The April 3, 2007 news article about the Department of Defense investigation into MIT’s Lincoln Laboratory misstated the nature of the review on which the charges were based. According to the MIT News Office, the review evaluated a piece of software developed by military contractor TRW that was intended for, but never used in, a missile defense flight test. The researchers used data from an earlier test to examine whether the software worked as claimed by TRW. They were not asked to evaluate the flight test.
A Department of Defense investigation into long-standing allegations of fraud in a Lincoln Laboratory-led review of national missile defense tests has concluded that neither the review team nor Lincoln Laboratory management is guilty of research misconduct. The DoD investigative report, released Friday, pointed out problems with how the review’s results were presented — namely that critical information was omitted in the study — but said that these actions “did not rise to the level of research misconduct.”
Entering its eighth year, the dispute regarding the Lincoln Laboratory review has raised questions about how MIT can manage and oversee classified research for the United States government if they are unable to conduct independent investigations because of the classified nature of the information involved.
The DoD report recommends that the charges of research misconduct be dismissed and that the two authors of the study, Charles K. Meins Jr. ’75 and Ming-Jer Tsai, both Lincoln Laboratory scientists, be publicly exonerated. Additionally, the report states that certain requirements in MIT research misconduct policies should be strengthened and that the DoD itself should prepare a “lessons-learned report” regarding this case several months from now.
MIT considers the matter closed and will now work toward addressing the recommendations made regarding MIT policies and procedures, said Claude R. Canizares, vice president for research and associate provost.
The full report is available at http://web.mit.edu/provost/reports/InvestigationReport01292007.pdf and http://www.defenselink.mil/home/pdf/Investigation_Report_27JAN.pdf.
The allegations center around the Lincoln Laboratory’s classified “POET” Study, which reviewed a 1997 Pentagon missile test. (For a timeline of events, see page 17.) The 1999 POET Study concluded that the test results were valid and that the Pentagon’s national missile defense system could distinguish actual warheads from warhead-shaped balloon decoys in outer space. This conclusion was questioned by critics, including MIT Professor Theodore A. Postol ’67, who believe that the original test results were exaggerated, that the POET Study has contradictions, and that the defense system would not work against a real attack.
“The report basically verifies every allegation I have made, but manages to conclude it was not scientific misconduct,” Postol said. “I find that very hard to understand.”
According to the report, Postol raised issues of financial irregularities at MIT and suggested that MIT administration and Lincoln Laboratory management may be guilty of misleading Congress and federal investigators. “I would hardly call this a closed matter,” Postol said. “I am ready to proceed to the next step.”
Postol alleges that MIT officials, including former President Charles M. Vest, have made false statements to members of Congress when describing the contents of a 2002 MIT inquiry report into the Lincoln Laboratory allegations as highly sensitive and classified. The DoD report also described certain parts of the 2002 report to be classified. Postol claims that the inquiry report did not contain highly classified information. Postol said that the actions of these administrators “could be considered obstruction of justice.”
Postol said that he brought these issues to the attention of Massachusetts Congressman John F. Tierney, chairman of the House subcommittee on national security and foreign affairs. Tierney is in the process of collecting information now, Postol said.
Tierney was out of the country yesterday and unable to comment. Canizares said that he was unaware of the congressman’s involvement and could not comment on the situation.
Provost L. Rafael Reif, in a letter to the MIT faculty (http://web.mit.edu/provost/letters/letter03302007.html), said that he has determined neither Meins nor Tsai engaged in academic misconduct, based on the conclusions of the investigation. He also said that he would “consider carefully” the recommendations of the report regarding MIT research misconduct policies and work closely with faculty and administration “on steps going forward.”
Reif did not respond to an e-mail request for comment.
“The investigator has done a great job with integrity and trying to get the truth,” Tsai said. “It is certainly a relief for me. It has been a long time. It was an unnecessary burden on me and my family. I don’t know how this came about and why it became such a big issue.” Meins did not respond to a phone call for comment.
“I think that the investigation done by the DoD was very complete,” Canizares said. “It was fair and intensive.” In some ways, the investigator was able to do more than an MIT investigator, he said, because many of the people interviewed would not have talked to MIT.
Postol said that MIT turning oversight responsibility over to the DoD was “unacceptable.” “MIT gets paid handsomely,” Postol said, for its involvement with Lincoln Laboratory. Part of what MIT is paid for is oversight, he said.
Postol has previously suggested that an MIT investigation could have been performed using only the publicly available information. Godfrey, in the DoD report, disagreed, stating that many key documents were classified and that a number of the investigation interviews were conducted at a classified level.
In an interview with The Boston Globe, Godfrey said Postol could not properly analyze the POET study because he did not have access to classified materials. He also said that the version of the report Postol used was a redacted version which had the classified information blacked out. “It’s simply unjust to be accusing people of misconduct based on half a report,” Godfrey said to the Globe.
Postol, in 2001, began sending letters to members of the MIT administration, calling the POET report “a serious case of scientific fraud.” This led to a preliminary inquiry by then-Course XVI Chair Edward F. Crawley ’76, who found that the allegations merited a full investigation because “sufficient inconsistencies, open issues, and needs for detailed rectification of facts” remained. This 2002 inquiry specifically identified six open issues that remained; these six issues were the focus of the DoD investigation.
Based on MIT policy, an investigation should have begun right away. However, the investigation was stalled because the Missile Defense Agency declared that all information related to the POET study “had to be classified in order to protect national security,” according to an MIT statement released in December 2004.
In March 2006, MIT and the Pentagon agreed that MIT would diverge from its written rules and the DoD would conduct its own investigation. At MIT’s insistence, Norman R. Augustine, former member of the MIT Corporation and former Chairman and CEO of Lockheed Martin Corp., was brought in as an advisor to the investigator, “to help assure an impartial and thorough investigation,” Reif wrote in an e-mail to the faculty at the time.
“In my opinion, the investigator, [Brandon B.] Godfrey [from the Air Force Office of Scientific Research], carried out his responsibilities very competently, with extraordinary diligence and a sincere effort to ferret out the truth,” Augustine wrote in a forward to the DoD report.
Augustine added in the forward to the report that he was granted full access to all information related to the investigations, including classified information, and that he actively participated in the interviews and reviewed over 100 documents. Canizares confirmed in an interview that Augustine had full access to all documents.
Godfrey wrote that there was an excessive delay in addressing the allegations, that “the DoD did not follow the federal requirement to initiate an investigation responding to the Inquiry Report in a reasonable period of time.” Augustine, in the forward, also wrote that the investigation was complicated by the length of time that had passed.
“In several instances it was difficult to locate relevant documents (although all documents sought eventually were obtained);” Augustine wrote, “in others the firms involved in the events ceased to exist as independent entities; and in still others, individuals retired and their security clearances lapsed, people changed jobs, memories — not unexpectedly — faded; and one individual died.”
The delay generally stemmed from a lack of agreement between the Missile Defense Agency (part of the DoD) and MIT regarding the handling and viewing of classified material, according to Chris Isleib from the DoD’s press desk. “The policy negotiations on that issue took some time.”
Canizares said that the delay was unfortunate and that MIT pushed the DoD to follow its own procedures.
According to Isleib, Godfrey will be preparing a lessons-learned document for the DoD in two or three months.
Study omitted sensor information
Godfrey investigated the six open issues identified in the 2002 MIT inquiry, which included concerns about the omission of critical information regarding the sensors used in the missile defense system, as well as alleged improper manipulation of flight data. None of the allegations of research misconduct were substantiated “by a preponderance of evidence” as is necessary based on the definition of research misconduct provided by the Office of Science and Technology Policy.
Regarding the sensors in particular, information about inconsistencies in calibration of the sensors was left out of the POET Study. Godfrey stated, in the DoD report, that “Omitting a discussion of such critical information in a scientific journal article might have constituted research misconduct. However, this was not a scientific journal article, and the two LL members of the POET team responded that the classified POET report was meant only for about 10 people, that the sensor problems were known to those involved in IFT-1A, and that the sensor produced adequate data anyways for at least 17 seconds, all of which were true.”
“Nonetheless, this investigation concludes that sensor performance was so important that it should have been discussed in the report,” the report continues. “This omission does not, however, rise to the level of research misconduct, due to the extenuating circumstances just described.”
Recommendations for MIT
Godfrey specifically criticized MIT’s handling of the 2002 preliminary inquiry of the allegations. The report states that the inquiry began without “clearly written allegations” and that the confidentiality of the process was compromised as “copies of a draft version of the Inquiry Report were not controlled adequately.” “Some information found its way into newspapers,” the report states. “Apparently, this violation of federal and MIT policy was not investigated.”
The DoD report also states that MIT failed to follow its own policy and did not include a summary of the rebuttal offered to MIT management by the alleged offenders, Meins and Tsai, in the inquiry report. According to the DoD report, Crawley, who led the inquiry, said that he was not aware of the rebuttal.
Canizares pointed out that most of the recommendations made in the DoD report are similar to the recommendations made by an ad hoc committee he led last year that looked into factors that complicated the resolution and investigation into the allegations. (The report is available at http://mit.edu/provost/letter-5-19-06.html.)
Canizares said that he decided to wait on the DoD report before discussing and possibly implementing the recommendations of the ad hoc committee. “It is definitely our intention to take a look at those,” Canizares said. “We are beginning to look at how to do that.” He said he hopes for expediency.