Evidence is not the same thing as being right

A decision is made about the material, not the events

The hardest idea for anyone in their first dispute is that the process does not determine what happened. It determines what has been proved to the required standard using the material put before it. Those usually coincide, which is the point of the exercise, but not always, and the gap between them is where most of the bitterness lives.

This is not a defect that could be engineered away. A decision-maker was not present at the events, has no independent means of investigating them, and must reach a conclusion within a finite time using what the parties supply. Any system with those constraints will sometimes get the answer wrong, and every system knows it.

Who has to prove what, and how convincingly

Two separate questions run underneath every dispute. The first is which party carries the burden of proving a particular point, and the general position in most systems is that whoever asserts something must establish it. The second is the standard: how convinced the decision-maker must be before treating it as established.

Civil standards are typically lower than criminal ones, commonly expressed as something close to more likely than not rather than beyond reasonable doubt. That difference has a consequence people find counter-intuitive: the same events can produce a finding of liability in a civil claim and no conviction in a criminal one, without either outcome being inconsistent with the other.

Documents made at the time outrank everything

A contemporaneous document is the strongest ordinary evidence there is, and the reason is structural. It was created before anyone had a stake in the answer, so it has not been shaped by the dispute. An email confirming a conversation, a photograph with a date, a delivery note, a diary entry — all of these carry weight out of proportion to how important they seemed when made.

Recollection is treated far more cautiously, and modern practice in several jurisdictions has become explicitly sceptical of memory in commercial disputes. This is not an accusation of dishonesty. Memory reconstructs rather than replays, it is influenced by what a person has read and discussed since, and confidence turns out to be a poor guide to accuracy. Witnesses can be entirely honest and entirely mistaken.

Metadata has quietly become part of this. Timestamps on messages, dates attached to photographs and the edit history of a document can establish when something was created, which is very often the contested point. It cuts both ways, since the same information may show that a document relied on was produced later than claimed. Parties are frequently surprised by how much of it survives after a file has been copied and forwarded several times, and by how much is lost when a screenshot replaces the original.

Handing over what damages you

Most systems require parties to disclose relevant documents, and in many that obligation extends to material that harms your own case. People find this astonishing when they first meet it. The rationale is that a process where each side reveals only what suits them produces decisions based on a curated fiction.

The practical consequences are worth understanding before a dispute starts. Deleting inconvenient material once a dispute is in prospect is treated very seriously and can be worse for a party than the document ever was. Messages, drafts and internal notes are frequently disclosable. And the volume of material is one of the main drivers of cost in larger disputes, which is a reason to keep records tidy rather than voluminous.

Opinion, expertise and the limits of both

Ordinary witnesses generally give evidence about facts rather than conclusions. Where a question requires specialist knowledge — the cause of a structural defect, the standard of professional work, the value of an asset — most systems allow expert evidence, usually with permission and usually subject to conditions.

The condition that matters most is that an expert’s duty is typically owed to the court or tribunal rather than to whoever is paying. An expert who argues a case rather than answering a question tends to be discounted, sometimes heavily. That is why an unfavourable expert opinion obtained early is genuinely useful information, even though it is the last thing anyone wants to hear.

What this means before anything has gone wrong

The unglamorous conclusion is that the outcome of a dispute is substantially determined long before it begins, by whether anybody wrote things down. Confirming conversations in writing, keeping receipts and photographs, and recording problems as they occur costs almost nothing and is the single most effective thing available to a non-lawyer.

Rules of evidence, disclosure obligations, the treatment of experts and the applicable standards all vary between jurisdictions, and this article states none of them as universal. If you are in a dispute that matters, take advice from a qualified lawyer where you live — early, because preserving evidence and meeting a limitation period are both things that can only be done in time.