By Zoë Hall

Published

The Evidence Does Not Speak

A fingerprint at a crime scene does not explain when it was left there or why it was left there. A DNA sample may yield a profile and help identify whose biological material is present without telling us how or when it arrived there. A fibre cannot tell a jury whether it was transferred during a crime, through innocent contact, or by subsequent contamination. Even a laboratory result still has to be interpreted before it can answer any of the questions being asked of it.

Episode I of SYSTEMIC, The Erosion Code, looked at the legal architecture itself. It asked what the state is allowed to do, how rights can be restricted, and how those powers become embedded into law. Now we examine what has to happen before those powers can be used against someone. First, the system has to decide what actually happened, who did it and what the evidence proves.

That is where Episode II begins.

Forensic evidence carries an extraordinary amount of weight. It often comes wrapped in statistics, probabilities, technical language, laboratory reports and expert testimony. That can make it sound far more authoritative than it really is, which is why the interpretation matters. Get that wrong and the consequences are very real for everyone involved.

Before evidence reaches a courtroom, it passes through a chain of human decisions that can, on occasion, introduce problems, risks and miscalculations. Someone has to collect it, someone else may preserve it, someone makes a decision about what to test, and an analyst then interprets the results. Police place those results inside an investigative theory. The Crown Prosecution Service (CPS) has to evaluate the evidence and decide how, if at all, it fits the prosecution case. Experts explain it, lawyers frame it within the competing cases being put before the court, and finally a jury has to decide whether the prosecution has proved guilt beyond reasonable doubt.

At any one of those stages, uncertainty can be preserved, tested, narrowed or lost. The possible can become the probable. Consistent with can begin to sound like identified as. A qualification can disappear. An interpretation can begin to sound like a measurement. A finding that supports a proposition can start to be treated as though it establishes it.

The problem is not that certainty necessarily increases at every handoff. It is that uncertainty can be lost, obscured or overstated somewhere between the evidence and the legal conclusion, even inside a process that contains safeguards specifically designed to test it, expose it and make its limits clear.

That is what I mean by forensic fiction. Uncertainty presented with the authority of fact.

And there is a reason I want to examine that system now. Before any irreversible punishment can be justified, we need to understand how the system reaches the certainty on which that punishment would depend, and what happens when later evidence shows that certainty was misplaced.

TL;DR

  • Evidence and interpretation are different things. A forensic observation can be accurate while the meaning attached to it remains uncertain, conditional or open to challenge.
  • Not every forensic failure is a forensic fiction. Laboratory-integrity failures, non-disclosure, interpretive overstatement and weaknesses in correction machinery are different problems and should be examined separately.
  • The safeguards matter. Modern forensic practice includes regulatory standards, disclosure duties, expert-evidence requirements, defence challenge, peer review, judicial scrutiny, appeal and post-conviction review. The question is where uncertainty can still survive despite them.
  • The vulnerability is still current. Recent Forensic Science Regulator material identifies live concerns around interpretation, validation, measurement uncertainty, checking and peer review, and current regulation still has to hold the line between detecting forensic material and deciding what its presence means.
  • Historical miscarriages show failures, not prevalence. Clark and George help expose how statistical reasoning, missing material and interpretive significance can distort the case. They do not establish how often those failures occur under the system operating today.
  • Correction is a system of its own. Sullivan and Randox show that finding a problem is only the beginning. Correction depends on surviving evidence, expertise, funding, review capacity and a legal route capable of reopening the question.

Verify to Unlock

This episode is free, but it requires a verified account.

Create Free Account

After verifying your email, this episode will unlock automatically.

The conversation is not behind the paywall

Read the public Community discussion around this episode for free. No account required.

Read the Community Discussion

Posting, voting and other participation still require a free account.


Prefer the full text now? You can buy this part and keep a permanent offline edition.

Download the PDF - £2.99

Members and Founding Members pay £2.49, applied automatically at checkout when signed in.