Canadian court ponders DNA pop stats admissibility

[Howard]

From Saturday's Toronto Globe and Mail
(Just a note for those unfamiliar with Canada's legal system. The fact that
the appeal court has consented to hear this ­ and will rule on it ­ puts it
a step beyond lawyers dreaming up loopholes. My own opinion is that they'll
dismiss it. An Ontario appeal court ruling is subject to further appeal in
the ultimate Supreme Court of Canada ... where it will likely go next in
any event.)

-hm

BAR STATISTICAL PROBABILITIES
ON DNA, CANADIAN COURT URGED

"Jurors distracted from seeking the truth
when presented with ratios in the millions"

BY THOMAS CLARIDGE
Courts Reporter
TORONTO ­ The Ontario Court of Appeal has been asked to prohibit DNA
experts from expressing their findings in terms of statistical
probabilities at criminal jury trials.
The request was made this week by lawyers for John Carlos Terceira, a
maintenance worker who was convicted of first-degree murder in the sex
slaying death of a six-year-old girl.
The asphyxiated victim, Andrea Atkinson, was found in the boiler room of
the apartment building in which she lived with her mother and where
Terceira worked.
The 1993 murder trial was one of the first in Canada dominated by DNA
evidence, with Crown attorney Paul Culver portraying DNA from the outset of
the trial as a "calling card" Terceira had left at the murder scene.
The DNA comparison was between semen found on the little girl's leotards
and on the floor near her body and a blood sample Terceira had volunteered
as one of several suspects in the case.
In describing the results of the profiling, Crown witness Pamela Newall
told the jury that the semen found on the floor had a profile that would be
expected to occur only once in a population of 8.3 million Caucasians. The
leotard stain had yielded a result that was slightly less conclusive, the
DNA profile being expected once in a population of 1.8 million.
The appeal court heard that at the time of Terceira's trial, there was no
general acceptance of the methodology to determine the statistical
likelihood of a random match.
"Even today, there is considerable controversy in the scientific community
surrounding this issue," lawyer David Tanovich told the three-judge panel
in a written submission.
Contending that the trial judge had erred in admitting the statistical
evidence, Tanovich said the "prejudicial impact of DNA numbers far
outweighed their probative value." He argued that introduction of such huge
numbers "would likely have had an overwhelming impact on the jury."
He went on to ask the court to impose a "bright line rule" ­ a mandatory
requirement ­ "that DNA statistical numbers are inadmissible in criminal
trials."
Tanovich said numbers often expressed in the millions, as in the Terceira
trial, were likely to distract jurors from their job of seeking the truth.
"Vividness theory suggests that people give inferential weight to evidence
in proportion to its vividness or memorability."
However, Crown lawyers Shawn Porter and Michal Fairburn noted that Tanovich
did not question the reliability of the statistical method used in the case
or find fault with the trial judge's charge to the jury on the DNA
evidence.
"Neither the expert testimony nor the jury charge provided a basis for an
inference that the jury had been left with the misapprehension that the
statistics predicted the likelihood of guilt,² they said, asking the court
to find that the judge was "correct in not depriving the jury of this
relevant, highly probative evidence."
The Crown lawyers said recent studies "bear out the assertion that juries
are not overwhelmed by statistical expressions of match significance in
relation to blood profiling."
They cited a U.S. National Research Council paper as concluding that
empirical research "does not support the common assertion that statistical
evidence is overvalued. To the contrary, several studies with mock juries
suggest that decision-makers generally make smaller adjustments in their
judgments in response to probability evidence than the statistical evidence
warrants."
The lawyers also warned that if testimony was limited to the use of
non-statistical terms such as "rare" or "common," the evidence would become
too ambiguous. "It is submitted that numbers provide a 'language' in which
the significance of the particular matches in an individual case can be
expressed."
The court has reserved judgment in the appeal.

[Lo52964]

>how...@the-wire.com (Howard ) writes

Made Simple

Anglo Saxon Courts don't like stats. "We are trying this case not
everybody else's." All that would be needed here is for the expert to
testify." to a reasonable scientific certanty." that theDNA came from the
defendant and leave the numbers out.

There also appears to be what in the US would be a "FRYE RULE" question.
Was the evidence generally accepted science at the time?

LO5 2964

[GLC1173]

Howard quoted:

>> The Ontario Court of Appeal has been asked to prohibit DNA
>>experts from expressing their findings in terms of >>statistical
>>probabilities at criminal jury trials.

DNA evidence is a good example of where and why the jury system is outmoded
- many jurors simply do not have the coursework to understand the concepts
involved in any trial involving complicated engineering-type issues, scientific
evidence, or medical malpractice.
Juries - at least in America - tend to be made up of people who are
disproportionately unlikely to have a college degree in anything, much less
math, chemistry, physics, or engineering. Few jurors are doctors. Few are
CPAs. This leaves a bunch of people left to be "impressed" by a magic show of
expert witnesses - as happened in both the O.J. trial and in the Hinckley
trial. (How is a juror listening to several shrinks tell him Hinckley is
insane and an equal number telling him Hinckley is sane to decide which to
believe - when that juror has no coursework in the field?)

[Lo52964]

glc...@aol.com (GLC1173) writes

SNIP

> This leaves a bunch of people left to be "impressed" by a magic show of
>expert witnesses - as happened in both the O.J. trial and in the Hinckley
>trial. (How is a juror listening to several shrinks tell him Hinckley is
>insane and an equal number telling him Hinckley is sane to decide which to
>believe - when that juror has no coursework in the field?)

In Hinkley it it called,'The policeman at the elbow test." If you would have
done it with a policeman etc then you are insane. Hinkley did it in front of
the Secret Service who were sure to kill or capture him.

You don't have to be a fud to figure this out.

In OJ, if the expert says this DNA is x's. OK. Other expert says this in
not proper procedure, you have reasonable doubt, if you want it.

Even with a jury of microbiologists, you still have reasonable doubt unless
they run the tests themselves, which juries don't do. Juries do their
wierdest in civil cases anyway.

LO5 2964

[GLC1173]

LO5 2964 quoted me:

> This leaves a bunch of people left to be "impressed" by a >magic show of
>expert witnesses - as happened in both the O.J. trial and in >the Hinckley
>trial. (How is a juror listening to several shrinks tell him >Hinckley is
>insane and an equal number telling him Hinckley is sane to >decide which to
>believe - when that juror has no coursework in the field?)

and wrote:
>> In Hinkley it it called,'The policeman at the elbow test." >>If you would
have
>>done it with a policeman etc then you are insane. Hinkley >>did it in front
of
>>the Secret Service who were sure to kill or capture him.

Even all the expert-witness shrinks in his trial did not agree that Hinckley
was - or was not - insane. If those highly-educated shrinks could not find one
true conclusion on the issue of Hinckley's sanity, can a jury composed largely
of those who never went beyond high school do so?
You view insanity-defense law from a very simplistic perspective that does
not resemble what happens in trials. Squeaky Fromme and Sara Jane Moore - both
disciples of Charles Manson - also tried to murder a president at close range
with the Secret Service present making for a high risk of their being killed -
and both Fromme and Moore were found guilty and are now in federal prison.
"Sure to get captured" does not add up to proof for a successful insanity
defense either - just being unskilled at being a criminal does not get you off
as insane.