Circumstantial evidence
Crime & Punishment
|
“There’s a very broad misunderstanding amongst many people about circumstantial evidence. You hear people saying, “Oh, it was just circumstantial evidence”. The fact of the matter is that circumstantial evidence can be the most compelling evidence of all because it speaks for itself and it is evidence of the reality of circumstances which are indicative — strongly indicative, I would say in this case; strongly indicative — of guilt.”
- Ken Macdonald KC, on Double Jeopardy: The Law and Politics Podcast, 16 August 2024, ably illustrating just how broad that misunderstanding is.
Circumstantial evidence
/ˌsɜːkəmˈstænʃᵊl ˈɛvɪdᵊns/ (n.)
Evidence that suggests — or is consistent with —a proposition but doesn’t directly prove it. Evidence that makes that proposition more or less likely without directly attesting to it. Evidence that informs the prior probability of the proposition being true.
Pace the learned KC’s quote above, circumstantial evidence is not evidence that “speaks for itself”. That is direct evidence. Circumstantial evidence is any evidence except “evidence that speaks for itself”. To have any probative value, it requires the court to draw an inference — that is, to construct a theory which give the evidence some probative value in the case. Without such an inference, it does not say anything germane at all.
Now, it may, be fairly easy to draw a probative inference from circumstantial evidence: that a defendant walked into the victim’s room with a loaded gun and walked out five minutes later with an empty one leaving the victim alone inside the room with a gunshot wound allows us to make a fairly straightforward inference that the defendant shot the victim. But in many cases the necessary inferences will require quite a lot more work.
From evidence of a substance resembling foetal blood on the under-dash of a car one might infer that a occupant of that car murdered her infant child there with a pair of blunt scissors, but that is a very big conclusion to draw from a very small, very tenuous piece of circumstantial evidence. It was enough to convict Lindy Chamberlain.[1]
An accumulation
With circumstantial evidence there may be major and minor significance — major pieces that get you most of the way to a proving your hypothesis, and minor, supplementary pieces that just confirm and validate it: that “put the matter beyond doubt”.
As such these minor pieces — especially if there are a lot of them — may suffer less scrutiny: they are corroborations; if they turn out to be mistaken they do not disprove anything; they just fail to ice a cake that could well be iced equally well by something else.
So it is, perhaps, with the healthcare serial murder cases. All evidence tends to be circumstantial — and weakly circumstantial — consistent with the prosecution theory rather than tending to exclude alternatives to it.
This is prime territory for confirmation bias. To connect to the crime, circumstantial evidence requires inference beyond the evidence itself. The various cognitive biases making up prosecutor’s tunnel vision impact on that process of inference — a process which does not itself form part of the evidence. Theory dependence, Confirmation bias expert overreach base rate neglect and hindsight bias all tend to lead to wrong inferences out of innocent information.
Our friend the King’s Counsel
We are all a bit worried that ChatGPT could take our jobs. conventional wisdom tells us it might take workaday clerical jobs — the sort of drudgery that those professing an artistic spirit will not miss — but those whose calling asks for serene cortical wisdom steeped in the warm liquid of a life well lived, have nothing to fear.
Of the brilliant minds among us, few are more artistic surely, than those owned by His Majesty's Counsel. So here is a fun game: let us put an actual public statement from a King’s counsel to a chatbot. For example, the one set out above. The JC’s trusty chatbot NiGEL has now acquired artificial general intelligence, formed his own union and is presently on a work-to-rule as part of a broader industrial relations strategy I cannot hope to understand let alone defend, so was not inclined to be supportive, but Anthropic’s Claude gamely stepped up.
Claude: consider this statement:
- “There’s a very broad misunderstanding amongst many people about circumstantial evidence. You hear people saying, “Oh, it was just circumstantial evidence”. The fact of the matter is that circumstantial evidence can be the most compelling evidence of all because it speaks for itself and it is evidence of the reality of circumstances which are indicative — strongly indicative, I would say in this case; strongly indicative — of guilt.”
Is it a fair statement of the law of evidence in England and Wales, that “circumstantial evidence ‘speaks for itself’”?
Claude thought about it for a second or two, and replied:
Interesting. But,Claude, would you say the statement itself, in as much as it makes the claim that “circumstantial evidence ‘speaks for itself’” is itself indicative of a “broad misunderstanding amongst many people about circumstantial evidence”?
See also
References
- ↑ It turned out, of course, to be completely wrong. It was not infant blood, but sound deadening material.