In 2013 (Elhaik 2013) I published a comprehensive study that contrasted two hypotheses proposed to explain the origin of European Jews, namely the Rhineland Hypotheses and the Khazarian Hypotheses. While these hypotheses were not the only ones proposed to answer this question, thus far, they were the only ones that explained the vast number of European Jews in Eastern Europe, the first by supernatural reasoning and the second by the influx of Khazar-converts. In my study, I carried a large number of genetic analyses using data from Jewish and non-Jewish populations and compared the outcomes to the prediction of each hypothesis. The results unequivocally supported the Khazarian Hypothesis. The findings were reported in the prestigious journal Genome Biology and Evolution and were broadly covered in the press. Over a year after its publication, the paper is still ranked #1 most read paper in GBE with its highlighted assay ranked #4. In fact, 2 other papers of mine are listed in the top 15 most read papers. The reviewer comments were published here.
Later in 2013, Behar and colleagues published a critical analysis supposedly meant to test my findings. Their findings were summarized in the title of their paper: “No Evidence from Genome-Wide Data of a Khazar Origin for the Ashkenazi Jews” (Behar et al. 2013). This paper appeared in a redundant depository called Digital Commons with an impact factor of a typical MySpace page as part of a cluster of publication whose sole intent seemed to be the refutation of my findings lead by Noah Rosenberg. Noah Rosenberg also happened to teach a most exciting course called From Generation to Generation: The Genetics of Jewish Populations (Bio 127) in Stanford where the bright minds of the future learn how to invent scientific evidence to validate biblical stories directly from teachers like Behar and colleagues. No need to explain, why my findings did not fit with their indoctrination.
As I never considered the work of Behar and colleagues to be of intellectual challenge of any sort, I delayed my response and waited for another paper of mine to be published. In this paper (Elhaik et al. 2014), myself and others have introduced GPS, the most accurate biogeographical tool that can trace the origin of one’s DNA with an accuracy of home village (in some cases) and resolution of up to 1,000 years. The paper compared GPS to SPA, a tool used by Behar and colleague in their analysis, which will be discussed later on. The findings of my paper were reported in the prestigious journal Nature Communications and were broadly covered by the press.
With that study published, I felt more comfortable discussing the fallacies of the Behar et al. (2013) study, but before delving into that mud, I would like to share with you a piece of evidence that would prove that Bugs Bunny exists. This evidence is based on a conversation between Elmer and Duffy Duck that you might have missed.
Elmer: I keep telling you that Bugs Bunny exists!
Duffy: And I keep telling you that Bugs Bunny is a fictional character.
Elmer: No! He is real and I can prove it to you.
Elmer: I propose a scientific experiment. You do believe in science, don’t you?
Duffy: Oh god yes! If you can do such scientific experiment I’ll believe you beyond and reasonable doubt.
Elmer: Good, I always knew you are a smart ducky! I propose a very simple experiment, let’s go to that hole and see if Bugs bunny exists.
Duffy: I agree!
[Walking to a hole in the ground]
Elmer: Bugs! Bugs come out! Come on Bugs I have a yummy carrot for you!
[Nothing happens… Elmer continues screaming… Duffy gets bored]
Duffy: Ok, I think it is time to admit you were wrong.
Elmer: Excuse me??? How so?
Duffy: Bugs Bunny doesn’t exist.
Elmer: Sure he does! He just can’t hear me, he is probably sleeping.
Elmer: I told you that I will prove to you that he exists and so I did.
Duffy: But your experiment failed! He didn’t come out – he doesn’t exist!
Elmer: I am sorry, maybe you need a refreshment on your scientific background. I never said I will test whether he doesn’t exist. All I said is that I will prove to you that he exists!
We’ll leave these two alone for now and evaluate the role of events. Does Duffy duck really need a refreshment on his scientific background or did Elmer cheat? Actually, it’s a little of both. In science we often do hypotheses testing in which we compare two or more hypotheses that together represent the universe of possibilities. Did Elmer’s test do that? No. Elmer has cleverly deceived Duffy by nullifying the alternative hypothesis. While Duffy perhaps assumed that the experiment was designed to test whether Bugs EXISTS OR DOES NOT EXIST, Elmer agreed to test only if Bugs is EXIST. Therefore, he could only conclude that Bugs EXISTS, as there was no alternative hypothesis.
In scientific terms Elmer proposed to test the following:
Does Bugs Bunny exist?
Null hypothesis: Yes
Alternative hypothesis: –
Whereas Duffy Duck assumed that the test is more of the following:
Does Bugs Bunny exist?
Null hypothesis: Yes
Alternative hypothesis: NO
So who is at fault here? That’s a judgment call. Elmer has deceived Duffy but Duffy has agreed to the rules and is therefore as guilty.
And from Loony Tunes we’ll switch to the genetics of Jews, which you may be shocked to find how little different they are. The reason is that almost every study aims to study the origin of European Jews has nullified the alternative hypothesis and concluded a Middle Eastern origin, in agreement with the null hypothesis. This is what a typical papers supposedly aims to study the origin of European Jews looks like:
Do Jews have a Middle Eastern origin?
Null hypothesis: Yes.
Alternative hypothesis: –
In that respect, my study (Elhaik 2013) was one of the few scientific studies that actually studied the origin of European Jews with the hypotheses phrased like that:
What is the origin of European Jews?
Null hypothesis: Middle Easte
Alternative hypothesis: Caucasus
When a hypothesis is phrased this way, rather than compare X to NOT X, one is at risk of missing alternative explanations. For that reason, I included only hypotheses that proposed an explanation to the existence of a large number of European Jews in Eastern Europe, certainly a valuable criteria and showed that the two hypotheses are distinct.
Let us now turn to the Behar et al. (2013) study. This paper is also “first” in something. It is the first study, in which the Rhineland Hypothesis was not the null hypothesis. Instead, the authors used the Khazarian Hypothesis as the null hypothesis. So what was their alternative hypothesis, you surely ask now, that’s right… it was nullified! Once again.
Are European Jews similar to the Khazar proxies?
Null hypothesis: No
Alternative hypothesis: –
Behar et al. (2013) never intended to test my findings (Elhaik 2013). Had they done so they would have found that they are correct. Behar et al. (2013) simply went Elmer and pretended to do a scientific experiment. So what exactly did they do? They looked at populations from the Caucasus and considered them Khazarian proxies. Just like Elmer proposing to prove that Bugs exists, Behar et al. (2013) proposed to prove that European Jews are dissimilar from the Khazarian proxies. Just like Elmer they avoided covering the entire universe of possibilities and just like Elmer who did not define what would be considered evidence to Bugs existence, the authors did not define how similar one needs to be from the Khazar proxies to be considered a Khazar. It also doesn’t matter of course. In the absence of an alternative hypothesis, one HAS to conclude that the null is correct. Therefore, when European Jews happened to cluster around Turkey the authors proudly and rightfully announced that the findings support their null hypothesis, or in other words, that European Jews are just too far for their taste from the Khazarian proxies to be considered Khazars.
Was this a scientific study? It is had we lived in Loony Tunes.
The authors committed a final fallacy called argumentum e silentio where the conclusion is based on the absence of evidence, rather than the existence of evidence by concluding (“Thus, analysis of Ashkenazi Jews together with a large sample from the region of the Khazar Khaganate corroborates the earlier results that Ashkenazi Jews derive their ancestry primarily from populations of the Middle East and Europe”). The intelligent reader may rightfully jump and cry “Wow, who said they came from the Middle East at all?” Indeed, no one. With the alternative hypothesis nullified and the null hypothesis defined as where Jews are NOT from, there is no scientific way of reasoning that European Jews came from the Middle East. Recall that previous studies did not test any alternative hypothesis and thereby one cannot rely on them. In fact, Jews could have come from Asia, Africa, Mars, or Looney Tunes for that matter, who knows? No one has tested any alternative analysis. At best (assuming the alternative hypothesis is implied) the authors can conclude that European Jews are not of Khazarian origin, but no one drew a line between what is and what is not Khazarian because this was not really the purpose of this piece. The authors cannot even cite the Bible as they did in the past, because biblical-based origins are not part of their universe of possibilities. Atzmon et al. (2010) used a similar technique of rejecting the Khazarian hypothesis without testing it. To that, Dan Graur, my PhD supervisor would say: if you know the answer why do the experiment? just write the damn paper and stop wasting my precious time! Having a sharp object within his vicinity could have spared you from this little speech. His irony may be lost on the authors of Atzmon et al. and Behar et al. who repeatedly shown that when it comes to studying the origin of Jews science can become much more colorful. The intelligent reader may take some comfort in the fact the authors agree, for the first time, to consider Europe (wherever that is) as the origin of European Jews, but that does not add validity to this piece.
One of the results by Behar et al. (2013) caught my attention. It was the SPA analysis, based on a tool published by Yang et al. (2012) in Nature Genetics. SPA was presented as a tool that can infer the biogeographical origin of populations using a PCA-like approach, however this tool is a hoax. Elhaik et al. (2014) compared the predictions of this tool to those of GPS and showed that it has ~2% accuracy. In other words, 98% of the time it would predict the wrong geographical origin of the population in question. Of course, no better way of proving that a tool is useless than its creators’ own results. We illustrated the complete lack of accuracy of SPA using the results that Yang et al. (2012) buried in their supplementary table. Halperin, a joint authors to the Behar et al. and Yang et al. papers had to be aware of the fact that SPA is absolutely useless. But, I supposed that if Yang and co-authors felt comfortable enough publishing a completely useless tool there is no reason why Halperin would not apply it to study the origin of European Jews. By the same reasoning, he can apply it to study the origin of Martians, Female Terminators, or Mickey Mouse, it would make no difference because SPA returns random results that depend on the populations included in the cohort. But again, let us assume that SPA does work and that Halperin’s analysis has some vague scientific merit. Do you truly believe that tracing European Jews to Turkey rules out their Khazarian origin, provided that the Khazars were “a conglomerate of mostly Turkic tribes” as the authors noted?
I seriously doubt that.
And so we come to the most important question: Is Bugs Bunny a Jew? Fortunately, we don’t need to ask Behar and colleagues that question as film expert David Yehuda Stern already answered it with a big YES.
Behar DM, Metspalu M, Baran Y, Kopelman NM, Yunusbayev B, Gladstein A, Tzur S, Sahakyan H, Bahmanimehr A, Yepiskoposyan L. 2013. No Evidence from Genome-Wide Data of a Khazar Origin for the Ashkenazi Jews.
Elhaik E. 2013. The Missing Link of Jewish European Ancestry: Contrasting the Rhineland and the Khazarian Hypotheses. Genome Biology and Evolution. 5:61-74.
Elhaik E, Tatarinova T, Chebotarev D, Piras IS, Calò CM, De Montis A, Atzori M, Marini M, Tofanelli S, Francalacci P. 2014. Geographic population structure analysis of worldwide human populations infers their biogeographical origins. Nature communications. 5.
Yang WY, Novembre J, Eskin E, Halperin E. 2012. A model-based approach for analysis of spatial structure in genetic data. Nat. Genet. 44:725-731.