where polygraph results have been admitted pursuant to stipulation, the Dean court noted such stipulations encourage discussion and eventual agreement with regard not only to the general subject of the reliability of polygraph testing but also to more specific subjects such as the qualifications required of the examiner, the designation of the examiner, the phrasing of the test questions, and the specification of the condition under which the test is to be given. 307 N.W.2d at 637. Defendants' unilateral test satisfied none of these safeguards and conditions.
Fourth. Even on their own terms, the results of the Gallup poll submitted by defendants are not impressive. Two out of three of those responding to the telephone poll had never used polygraphs to determine whether an individual was telling the truth, and only 33 percent of all the respondents described themselves as "very informed" regarding polygraph testing.
Thus, the expertise of the body of those polled is somewhat suspect. Moreover, only 62 percent of those responding stated that they regarded polygraph tests as a useful diagnostic tool when considered with other information; 35 percent found them to have "little weight" or no usefulness.
That is hardly the stuff which would justify one in concluding that now, unlike in the past, polygraph test results have found broad scientific acceptability.
Fifth. It is obvious that, had the results of the polygraph examination shown that Drinkhall was not telling the truth, neither plaintiffs nor the Court or jury would have ever heard about the examination or the results.
Whatever other consequences might have flowed from this peculiar method of insuring scientific objectivity, it was bound to have an impact on the state of mind with which Drinkhall approached the tests. One of the important factors that many of those who rely on polygraphs insist on is that the individual taking it should know that something real is at stake. Absent such conditions, no results of significance can be expected.
The Dow Jones defendants respond that Drinkhall was under just such pressure because he knew that his employers would not take kindly to lying and that, if he did not pass the test, they would take appropriate action.
It would not be fair to the plaintiffs or to the search for truth in this litigation to depend on that assumption for equating the Drinkhall test with one that is taken with full knowledge of and participation by both parties. No one can know whether, had Drinkhall failed the test, he would have been subjected to consequences remotely similar to those which would have followed from his failure to pass a polygraph test conducted under impartial auspices -- consequences including a possibly very large verdict against both him and his employers, with the attendant loss of prestige and credibility. The Drinkhall test was not a fair test, and even if, contrary to the Court's conclusion, polygraphs in general are valid scientifically, the results of this particular test cannot be regarded as accurate for admission into evidence.
Defendants would use the results of one flawed instrument -- the public opinion poll -- to validate the scientific acceptability of another flawed instrument -- the polygraph. That will not do.
There is a second major obstacle to the use of polygraph tests as evidence in judicial proceedings. A considerable number of courts have held that public policy reasons militate against the admission into evidence even of scientifically-validated polygraph results because such results are likely unduly to sway the jury.
These courts suggest that the relevance of the polygraph results is outweighed by the possibility that the jury will be misled, prejudiced, and indeed displaced by the polygraph and its operators.
This Court endorses the observation of Chief Judge Gibson for the Eighth Circuit in United States v. Alexander, supra, 526 F.2d at 169, to the effect that
The jury institution was created and maintained due to the public reluctance 'to entrust plenary powers over the life and liberty of the citizen to one judge or to a group of judges. . . . However, in many cases where polygraph evidence is admitted, a single person, the polygraphists, will give testimony which will often be the determinative factor as to the guilt or innocence of a defendant in a jury-tried case. This would deprive the defendant of the common sense and collective judgment of his peers, derived after weighing facts and considering the credibility of witnesses, which had been the hallmark of the jury function.
Defendants recognize the danger that the jury will give undue credence to the machine (and its human operator), and on that issue, too, their answer is a Gallup poll. In December 1982, the Gallup organization interviewed by telephone 1,004 potential District of Columbia jurors. 15 percent of those questioned stated that they would pay more attention to polygraph results than to the testimony of a witness,
11 percent responded that they would ignore the polygraph results, and 72 percent stated that they would pay greater attention to what the witness said but would also consider carefully the polygraph results. Defendants argue on the basis of the fact that 72 percent gave what they regard as the "common sense answer" there is no danger that the jury would be unduly influenced or swayed. That conclusion is likewise unpersuasive.
One clue to the difficulty is precisely that 72 percent gave the "common sense answer." No one with any common sense would, when interviewed by a polling organization, give anything but the answer that he would consider both the witness and the polygraph. Indeed, almost any person would be embarrassed to state that, if called to jury duty, he would give greater weight to a machine than to a witness. It is, however, an enormous leap from an answer given to a telephone inquiry by a poll-taker to the conclusion that the same individual, after sitting as a juror for six to ten weeks, would not decide that the machine was more trustworthy than the confusing stories of several dozen witnesses.
The credibility of Jim Drinkhall shapes up as a crucial issue in this action. It appears that on many issues it will be his word against that of others, and the surrounding circumstances are likely to be confused and confusing. If the Court were to admit the polygraph evidence, defendants could be expected to hammer home to the jury that the machine was the best, the only way out of the human contradictions, and that, indeed, the jury would be derelict in its duties were it to reject the machine's verdict. In short, in significant measure, trial by machine would replace trial by jury.
The polygraph is not scientifically reliable. Yet there is here a real danger that, notwithstanding the lack of reliability, it (or rather its operators) will be the real finders of fact in this case.
For the reasons stated, defendants' motion to admit polygraph evidence will be denied, first because as a matter of law and precedent in this jurisdiction such evidence is not admissible and second, because, if the Court has discretion, it concludes that the scientific reliability of such evidence is substantially outweighed by its confusing effect on the jury. See Rule 403, F.R. Evid.