The debate rages over the reliability of forensics

Forensics is far from a settled discipline. Controversy has raged over the reliability of fingerprints and even DNA.

When it comes to blood pattern analysis, a case in Indiana is stirring strong debate.

David Camm was a former state trooper accused of fatally shooting his wife and three children in their garage in 2000.

Camm claimed that he discovered the bodies after the murders and that tiny blood stains on his shirt were from brushing against his daughter.

Prosecutors, however, said the blood marks were spattered from the impact of the gunshot.

Ten forensic witnesses testified on that point -- five for the state, five for the defense. They were in complete disagreement over whether the blood was an impact pattern from a gunshot, or a transfer pattern from contact.

Despite such discrepancies, Camm has twice been convicted.

At a national conference last year, blood pattern analysts discussed issues the case raised about witness qualifications and errors.

As shows like "CSI" have made crime investigation popular, more people have entered the field, with glaring differences in expertise. Some are working after just a one-week course in a specialty like blood pattern analysis. Others have Ph.D.s in the sciences and a lifetime of experience.

As a consequence, bloodstain analysts are debating what qualifications establish someone as an expert in the field.

There are two schools of thought: that forensic testimony should be restricted to scientists who work in labs, or that police officers who know the way around a crime scene should be considered experts.

Currently, anyone with a 40-hour training course in blood pattern analysis can be established as an expert to give court testimony. By contrast, in Canada, analysts are required to undergo advanced training and become an understudy for a year before qualifying as an expert.

Joe Slemko, a constable and prominent bloodstain analyst from Edmonton, says it takes years of experience in the field to compare textbook blood patterns with actual crime scenes.

At the same time, an expert should be able to replicate in the lab any theory of how a bloodstain was produced in the real world.

"Blood pattern analysis can be a very dangerous science in the hands of the inexperienced," he said.

As an example, he cited the case of Sion Jenkins in England, who served seven years in prison for a crime that Slemko testified he could not have committed.

The prosecution witness, with limited training, testified that blood on Jenkins was impact spatter and meant Jenkins was at the murder scene, but Slemko concluded it was expirated, or breathed, blood from Jenkins attempting to help the victim. The defendant was acquitted and freed after serving seven years on his original conviction.

To remedy such discrepancies, Jeff Gurvis, a former crime scene coordinator for what's now the Northeastern Illinois Regional Crime Lab, is part of a group of forensic scientists working for the FBI to develop minimum standards for education, training and working procedures for blood pattern analysis.

One area for potential research involves testing how blood reacts when spilled from real human bodies, or from live (unharmed) participants touching, wiping or rolling in blood.

Gurvis, who like most analysts is not in the least grossed out by blood, looks forward to such work.

"It would be fascinating," he says. "That's one thing this field definitely needs -- a lot of research."