertion that the article was “biased”; the evidence is strong because the foundation of the problem is the failure of the plaintiffs to subject their “discovery” to a proper scientific review and methodology; the evidence confirming (or rejecting) the test does not exist because the work was not done.

[176] The evidence is clear. The defendants communicated responsibly. The defence of responsible communication is available to them.

MALICE

[177] The issue of malice remains. Defendants cannot rely on the defence of qualified privilege or fair comment if they were motivated by malice. The presumption is that they were not. The onus is on the plaintiffs to show they were and, by that means, if they succeed, to prevent the defendants from relying on the two defences. There is no evidence to support a finding that the defendants bore or were motivated by any ill will directed to the plaintiffs. In particular there is nothing I am aware of that would hint at, much less demonstrate, that counsel for the defence demeaned and humiliated Yuri Melekhovets. It is unfortunate that counsel for the plaintiffs felt it necessary to include such a submission. It does them no credit and their client no benefit.[^224] I say the same with respect to the bald and unsupported allegation that Peter Aldhous was on a smear campaign.[^225] There was no malice. If it were necessary the defendants have established and can continue to rely on the defences of “qualified privilege” and “fair comment”. As I see it, this is not necessary as the defences of both “justification” and “responsible communication” are also made out.

[178] For the reasons reviewed herein the action is dismissed.

COSTS

[179] If the parties are unable to agree as to costs. I will consider written submissions on the following terms:

(1) On behalf of the defendants, no later than 15 days after the release of these reasons, such submissions to be no longer than 5 pages double spaced not including any Costs Outline, Bill of Costs or case law that may be included.

(2) On behalf of the plaintiffs, no later than 10 days thereafter, such submissions to be no longer than 5 pages double spaced not including any Costs Outline, Bill of Costs or case law that may be included.

(3) On behalf of the defendants, no later than 5 days thereafter, in reply if necessary, such submissions to be no longer than 2 pages double spaced.

Lederer J.

Released: December 7, 2018

[^1]: Affidavit of Yuri Melekhovets at Exhibit EE taken from Blood, Vol. 88, No 11 (December 1), 1996: pp. 4390-5395

[^2]: Affidavit of Yuri Melekhovet, at Exhibit FF taken from The Lancet Vol. 350, August 16, 1997: pp. 485-487

[^3]: Affidavit of Yuri Melekhovets at para. 60

[^4]: Trial Transcript, Peter Aldhous pp. 2131-2132

[^5]: Affidavit of Peter Aldhous at para. 110 and Exhibit G, Wagner, Dzijan, Marjanovic and Lauc, Non-invasive prenatal paternity testing for maternal blood Int J. Legal Med (2009)123:75-79

[^6]: Ibid, Exhibit G, Wagner, Dzijan, Marjanovic and Lauc, Non-invasive prenatal paternity testing for maternal blood at p. 78 under heading “Conclusions”

[^7]: Ibid at para. 111

[^8]: Affidavit of Karl Reich at paras. 20-21 and the Affidavit of Peter Aldhous at paras. 91-95

[^9]: Affidavit of Peter Aldhous at paras. 73-74 and Trial Transcript, Peter Aldhous at pp. 2100-2102

[^10]: Affidavit of Denise Syndercombe Court at paras. 1, 5-11

[^11]: Trial Transcript, Denise Syndercombe Court at p. 1622

[^12]: Affidavit of Denise Syndercombe Court at para. 31

[^13]: Affidavit of Peter Aldhous at para. 40.

[^14]: Affidavit of K… F… at paras. 22-23

[^15]: Affidavit of Denise Syndercombe Court at paras. 48-49

[^16]: Ibid at paras.48, 49, 51 and Exhibit H, and see Trial Transcript, Denise Syndercombe Court at p. 1681-2

[^17]: Ibid (Affidavit of Denise Syndercombe Court) at paras. 51-52

[^18]: Affidavit of Yuri Melekhovets at para. 4

[^19]: Ibid at para. 6

[^20]: Affidavit of Peter Underhill at paras. 23-25

[^21]: Ibid at para. 63 and Affidavit of Christopher Tyler-Smith at para. 44

[^22]: Trial Transcript, Dean Stetler at pp. 1446-1447

[^23]: Expert Report of Dr. [Monte] Miller at p. 8

[^24]: Affidavit of Bruce Budowle at Exhibit 1 (Expert Witness Report of Prof. Bruce Budowle) at p. 46

[^25]: Trial Transcript, Yuri Melekhovets at pp. 238-239

[^26]: Trial Transcript, Peter Underhill at p. 1704

[^27]: Affidavit of Peter Underhill at paras. 36, 37 and 43

[^28]: Affidavit of Denise Syndercombe-Court at para. 70

[^29]: Affidavit of Bruce Budowle at Exhibit 1A (curriculum vitae of Prof. Bruce Budowle) and see Trial Transcript, Bruce Budowle at pp. 1817-1840

[^30]: Ibid (Trial Transcript, Bruce Budowle) at p. 1825-1828. Bruce Budowle was tendered as an expert as follows:

I tender him as an expert in STRs including how they are inherited, polymorphisms, mutations in STRs and population data relating to STRs.

Analyzing low-grade DNA samples, analyzing degraded DNA samples. Analyzing DNA samples that are extras. Establishing, conducting and validating validation studies, and establishing, conducting and validating quality assurance and quality control processes for laboratories.

Conducting relationship testing and/or DNA analysis, including pursuant to ISO 17025.

And developing protocols for all aspects of relationship testing, including the extraction of DNA samples and typing up DNA samples. And validating and applying PCR technology, including real-time PCR and quantitative PCR, and all aspects of relationship testing, including the paternity testing.

(Trial Transcript, Bruce Budowle at pp. 1834-18345)

After some clarification initiated by the court, Bruce Budowle was accepted as an expert as tendered, without objection or any expressed concern.

[^31]: Ibid (Trial Transcript, Bruce Budowle) at p. 1962

[^32]: Affidavit of Yuri Melekhovets at para. 4

[^33]: Peter A. Downard, Libel, Third Edition © LexisNexis Canada Inc., October 2014 at p. 1 referring to Bou Malhab v. Diffusion Metromedia CMR Inc. [2011] S.C.J. No. 9, 2011 SCC 9 at paras. 33-37 (S.C.C.); Crookes v. Newton 2011 SCC 47, [2011] S.C.J. No. 47, [2011] 3 S.C.R. 269 at para. 39 (S.C.C.); Botiuk v. Toronto Free Press Publications Ltd., 1995 CanLII 60 (SCC), [1995] S.C.J. No. 69, [1995] 3 S.C.R. 3 at 24 (S.C.C.); Sim v. Stretch, [1936] 2 All E.R. 1237 at 1240 (H.L.); Skuse v. Granada Television Ltd., [1996] EMLR 278 at 286 (C.A.) and Modi v. Clarke, [2011] EWCA Civ 397 at para. 10 (Eng. C.A.)

[^34]: Grant v. Torstar Corp. 2009 SCC 61, [2009] SCC 61, [2009] 3 S.C.R. 640 at para. 28

[^35]: Ibid at para. 29

[^36]: Ibid at para. 28

[^37]: Brown, Law of Defamation in Canada, 2nd ed. (Scarborough, Ont.: Carswell, (1994) (looseleaf), vol. 1, 5.3(6)(a) at 5-55, see also: Martini v. Smith Lyons LLP, 2003 CanLII 22736 (ON CA), 171 O.A.C. 375, 2003 CarswellOnt 1731 (O.C.A.) at para. 14; Bernstein v. Poon, 2015 ONSC 155 at para. 44

[^38]: United Soils Management Ltd. v. Mohammed, 2017 ONSC 4450 39 CCLT (4th) 304; 7 CPC (8th) 58 at para. 21 referring to Rutman v. Rabinowitz, [2016] O.J. No. 6309 at para. 133 in turn referring to Cusson v. Quan (2007) ONCA 771, 286 D.L.R. (4th) 196 at para. 34

[^39]: Peter A. Downard, Libel, Third Edition, supra (fn. 33) at p. 105 referring to Littlejohn v. Hamilton 1974 CanLII 438 (ON CA), [1974] O.J. No. 1955 4 O.R. (2d) 283 at 286 (Ont. C.A.) leave to appeal to S.C.C. refused 4 O.R. (2d) 283n (S.C.C.); Govenlock v. London Free Press Co. 1915 CanLII 555 (ON CA), [1915] O.J. No. 15, 35 O.L.R. 79 at 83 (Ont. C.A.); Jameel v. Wall Street Journal Europe Sprl, [2005] EWCA Civ 74 at para. 4 (Eng. C.A.) rev’d on other grounds [2007] 1 A.C. 359 (H.L.)

[^40]: Raymond Brown, Brown on Defamation (Canada, United Kingdom, Australia, New Zealand and United States), Second Edition (Toronto: Thomson Reuters Limited) at ch. 10, para. 10.2

[^41]: Leenen v. Canadian Broadcasting Corp. 2000 CanLII 22380 (ON SC),48 OR (3d) 656; [2000] OTC 672; 50 CCLT (2d) 213 at par. 92

[^42]: Grant v. Torstar Corp., supra (fn. 34) at para. 32

[^43]: Cimolai v.Hall 2005 BSCS 31 at para. 172 referring to Bank of British Columbia v. Canadian Broadcasting Corp. (1995), 1995 CanLII 7442 (BC CA), 126 D.L.R. (4th) 644, 10 B.C.L.R. (3d) 201 at ¶57 (C.A.); P.G. Restaurant Ltd. v. Northern Interior Regional Health Board (2004), 25 B.C.L.R. (4th) 242, 2004 BCSC 294

[^44]: Ibid (Cimolai v. Hall) at para. 173 referring to Hodgson v. Canadian Newspapers Co. Ltd. (1998), 1998 CanLII 14820 (ON SC), 39 O.R. (3d) 235, [1998] O.J. No. 2682 (Gen. Div.), Trozzo Holdings Ltd. v. M.V.S. Construction Ltd. (2002), 18 C.L.R. (3d) 159, 2002 BCSC 1202 at ¶92; Roberge v. Tribune Publishers Limited (1977), 1977 CanLII 2320 (NB SC), 20 N.B.R. (2d) 381, [1977] N.B.J. No. 299 at ¶6 (S.C.).

[^45]: Trial Transcript, Michelle Beckwith at p. 2163

[^46]: Affidavit of Joseph Gorman at para. 53

[^47]: Trial Transcript, Michelle Beckwith at p. 2171

[^48]: Ibid at p. 2169, This problem was explained by Joseph Gorman:

So the way that the paternity test works is, you determine if the tested man has the necessary obligate allele, then you determine how rare or common that allele size is in the population, which leads to your paternity test index for that one DNA locus, but Dr. Melekhovets didn't use that at all. He simply used the manufacturer's average frequency of the alleles, which is simply a statement about, on average how many people would match. It says nothing about this individual person and man and child and the allele size that they matched at. So he doesn't have a basis to even calculate a paternity test result. (Trial Transcript, Joseph Gorman at p. 2105)

[^49]: Trial Transcript, Michelle Beckwith, at pp. 2169-2170

[^50]: Affidavit of Joseph Gorman at paras. 58-59

[^51]: Ibid at para. 60

[^52]: Trial Transcript, Michelle Beckwith at p. 2174

[^53]: Trial Transcript, Joseph Gorman at p. 2104

[^54]: Affidavit of Joseph Gorman at para. 50, Trial Transcript, Michelle Beckwith at pp. 2174-2175

[^55]: Trial Transcript, Joseph Gorman at p. 2106

[^56]: Affidavit of Bruce Budowle at Exhibit 1 (Expert Witness Report of Prof. Bruce Budowle) at p. 24

[^57]: Trial Transcript, Dean Stetler at p. 1484

[^58]: Affidavit of Bruce Budowle at Exhibit 1 (Expert Witness Report of Prof. Bruce Budowle) at p. 24

[^59]: Ibid at pp. 24-25

[^60]: Ibid at p. 26

[^61]: Trial Transcript, Yuri Melekhovets at p. 323

[^62]: Ibid at p. 323

[^63]: Affidavit of Dr. Peter Aldhous at Exhibit GG, Affidavit of Rowan Hooper at Exhibit Q

[^64]: Trial Transcript, Yuri Melekhovets at pp. 111-112

[^65]: Affidavit of Yuri Melekhovets at para. 30

[^66]: Trial Transcript, Yuri Melekhovets at p. 301

[^67]: Affidavit of Bruce Budowle at Exhibit 1 (Expert Witness Report of Prof. Bruce Budowle) at p. 6

[^68]: Ibid at p. 31

[^69]: Ibid at p. 32

[^70]: Ibid a p. 38

[^71]: Ibid at p. 36

[^72]: Ibid at pp. 32-33

[^73]: Ibid at p. 33

[^74]: Ibid at p. 33-34

[^75]: Ibid at p. 34

[^76]: Ibid at p. 34

[^77]: Ibid at p. 31

[^78]: Trial Transcript, Yuri Melekhovets at pp. 293-294

[^79]: Ibid at p. 294

[^80]: Ibid at p. 294

[^81]: Ibid at pp. 294-295

[^82]: Ibid at p. 122

[^83]: Ibid at p. 121-122

[^84]: Affidavit of Rowan Hooper at paras. 6 and 36

[^85]: Ibid at Exhibit U

[^86]: Ibid at Exhibit U

[^87]: Examination for Discovery of Yuri Melekhovets at Q.1789

[^88]: Trial Transcript, Yuri Melekhovets at p.1273

[^89]: Ibid at p. 1275

[^90]: Ibid at p. 347

[^91]: Ibid at p. 348

[^92]: Affidavit of Bruce Budowle at Exhibit 1 (Expert Witness Report of Prof. Bruce Budowle) at p. 8

[^93]: Ibid at p. 22

[^94]: Trial Transcript, Yuri Melekhovets at p. 296

[^95]: Affidavit of Yuri Melekhovets at Exhibit S

[^96]: Affidavit of Bruce Budowle at Exhibit 1 (Expert Witness Report of Prof. Bruce Budowle) at p. 5

[^97]: Ibid at p. 34

[^98]: Ibid at p. 37

[^99]: Ibid at p. 38

[^100]: Ibid at p.39

[^101]: Ibid at p. 41

[^102]: Ibid at p. 42

[^103]: Ibid at p. 42

[^104]: Ibid at p. 42

[^105]: Ibid at pp. 45-46

[^106]: Trial Transcript, Yuri Melekhovets at pp. 401-403

[^107]: Affidavit of Bruce Budowle at Exhibit 1 (Expert Witness Report of Prof. Bruce Budowle) at pp. 35-36

[^108]: As I understand it the plasma fraction is the liquid (as opposed to the solid) component of blood).

[^109]: Trial Transcript, Yuri Melekhovets at pp. 1278

[^110]: Affidavit of Bruce Budowle at Exhibit 1 (Expert Witness Report of Prof. Bruce Budowle) at p. 36

[^111]: Ibid at p. 36

[^112]: Ibid at pp. 36-37

[^113]: Ibid at p. 7

[^114]: Ibid at p. 23

[^115]: Read in from: Examination for Discovery of Yuri Melekhovets at p.476 Q. 2203 – p. 477 Q. 2208

[^116]: Affidavit of Bruce Budowle at Exhibit 1 (Expert Witness Report of Prof. Bruce Budowle) at p. 23

[^117]: Ibid at pp. 23-24

[^118]: Trial Transcript, Yuri Melekhovets at p. 1280

[^119]: Ibid at pp. 1280-81

[^120]: Affidavit of Bruce Budowle at Exhibit 1 (Expert Witness Report of Prof. Bruce Budowle) at p. 43

[^121]: Trial Transcript, Robert Allen at pp. 2048-2049

[^122]: Ibid at pp. 2077-2080

[^123]: Trial Transcript, Monte Miller at pp. 963-967

[^124]: Trial Transcript, Dean Stetler at p. 1445

[^125]: Ibid at 1447

[^126]: Ibid at p. 1444 and. Affidavit of Bruce Budowle at Exhibit 2(Agreement as to Issues: Budowle and Stetler) at para. 63

[^127]: Affidavit of Bruce Budowle at Exhibit 3 (Agreement as to Issues: Budowle and Miller) at para. 63

[^128]: Ibid at para. 68 and 69

[^129]: Affidavit of Bruce Budowle at Exhibit 2(Agreement as to Issues: Budowle and Stetler) and Exhibit 3 (Agreement as to Issues: Budowle and Miller) both at para. 68

[^130]: Ibid both at para. 69

[^131]: Ibid both at para. 69

[^132]: Affidavit of Bruce Budowle at Exhibit 1 (Expert Witness Report of Prof. Bruce Budowle) at p. 43

[^133]: Affidavit of Rowan Hooper at paras. 20-21

[^134]: Trial Transcript, Bruce Budowle at p. 1955

[^135]: Trial Transcript, Dean Stetler at p.1517

[^136]: Affidavit of Bruce Budowle at Exhibit 1 (Expert Witness Report of Prof. Bruce Budowle) at p. 43

[^137]: Trial Transcript, Dean Stetler at p.1517

[^138]: Ibid at pp. 1518-19

[^139]: Affidavit of Peter Underhill at para. 17 and Affidavit of Bruce Budowle at Exhibit 1 (Expert Witness Report of Prof. Bruce Budowle) at p. 17

[^140]: Affidavit of Bruce Budowle at Exhibit 1 (Expert Witness Report of Prof. Bruce Budowle) at pp. 27-28

[^141]: Ibid (Expert Witness Report of Prof. Bruce Budowle) at p. 44 quoting from Transcript of the Further Examination for Discovery of Y. Melekhovets at p. 47

[^142]: Ibid at p. 44

[^143]: Affidavit of Denise Syndercombe-Court at para. 33

[^144]: Trial Transcript, Dean Stetler at p. 1477

[^145]: Trial Transcript, Yuri Melekhovets at p. 1386 and Affidavit of Yuri Melekhovets at Exhibit W at p. 4 There is no explanation for this but I note that at p. 6 the same document says the test is 98%-99% accurate. It prepares the employee to address why this is different from competitors who claim an accuracy of 99%..

[^146]: Trial Transcript, Bruce Budowle at pp. 1913-1914

[^147]: Ibid (Exhibit 1 (Expert Witness Report of Prof. Bruce Budowle)) at p. 44

[^149]: Ibid at para. 65

[^150]: Ibid at para. 66

[^151]: Ibid at para. 98

[^152]: Ibid at para. 65

[^153]: Ibid at para. 100

[^154]: Ibid at para.128

[^155]: Ibid at para. 100

[^156]: Ibid at paras. 122, 123 and 126

[^157]: Ibid at para. 103

[^158]: Ibid at 131

[^159]: Closing Submissions and Factum of the Plaintiffs at paras.35, 36, 37, 39 and 40

[^160]: Affidavit of Milan Jeknich at para. 5

[^161]: Affidavit of Kimberly Robbins at para. 7

[^162]: Ibid at para. 9-10 and Affidavit of Milan Jeknich at para. 27

[^163]: Ibid (Milan Jeknich) at para. 28 and (Kimberly Robbins) at para. 20

[^164]: Ibid (Milan Jeknich) at para. 29 and ( Kimberly Robbins) at para. 22

[^165]: Ibid (Milan Jeknich) at para. 33 and (Kimberly Robbins) at para. 23

[^166]: Ibid (Milan Jeknich) at paras. 34 and 35 and (Kimberly Robbins) at para. 23

[^167]: Ibid (Milan Jeknich) at para. 50

[^168]: Ibid at para. 41 and Affidavit of Kimberly Robbins at para. 29

[^169]: Ibid (Milan Jeknich) at para. 45

[^170]: Trial Transcript, Yuri Melekhovets at pp.463-467

[^171]: Affidavit of Milan Jeknich at paras. 47-48 and Affidavit of Kimberly Robbins at para. 30

[^172]: Ibid (Kimberly Robbins) at paras 31 and 32 and (Milan Jeknich) at paras 50-51and 68-71.

[^173]: Ibid (Milan Jeknich) at para. 55

[^174]: Ibid at para. 54

[^175]: Trial Transcript, Yuri Melekhovets at p. 473-474

[^176]: Trial Transcript, Yuri Melekhovets at pp. 422, 424 and 428-432

[^177]: Affidavit of Milan Jeknich at Exhibit M

[^178]: Summary Chart of Customer Cases, Defendants’ Closing Compendium at Tab B

[^180]: Ibid at para. 32

[^181]: Brown, The Law of Defamation, 2nd ed. (looseleaf) (Toronto: Carswell, 1999) at p. 13-14 quoted in Leenen v. Canadian Broadcasting Corp., supra (fn. 41) at para. 107

[^182]: Myers v. Canadian Broadcasting Corp.1999 CarswellOnt 3735, [1999] O.J. No. 4380, 103 O.T.C. 81, 47 C.C.L.T. (2d) 2712 at para. 67 referring to Botiuk v. Toronto Free Press Publications Ltd. (1995), 1216 D.L.R. (4th) 609 (S.C.C.) at 626. Myers was considered by the Court of Appeal at 2001 CanLII 4874 (ON CA), 54 OR (3d) 626; 147 OAC 310; 6 CCLT (3d) 112; [2001] OJ No 2229 (QL). The appeal says little in respect of the defence of “qualified privilege”.

[^183]: Peter A. Downard, Libel, Third Edition, supra (fn. 33) at p. 5

[^184]: United Soils Management Ltd. v. Mohammed, supra (fn. 38) at para. 56 referring to D’Addario v. Smith [2015] O.J. No. 6459, 2015 ONSC 6652 at para. 55 and Cusson v. Quan (2007) 2007 ONCA 771, ONSC 771, 286 D.L.R. (4th) 196 at para. 39

[^185]: Myers v. Canadian Broadcasting Corp., supra (fn. 182) at para. 70, 71 referring Milmo & Rogers, Gatley on Libel and Slander, 9th Ed., 1998 p. 392 (“The media is no different from anyone else and must show the relevant reciprocity of duty and interest”), Jones v. Bennett (1968), 1968 CanLII 126 (SCC), [1969 S.C.R. 277 (S.C.C.) (the defence cannot succeed where the words complained of are published to the public generally: Myers at para. 71) and Hodgson v. Canadian Newspapers Co. (1998), 1998 CanLII 14820 (ON SC), 39 O.R. (3d) 235 (Ont. Ge. Div.) (the defence is not available to newspapers in Canada: Myers at para. 71)

[^186]: Ibid at para. 72

[^187]: (May 28, 1997) Doc A CA C16116 (Ont. C.A.)

[^188]: (1998), 1998 CanLII 14936 (ON SC), 167 D.L.R. (4th) 554 (Ont. Gen. Div.)

[^189]: Grant v. Torstar, supra (fn. 34) at para. 35 and 37

[^191]: Ibid at para. 113 referring to Teskey v. Canadian Newspapers Co. (1989), 1989 CanLII 4392 (ON CA), 68 O.R. (2d) 737, 59 D.L.R. (4th) 709 (C.A.)

[^192]: Ibid at para. 113

[^193]: Affidavit of Rowan Hooper at Exhibit I

[^194]: Trial Transcript, Rowan Hooper at p.1567

[^195]: Affidavit of Peter Aldhous at Exhibit A

[^196]: Affidavit of Tatiana Volossiouk at paras. 1, 5 and 6

[^197]: Myers v. Canadian Broadcasting Corp. supra (fn. 182) at para. 79

[^198]: Grant v. Torstar, supra (fn. 34) at para. 31, Bernstein v. Poon, supra (fn. 148) at para. 103

[^199]: Mainstream Canada v. Staniford, 2013 BCCA 341 at para. 24

[^200]: Ibid at para. 24 quoting from Roos v. Stent and Pretoria Printing Works, Ltd., 1909 T.S. 988 at 998 a decision of the Transvaal Supreme Court (South Africa).

[^201]: WIC Radio Ltd. v. Simpson, [2008] 2 SCR 420, 2008 SCC 40, 293 DLR (4th) 513; 376 NR 80; [2008] 8 WWR 195; 80 BCLR (4th) 1; 256 BCAC 1 at para. 31 referring to Price v. Chicoutimi Pulp Co. (1915), 1915 CanLII 66 (SCC), 51 S.C.R. 179 at p. 194

[^202]: Ibid at para. 26 referring to Ross v. New Brunswick Teachers’ Assn. (2001), 201 D.L.R. (4th) 75, 2001 NBCA 62 at para. 56,

[^203]: Peter Aldhous consulted with the following experts (excluding customers):Dr. Denise Syndercombe Court; Joe Gorman; Kim Gorman; Barry Lenett; Dr. Yuri Melekhovets; Dr. Ali Al-Salih; Dr. Harvey Tenenbaum; Dr. Tom Reid; David Hartshorne; Paul West; Dr. Robert Allen; Lars Mouristen; Dr. Karl Reich; Dr. Rick Staub; Dr. Bruce Budowle; Dr. Gordon Lauc; Dr. David Reich; Dr. Chris Tyler-Smith; Dr. Peter Underhill; Gail Javitt; Beth Goldberg; Eduardo Nunes; Dr. Elizabeth Panke; Dr. Wes Burkhardt;

[^204]: Affidavit of Peter Aldhous at para 34.

[^205]: Closing Submissions and Factum of the Plaintiffs at para. 54

[^206]: Bernstein v. Poon, supra (fn. 37) at para. 115 referring to WIC Radio Ltd. v. Simpson, supra (fn. 201) at para. 34

[^207]: Mainstream Canada v. Staniford, supra (fn. 199) at para. 25 quoting from WIC Radio Ltd. v. Simpson, Supra (fn. 201) at para. 31

[^208]: This is the trial decision found at 2012 BCSC 1433. The appeal citation is at supra (fn. 199)

[^209]: Ibid (BCSC)at paras. 6-7

[^210]: Ibid (BCSC) at paras. 180-181

[^211]: Ibid(BCSC) at para. 183

[^212]: Ibid (C.A.) at para. 43

[^213]: Ibid (C.A.) at para. 45

[^214]: Ibid (C.A.) at para. 47

[^215]: Charter of Rights and Freedoms s. 2(b)

[^216]: Grant v. Torstar, supra at (fn. 34) at paras. 1-3

[^217]: Ibid at para. 7

[^218]: Ibid at para. 112

[^219]: Ibid at para. 111

[^220]: Ibid at para. 113

[^221]: Ibid at para. 114

[^222]: Ibid at paras. 116-117

[^223]: Ibid at paras. 118

[^224]: Closing Submissions and Factum of the Plaintiffs at para. 111

[^225]: Ibid at para. 111