Editorial: Statistics and "The lost tomb of Jesus"

The Annals of Applie d Statistics 2008, V ol. 2, No. 1, 1–2 DOI: 10.1214 /08-A OAS162 c  Institute of Mathematical Statistics , 2 008 EDITORIAL: ST A TISTICS AND “THE LOST TOMB OF JESUS” By Stephe n E . Fienberg Carne gie M el lon University What mak es a problem suitable for statistical analysis? Are historical and religious questions addressab le usin g statistica l calculatio ns? Suc h issu es ha ve long b een debated in the statistical communit y and statisticians and others ha ve used h istorical inf ormation and texts to analyze su c h questions as the ec onomics of sla v ery , the authorship of t he F ederalist P ap ers and the question of the existence of Go d. But w hat ab out historical and r eligious attributions asso ciate d with inf orm ation gathered f r om arc heologic al finds? In 1980, a construction crew w orking in the J erusalem neigh b orho o d of East T alpiot stu m b led up on a c rypt. Arc haeologist s from the Isr ael A n tiqui- ties Authority came to the scene an d found 10 limestone burial b oxes, kno w n as ossuaries, in the cryp t. Six of these h ad inscrip tions. The remains found in the ossuaries w ere reburied, as required by Jewish religio u s traditio n, a nd the ossu aries w er e catalogued and stored in a wa rehouse. The inscriptions on th e ossu aries w ere catalog ued and pub lished by Rahmani ( 1994 ) and b y Kloner ( 1996 ) bu t th ere rep orts did not r eceiv e w idespread p u blic atten tion. F ast f orward to M arc h 20 07, when a television “do cudrama” aired on The Disc overy Channel ent itled “The Lost T om b of Jesus” 1 touc hed off a p ublic and religious con trov er s y—one only need think ab out the title to see why there migh t b e a con tro v ersy! The pr ogram, and a simultaneously published b o ok [Jacob ovic i and Pelleg r ino ( 2007 )], describ ed the “rediscov ery” of the East T alpiot archeo logica l find and they presented interpretatio n s of the ossuary inscriptions from a n um b er of p ersp ectiv es. Among these wa s a sta- tistical calculation attribute d to the statistician Andrey F euerv erger: “that the o dds that all six names wo u ld app ear together in one tom b are 1 in 600, calculate d conserv ativ ely—or p ossibly ev en as m u c h as one in one million.” A t ab out this time, F euerv erger su bmitted a pap er to The Anna ls of Applie d Statistics (A OA S) for review, but its conte nts remained confi den- tial and only a r ough outline of the details of his calculations w as publicly Received January 2008; rev ised January 2008. This is an e le c tronic repr int of the orig inal a rticle published by the Institute of Ma thematical Statistics in The Annals of Applie d Statistics , 2008, V o l. 2, No . 1, 1– 2 . This repr int differs from the original in pa gination and t ypo graphic detail. 1 “The Lost T omb of Jesus.” Disc overy Channel , March 4, 2007. http://dsc.dis co very . com/con vergence/tom b /tom b.html . 1 2 S. F. FIENBERG a v ailable [Mims ( 2007 )]. Commentary regarding F euer verger’s statistical cal- culation quic kly app eared on the web. W as it really a Bay esian calculatio n? On w h at assu m ptions were the statistical argument s b ased? Most criticism fo cused not directly on t he actual statisti cal arguments but on ho w they w ere p ortray ed by the do cumen tary’s pro d ucers and inte r preted b y others. And the con tr o ve rsy o ver the broader interpretati on and claims rega rding the origin of th e East T alpiot tom b raged on. In July 2007 at the J oin t Statistical Meetings in Salt Lake Cit y , F euerve rger ga v e the first public airing of the details of his work and three discussants present ed alternativ e p ersp ectiv es. The pap er itself und erw ent an extensiv e review pro cess a nd a substantia lly revised v ersion app ears in this issue of A OAS [F euerv erger ( 2008 )]. It includes photographs, d etailed d iscu ssion of p ossible data on n ames from ancient sources, th e assumptions up on whic h the analysis was based, and a nov el p -v alue cal culation. The pap er is ac- companied b y a series of d etailed discus s ions and critiques, sev eral of which reframe the statistical problem f rom a Ba yesia n p ersp ectiv e. The AO AS editors encourage our readers to jud ge for themselv es the p er- suasiv eness of the assumptions, the data, and the calculatio ns p erf ormed by F euerverge r, esp ecially in ligh t of the criticisms voi ced in the extended dis- cussion that follo ws his pap er, and his resp onse. Intereste d readers ma y then wish to explore the extensive n onstatistical discussion of the East T alpiot a v ailable in print and on the web. REFERENCES Feuer verger, A . (2008). Statistical analysis of an archeologica l find (with discussion). Ann . Appl. Statist. 2 3–54. Ja cobo vici, S. and Pellegrino, C. (2007). The Jesus F amily T omb : The Disc overy , the Investigation, and the Evidenc e That Could Change History. Harp erone, New Y ork. Kloner, A . (1996). A tomb with inscribed ossuaries in East T alpiyot, Jerusalem. Atiqot 29 15–22. Mims, C. (2007). Q&A with the statistician who cal culated the o dds that this tom b b elonged to Jesus. Scientific Americ an I n F o cus , Marc h 2, 2007. A v ailable at http://sci am.com/ar ticle.cfm?articleID=13C42878- E7F2- 99DF- 3B6D16A9656A12FF . Rahmani, L. Y. (1994). A C atalo gue of Jewish Ossuaries in the Col le ctions of the State of I sr ael. Israe l An tiqu ities Authority , I srael A cademy of S ciences and Humanities, Jerusalem. Dep ar tment of St a tistics and Machine Learning Carnegie Mellon Un iversity Pittsburgh, Pen nsyl v ania 15213 USA E-mail: fien b erg@stat.cmu.edu