 |
 |
to fill out a simple form to connect to DUI Lawyers in your area.
|
|
|
|
|
Reed v. Maryland9/6/1978
was sufficiently reliable for courtroom acceptance. Acting under a grant given him by the United States Department of Justice Dr. Tosi did undertake just such a study as he described as being needed in the field, which resulted in some 35,000 trials of voice identification. I will not go into all of the details of his study since the details are in the record and have been discussed by courts elsewhere."
Judge McAuliffe commented relative to these tests:
"These tests, which took approximately two years, were completed in December of 1970 and the statistical data was obtained through the utilization of computers. The most significant result for our purposes is that there was a mean of 6.3 percent false identifications and a mean of 12 percent false eliminations.
"It is terribly important to keep in mind the difference between false identification and false elimination. False identifications are simply those in which the examiner says that these two samples match when in fact they do not. False eliminations of course occur when one fails to say that two samples match when in fact they do. Forensically, and I am not talking about the investigatory stage now, I am talking about the prosecution in court of a criminal case, we are vitally concerned with errors of a false identification. We are generally and certainly in this case before the Court not very much concerned with errors of false elimination, although it is a factor that we must consider in making this preliminary determination. But it has nowhere near the significance of errors of false identification. The reasons are obvious. If the examiner is presented with a recording of the actual perpetrator's voice and with a recording of a defendant's voice and he erroneously eliminates a defendant as being the same voice as the offender, he has not thereby caused an innocent man to be convicted. However, if he makes a false identification and says they are the same when in fact they are not, the consequences are much more grave. At least that is our system of American jurisprudence. We are very much concerned that we not convict innocent people and that is our primary concern.
". . . n this country are terribly interested in errors of false identification. Obviously the public
at large has an interest in errors of false elimination, but this will be more of a consequence on the investigatory level than it would be at trial.
"Now we note that these results were reached when considering all of the testing done as to the decisions reached in these four categories. Tosi then asked the computer what the percentage of error of false identification would have been if we considered only those responses of the examiners which were almost certain or fairly certain, thereby not considering the fairly uncertain or almost uncertain responses. This was a valid projection because it equates to forensic situations. We would not permit any expert to testify that he was almost uncertain or fairly uncertain in attempting to make identification. In the medical field we have always required reasonable medical probability and in the scientific field reasonable scientific probability. This is sometimes what I think erroneously is spoken of as reasonable scientific certainty. The answer to that question was that considering only the results generated by the almost certain and fairly certain responses we reduced the error of false identification to 2.4 percent. Furthermore, Tosi has extrapolated, if you will, his finding or projected finding on what he considers to be proper scientific thesis, that the reliability would further be increased and the percentage of error on a false identification would further
Page 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 Maryland DUI Attorneys
DUI Lawyers
|
| |
to fill out a simple form to connect to DUI Lawyers in your area.
|
|