Probability & Confidence Documentation

Probability and confidence documentation records how certain an authentication conclusion is, why it sits at that level, and what might change it. It is the difference between a collection record that says “genuine” and one that says “probably period, strongly consistent in materials and construction, but maker attribution not confirmed.”

This matters because confidence often hardens in storage. A careful note written after inspection can become a definite claim in a report, then a simplified label in a handover pack, then an inherited truth for the next owner. Good documentation keeps uncertainty visible without making the record feel weak. It shows the collector has thought carefully enough to know where confidence ends.

The confident note with no ladder beneath it

Imagine two collectors reach the same view that an object is probably authentic. The first records only “authentic, checked against examples.” The second records: “High confidence on period and production method; moderate confidence on originality of one component; low confidence on named attribution. Confidence would increase if a documented reference example with the same component pattern is found.”

Both collectors may be equally capable, but only the second record can be reviewed. It shows which parts of the conclusion are strong, which are provisional and what future evidence would matter. That is the quiet power of confidence documentation: it makes judgement reusable.

Understanding the topic

Confidence is not a decoration added after the conclusion

Confidence is part of the conclusion itself. “Authentic” and “probably authentic” are not the same record. “Consistent with the period” and “confirmed as the period” are not the same claim. A collector who records only the conclusion removes the scale that tells the next reader how much weight the conclusion can safely carry.

Good documentation treats confidence as evidence-shaped. It asks whether the conclusion rests on one strong clue, several independent clues, a reliable expert opinion, comparison with secure examples, scientific testing, provenance, or only a plausible surface impression. The level of confidence should rise because evidence converges, not because the claim is desirable.

Different parts of the same object may deserve different confidence levels

Authentication conclusions are often layered. A collector may be highly confident that the material is period-correct, moderately confident that the object is as-issued, uncertain about one replaced component, and unable to support a named-maker attribution. A single confidence label can hide that structure.

This is why useful records avoid one large confidence statement when the evidence is uneven. They document confidence by claim: identity, date, materials, maker, edition, originality, completeness, repairs, provenance and attribution where relevant. The reader can then understand exactly what is strong and what remains open.

The hidden question is what would change your mind

A confidence note becomes much stronger when it states what would move the conclusion. Would a better reference example increase confidence? Would inconsistent measurements lower it? Would an expert opinion resolve attribution but leave originality unanswered? Would imaging of a hidden area matter?

Experienced collectors often think this way even when they do not write it down. They know that confidence is not a mood. It is a current position in a continuing enquiry. Recording the evidence that would change the position protects the record from becoming frozen certainty.

Why it matters

Probability and confidence documentation matters because authentication records are used for different decisions. A private research note can tolerate more uncertainty than a sale description. An insurance schedule may need clear disclosure of uncertainty that would be harmless in a personal checklist. A handover pack may need enough detail for a future owner to avoid overstating the collection.

It also protects against the illusion of precision. Collectors sometimes replace uncertainty with confident wording because uncertain wording feels less professional. In reality, a carefully explained probable conclusion is often more trustworthy than a definite conclusion with no visible reasoning behind it.

Most importantly, confidence documentation teaches future readers how to think with the evidence. It does not merely preserve what the collector believed. It preserves why the belief was proportionate at the time.

Practical guidance

Record confidence beside the specific claim it belongs to

Do not let confidence float separately from the claim. A note saying “confidence: high” is not enough if the object contains several authentication questions. High confidence in date may not mean high confidence in originality. High confidence in maker may not mean high confidence in untouched condition.

A practical record names the claim, gives the confidence level, and states the reason. This turns uncertainty from a vague feeling into a reviewable judgement.

Name the exact claim: date, maker, edition, material, originality, completeness, provenance or attribution.
State the current confidence level in plain language rather than hiding behind unexplained scores.
Record the evidence that supports that level of confidence.
Identify the main reason confidence is not higher, if there is one.
Note what future evidence would raise, lower or resolve the conclusion.

Use wording that can survive being copied

Authentication wording often travels into summaries, reports, labels and sale notes. If the confidence is buried in a private paragraph, the copied version may become too strong. The safest approach is to keep confidence wording close to the claim wherever the claim appears.

For example, “probably original 1970s issue; confidence moderate-high because printing, paper and measurements align with secure examples; ownership history unverified” is safer than “original 1970s issue” with uncertainty stored elsewhere. The first wording resists drift because the confidence and limit are built into the sentence.

Date confidence changes and explain why they changed

Confidence should be allowed to move. New reference examples, expert opinions, scientific tests, better photographs, market discoveries or known counterfeits may all change a conclusion. The record should show the date and reason for the change rather than silently overwriting the old view.

A dated revision note is valuable even when confidence decreases. It shows intellectual honesty and helps a future reader understand why older descriptions may differ. Authentication records are stronger when they show careful revision rather than pretending the final wording was always obvious.

Common mistakes and risks

Using certainty words because they sound cleaner

Words such as genuine, confirmed, original and authentic can make a record look tidy, but they may overstate what has actually been shown. If the evidence supports probability, the record should say probability. Clean wording is not worth more than accurate wording.

Letting one strong clue raise confidence across every claim

A strong mark, certificate, material result or provenance note may raise confidence in one part of the authentication question. It should not automatically raise confidence in everything else. The record should show exactly which claim the strong clue supports.

Failing to record why confidence is limited

Collectors often write the evidence that supports a claim but omit the reason they stopped short of certainty. That missing explanation is important. It tells future readers whether the limit came from lack of evidence, conflicting evidence, limited access, unresolved variation or a deliberate decision not to overstate.

Advanced considerations

When confidence wording becomes part of disclosure

Confidence language becomes especially important when the record leaves private use. In valuations, insurance schedules, exhibition material, sale descriptions and handover packs, a confidence statement is not just internal thinking. It shapes how another person relies on the object.

Where value or risk depends on a conclusion, the record should identify whether confidence is strong, moderate, provisional, disputed or dependent on a particular source. This does not mean weakening every claim. It means making the strength of each claim visible enough that another person does not rely on it beyond the evidence.

Key takeaways

Document confidence beside the exact authentication claim it belongs to.
Different claims about the same object may deserve different confidence levels.
A probable conclusion with clear reasons is stronger than a definite conclusion with no reasoning trail.
Record what would raise, lower or resolve confidence in the future.
Date confidence changes so the record remains honest, reviewable and useful.

Continue learning

Attribution Documentation

Review how to document attribution levels, wording strength and the origin of named claims.

Back to Authentication Documentation

Return to the Authentication Documentation sub-domain and its full topic list.

Burden of Proof Documentation

Continue to recording what evidence is needed for claims that carry value, sale or insurance risk.