Snake Oil, Scapegoats and the Hare Krishnas: The CPO's Use of Bogus Science
Posted June 9, 2009
While completing my doctoral training in psychology, I had an assessment case where a couple had been accused of child abuse and subsequently had a child removed from their care by Child Protective Services. I was to administer a psychological evaluation to both parents in order to shed light on their guilt or innocence. I had no idea what kind of psychological test might be helpful since psychological tests are usually made to measure things like depression or psychotic thinking, not child abuse potential. My supervisors gave me carte blanche to communicate with any expert in the U.S. in order to find an appropriate tool to use. I talked with many experts. The overwhelming conclusion I heard was that there is no such test. In other words, there is no psychological test that is capable of determining if person A abused child B. I was warned to be skeptical of those who say they have devised such a test.
Imagine my surprise when a short time later it came to my attention that the CPO was training its investigators and temple presidents to use the Luscher Color Test to identify child abusers. Essentially, this test asks a person to identify their color preferences from an array of colors. A particular set of colors was said to be characteristic of a child abuser. This test has no accepted validity whatsoever among professionals in the field. Nevertheless, it was being used as evidence by the CPO to allege that certain individuals were child abusers! I argued vehemently against its use. After about a year, the head of the CPO at the time reluctantly agreed to discontinue the use of the test. Nevertheless, another year later, I was present when the CPO head received a telephone call from a devotee in India who had just administered the Luscher Color Test and had “discovered” a child abuser. This ridiculous test falsely identified many devotees as child abusers.
Later I was concerned to discover an ISKCON sannyasi had been required by the CPO to undergo a psychological evaluation at his own expense in the course of the CPO investigation against him. Correspondence forwarded to me from a North American GBC member indicated the contention of the CPO was that the sannyasi was somehow a “demon in the body of a devotee” who took pleasure in harming children. This was to be “proven” through psychological testing. The psychological term for “demon” is “psychopathic deviate”. I spoke to the sannyasi and informed him there was no psychological test that could determine if he were a child abuser. He elected to be humble, go along with the CPO demands, and undergo psychological testing anyway. I advised him that prior to the evaluation, he should at least get a commitment from the CPO that if the results did not show him to be a child abuser, his case would be dropped. If test results have the power to show a person’s guilt, shouldn’t they also have the power to exonerate him? I also recommended he receive the evaluation from a neutral evaluator, since the one chosen by the CPO would likely be biased, having been briefed by the CPO. The sannyasi made these requests but the CPO disregarded them. To his detriment, the sannyasi chose to be cooperative with the CPO, wanting to accept responsibility for his actions, and underwent psychological testing through someone chosen by the CPO.
I was interested in the test results, so the sannyasi sent them to me. The test results showed the sannyasi did not have psychopathic tendencies. If one accepted the CPO’s premise that the test could differentiate between an abuser and non-abuser, the sannyasi should have been exonerated, but the evaluator disregarded the results of the tests! (We can assume that if the results had been otherwise they would have been used as a basis for his conviction). Using information drawn from discussions with prosecution witnesses, the evaluator twisted his findings to conclude the sannyasi did, in fact, have psychopathic tendencies! In reviewing the report it was clear the evaluator did not even take the time to interview witnesses for the defense. He did not have the required qualifications to interpret psychological tests; he was not a psychologist at all. And lastly, the CPO-chosen evaluator turned out to be an ex-colleague and friend of a CPO investigator who had a well-known agenda to convict the sannyasi!
In my profession, a report done in this manner by an evaluator without the required qualifications and with such a conflict of interest would be considered malpractice, plain and simple. Nevertheless, the report was entered uncritically into evidence by the CPO to support a conviction against the sannyasi. Hence, it is another good example of the CPO using a scientific façade to gain a conviction.
In another case, instead of weighing the actual evidence, the judges were directed to use a junk science method, the SCAN-View Questionnaire. This test had no proven track record whatsoever in accurately separating the guilty from the innocent (in test parlance this is called a test’s reliability and validity. See below for a more detailed explanation*). There is a clear and timely example of this test’s resulting in false conclusions about guilt. On the test author’s own website he boldly declares that his test methods prove JonBenét Ramsey’s father is guilty of her murder. However, DNA evidence recently exonerated the father.
Even if someone unwittingly argues there was nothing wrong with the test, there are still problems. The test was not used as it was supposed to be used. If you don’t follow the instructions on how to use the test, the results are, of course, invalid. The test author warned that its accuracy would be suspect if it were used to investigate sex-related crimes. His letter to that effect is contained in the accused devotee’s case file; nonetheless, the CPO chose to use the method in a sex crime investigation. In addition, the test author’s procedure indicates the test is to be administered before someone has been informed of their being a suspect in a crime. After being identified as a suspect, a person’s heightened anxiety about being falsely accused will likely affect his responses. Nevertheless, the test was administered after the devotee was accused. To add insult to injury, the case file shows the CPO director carried on back-and-forth correspondence with the test author about the case prior to the author’s interpretation of the responses, eliminating any façade of impartiality on the part of the test interpreter. That the results of this so-called test were used to obtain a conviction is ludicrous.
The above examples show a disturbing pattern. With the prevailing sentiment in ISKCON against child abuse, it appears the CPO has repeatedly been using pseudoscience to generate evidence and lend a veneer of scientific authority to their procedures in order to gain convictions. Regardless of whether someone is guilty or innocent, a bad system of justice is another embarrassment for ISKCON.
My suggestion to ISKCON is simple. Use the civil authorities and court system to investigate and prosecute child abuse in the United States and in other countries where such systems exist. It is shortsighted and pretentious to think ISKCON has the money, expertise and experience to develop a more effective system. If a person is found guilty in the civil system, the CPO can make additional decisions.
In my professional opinion, the CPO has never properly considered the possibility that devotees they investigated may have been falsely accused. Their understanding of child abuse was misinformed and apparently overly influenced by prevailing public sentiment. This resulted in their having little regard for the accused devotees’ spiritual or material welfare. The CPO has misapplied psychological science and therefore it is likely they have arrived at erroneous conclusions.
Through the use of the methods cited above, I believe the CPO has been responsible for causing an inordinate amount of collateral damage in their attempts to protect children. This has resulted in the serious and unwarranted traumatization of accused devotees and their families, who have been suffering sanctions, shame and vilification for years from a CPO decision guided by pseudoscience and quackery. Enough is enough.
I appeal to the International GBC body to please take the steps necessary to address the problem. I believe there is ample evidence to support overturning the CPO decision in at least one case. In addition, I believe consideration should be given to suspending the investigative and judicial functions of the CPO immediately and until it can be determined if, how and when those functions can be carried out it a judicious manner.
* Reliability. Reliability refers to the ability of a test to yield the same results over time no matter who interprets the test. So, if trained persons A and B interpret a test and their conclusions differ markedly from one another, the test is not reliable. Test reliability is one of the very first ways to see if a test is worth anything. Measures of reliability are not difficult to obtain. All the author of a test has to do is get a number of people to interpret the same test and see if their conclusions match. If they do, the test is reliable. If not, it is not. The author of the test the CPO used to convict has not published any information about its reliability. It is standard practice in the field to publish reliability data on any assessment tools. Why has the author of the test used not done so? Undoubtedly because it is not credible and certainly not recognized. Secular authorities who have reviewed the test have concluded that there is no evidence that it is reliable. To illustrate, suppose you were asked to test the usefulness of a new experimental blood-pressure-measuring device. If every person who used the meter on the same patient at about the same time got a markedly different reading, what would you conclude? You wouldn’t know if the patient had high blood pressure, low blood pressure, or normal blood pressure because the test is not reliable. The same is true of the test used to convict the devotee. Its conclusions are worthless because it has no demonstrated reliability.
* Validity. The validity of a test refers to the test’s ability to accurately measure what it is supposed to measure. Was the method used in this case really able to assess guilt? It is well known that if a test is not reliable, it cannot be valid. In addition, it is standard practice for authors of assessment instruments to publish data about a test’s validity. The test author could have easily published his research. The fact that he did not leads to the conclusion that the test is unscientific. From the point of view of standard principles of assessment, the test results must be disregarded.