Authentication through comparison is the disciplined examination of an uncertain collectible against securely identified examples, documented production norms and credible historical records. Its purpose is not to find a convenient resemblance. It is to test whether the complete pattern of materials, manufacture, construction, markings, ageing and documented history makes better sense under the claimed identity than under the plausible alternatives.
Comparison is among the most powerful methods available to collectors, but it is also easily misused. A counterfeit is expected to resemble a genuine object. A legitimate regional issue may differ from the example in a collector's cabinet. A genuine object may contain replacement parts, restoration or later additions. The method therefore begins with a precise claim, evaluates the quality of the references, controls the variables being compared and records both agreement and contradiction before any conclusion is stated.
Governing principle
Similarity establishes compatibility, not identity. The unknown object must be tested against the reference evidence; the references must not be selected to defend the object.
Orientation
The vague question 'Is it real?' hides several different propositions. A comparison becomes useful only when the collector states which identity, date, issue, component and degree of originality are actually being tested.
Two objects may resemble one another because both are genuine, because one is a legitimate later issue, because a counterfeit copied a genuine example, because both came from the same counterfeit source or because they share a broad period style. The more visible a feature is, the more likely a competent reproducer can copy it. Strong comparison therefore moves beyond the intended appearance and looks for traces produced by manufacture: plate wear, mould seams, tooling, assembly sequence, fibre structure, hidden layers and interactions between materials.
“Does this look genuine?”
“What should be present, absent or variable if this exact claim is true?”
Method
A strong assessment usually combines more than one form. Surviving objects show what was made, production evidence explains how it was made, and contemporary records establish what was intended, distributed or documented at the time.
Object to object
Direct comparison can reveal dimensional agreement, shared tooling or plate characteristics, legitimate variation, replacement components and anomalous construction. It is strongest when the reference object has secure provenance, early documentation and little undisclosed restoration.
Another market example is not automatically a known genuine example.
Object to production model
The examiner asks which materials, technologies, factories, subcontractors, tolerances, assembly methods and recurring defects belong to the claimed production history. This often exposes objects that reproduce the correct design but not the correct process.
Process evidence is frequently harder to imitate than visible appearance.
Object to record
Catalogues, factory drawings, advertisements, invoices, dated photographs, packaging illustrations, accession records and production correspondence may establish that a version existed, identify when a feature changed or show that a supposed issue date is impossible.
Records should be assessed for date, authorship, purpose and relevance to the exact claim.
Reference discipline
A folder of similar photographs is not automatically a reference set. The collector must know why each example is trusted, what exact identity it represents and which limitations travel with it.
The circular-reference problem begins when Object A is accepted because it resembles Object B, Object B was accepted because it resembles Object C, and Object C has no secure foundation. Increasing the number of mutually similar examples does not repair the chain. They may all derive from one reproduction run, one incorrect catalogue attribution or one collector convention repeated until it feels established.
Tier 1
Factory-retained examples, archival production samples, objects acquired directly from the maker or original recipient, original dies or plates, and museum-accessioned objects supported by detailed records.
Tier 2
Objects with credible dated receipts, examples illustrated in contemporary catalogues, long-held collections formed before counterfeiting became attractive, and objects examined and published by recognised specialists.
Tier 3
Examples widely accepted by experienced specialists or recurring in respected catalogues, but without the complete documentary foundation of primary references.
Tier 4
Dealer listings, auction photographs, forum posts, social-media images and unattributed collection photographs. These can reveal patterns and lead to better sources, but should not silently become verified ground truth.
Reference warning
Digital images can conceal scale, true colour, gloss, depth, edge structure, watermark appearance, material layering, restoration and compression artefacts. Remote comparison is useful for triage, obvious inconsistencies and deciding where to examine next. It usually cannot establish material composition, weight, internal construction, microscopic print process, fluorescence behaviour or concealed restoration.
A remote opinion should therefore say “apparently consistent”, “visibly inconsistent” or “requires physical examination” rather than presenting image review as equivalent to handling the object.
Control the variables
The correct question is not whether the questioned object differs from a collector's example. It is whether an object with the exact identity claimed should be expected to match that particular comparator.
Two apparently identical game boxes may differ because they were printed in different countries, assembled by different subcontractors, produced in separate years, sold through different channels or made during a material shortage. A mismatch against the wrong production state is not evidence of forgery. Equally, an exact visual match may be weak evidence when a reproduction was copied from a genuine object.
Examination
Separating observations into axes prevents one persuasive feature from dominating the assessment. Each axis asks a different question and carries different limitations.
Examine
Meaning
Recorded measurements can support or challenge a claimed issue when compared with an appropriate expected range.
Collector caution
Allow for production tolerances, mould shrinkage, paper movement, trimming, wear, repair and uncertainty in the measurement method.
Examine
Meaning
Material evidence may be consistent, inconsistent, inconclusive, apparently original, apparently later or mixed.
Collector caution
Old materials can be reused, while genuine objects may contain later conservation, repair or replacement materials.
Examine
Meaning
Manufacturing evidence tests whether the object could plausibly have emerged from the historical process claimed for it.
Collector caution
The visible design may be copied accurately even when the historical production method is wrong.
Examine
Meaning
Magnification and controlled illumination can distinguish manufacture, later treatment and different material histories.
Collector caution
Micro-features require suitable reference evidence and experience; unfamiliarity is not itself proof of forgery.
Examine
Meaning
Copied artwork may preserve the design while adding a second generation of reproduction, softened detail, moiré or a new print structure.
Collector caution
Indicators vary by process. Generic statements such as 'the dots look wrong' need technical explanation.
Examine
Meaning
Colour can help identify production differences, ageing pathways or restoration when compared under controlled conditions.
Collector caution
Lighting, cameras, white balance, screens and image processing can create differences that are not present in the objects.
Examine
Meaning
Reverse, internal and hidden construction may be more discriminating than the display face because it follows the production process.
Collector caution
Opening or dismantling an object can cause damage. Non-invasive examination comes first.
Examine
Meaning
The marking, the object and the moment of application must be assessed separately before they are combined into one claim.
Collector caution
A valid serial number can be copied, a genuine stamp applied later and an authentic signature transplanted onto an inauthentic object.
Examine
Meaning
Plausible ageing follows the object's use, construction, contact points and environmental exposure rather than appearing as a uniform decorative effect.
Collector caution
Condition is highly variable. A well-preserved genuine object may look newer than a badly stored reproduction.
Weight of evidence
Matching characteristics do not all carry equal value. The method must explain not only that a feature matches, but why that match separates the claimed identity from its alternatives.
Low discrimination
Overall design, broad colour, basic dimensions, expected wording, visible logos and general style may establish compatibility but often occur on genuine issues, later reissues and counterfeits alike.
Medium discrimination
Exact typography, print registration, material thickness, assembly sequence, minor placement, recurring production flaws and characteristic finishing narrow the possibilities more effectively.
High discrimination
Specific die fractures, progressive plate wear, repeatable mould damage, distinctive layer construction, microscopic tool relationships and physically matching separated components can carry substantial weight when the references are secure.
Class characteristics
Dimensions, material type, print process, typeface, standard mould design, expected colour range and normal component type may show that an object could belong to a category.
More source-specific characteristics
Particular plate wear, an accidental engraving defect, a damaged mould, recurring misalignment, batch-specific material combinations or specific tool marks may support a more restricted conclusion.
Interpretation
Recording every similarity as positive evidence creates false confidence. Findings should be separated according to whether they support, contradict, fail to distinguish or cannot reliably be assessed.
Positive
The correct period printing process is observed. This supports the claim, although its strength depends on whether the process is exclusive to the claimed issue.
Negative
The paper contains a feature introduced after the claimed production date, or a construction joint inherent to the supposed process is absent.
Neutral
Dimensions fall within a broad range shared by genuine examples and known reproductions. The observation belongs in the record but adds little weight.
Indeterminate
The relevant area is obscured, the image lacks resolution, the comparator is unsuitable or the expected range has not been established.
A missing feature matters only if it genuinely should have been present and observable. Required legal wording, an edition-specific page or a construction joint inherent to the production method may be obligatory. A feature present on most examples may be normally expected but subject to exceptions. Other features are permitted variation or merely incidental.
Before treating absence as evidence, ask whether the entire area was examined, whether wear or restoration could conceal it, whether it belongs only to some production runs and whether the available image is adequate. Record “not present”, “not visible”, “not examined”, “obscured”, “removed” and “indeterminate” as different findings.
Judgement
An anomaly may indicate a counterfeit, but it may also result from transitional production, subcontracting, factory rework, supply shortage, repair, storage damage, owner modification or a mistaken reference classification.
Repeat the observation and exclude glare, image processing, scale error, measurement error or damage that merely changes the apparent feature.
Ask whether the reference is the same issue, factory, period and state. A mismatch against the wrong comparator has little authentication value.
Determine whether the feature is isolated, recurring, batch-specific, transitional or already associated with a known reproduction or counterfeit source.
Consider subcontracting, supply shortage, factory rework, repair, restoration, owner modification, mixed components and storage damage.
An undocumented feature may be rare without being historically impossible. A feature that requires a material or technology unavailable at the claimed date carries different weight.
Collector rule
Rarity does not explain an inconsistency. The surrounding production, documentary and material evidence must explain it.
Bias control
The seller's story, rarity, price and desired conclusion can shape what the collector notices. A mature method introduces deliberate friction before the conclusion becomes emotionally owned.
Diagnostic question
If the honest answer is “nothing”, the process is no longer authentication. It has become advocacy for a conclusion already chosen.
Counterfeit-aware comparison includes confirmed genuine examples, known later issues, documented reproductions, known counterfeits, altered genuine objects and uncertain examples kept outside the authenticated set. Comparing only with genuine objects shows what genuine objects may look like; it does not reveal which features false objects routinely reproduce or which counterfeit signatures recur.
Repeatability
A counterfeit may pass at the level of overall shape and design while failing in edge finishing, print structure, fibre, tooling, underlayers or concealed assembly.
Macro
Shape, layout, proportion, broad colour and major components visible without magnification.
Meso
Typography, seams, folds, fasteners, edge finishing, print alignment and component relationships.
Micro
Fibres, pigment particles, print dots, tool marks, corrosion layers, coatings and microscopic wear.
Internal
Underlayers, concealed repairs, support materials, assembly sequence and internal construction.
Standardised visible-light photography, raking light, transmitted light, macro photography, microscopy, ultraviolet or infrared imaging and three-dimensional surface capture may reveal different classes of evidence. The purpose is not to make attractive pictures; it is to preserve comparable observations.
Practical method
The workflow is deliberately sequential. Each stage reduces the risk that the collector will select evidence around a conclusion formed too early.
Write the identity precisely. Avoid undefined terms such as 'original'.
Document the object before the comparator dominates perception.
Record the identity, source, provenance strength, reason for inclusion, known restoration and limitations of every comparator.
Do not inspect every possible feature indiscriminately. Select the dimensions, materials, processes, construction, markings, ageing and records that can distinguish the alternatives.
Document recurring characteristics, obligatory features, permitted variation, production tolerances, uncertain features and known counterfeit indicators.
For each feature record the observation, reference expectation, match or difference, possible explanations and evidential significance.
Evaluate the claimed issue against legitimate variation, later reissue, altered genuine object, composite object, reproduction, counterfeit and insufficient evidence.
Look for convergence across independent visual, technical, documentary, provenance and scientific lines. Do not reduce the result to a vote count.
Identify unavailable references, inaccessible areas, lack of sampling, unreliable colour, uncertain provenance and incomplete production history.
State exactly what the evidence supports, what remains unresolved and what next action would materially improve the judgement.
Working record
A compact matrix keeps observation, expectation and interpretation separate. It also makes contradictions and low-value similarities visible rather than allowing them to disappear inside a narrative conclusion.
| Comparison area | Observation | Reference expectation | Interpretation |
|---|---|---|---|
| Dimensions | 210 x 148 mm | 209-211 x 147-149 mm | Consistent |
| Printing | Modern stochastic pattern | Period examples use letterpress | Strong contradiction |
| Paper | Fibre appears compatible | Compatible documented stock | Supportive, not exclusive |
| Marking | Correct wording | Also copied by reproductions | Low evidential value |
| Edge | Uniform laser-like darkening | Genuine examples mechanically cut | Raises concern |
| Provenance | Begins in 2024 | No earlier documentation | Neutral to weak |
| Wear | Abrasion conflicts with protected assembly | Wear should postdate assembly | Possible artificial ageing |
The matrix is a reasoning record, not a scorecard. One historically impossible feature may outweigh several generic similarities.
Failure patterns
These statements sound persuasive because each contains a fragment of potentially useful evidence. The failure lies in allowing that fragment to authenticate more than it can support.
Reference failure
The published object may be misidentified, restored, from a different issue, reproduced at the wrong scale or itself unauthenticated. Publication is not a substitute for reference criticism.
Circular confidence
Many apparently independent listings may trace back to one copied description, one catalogue error or one counterfeit source. Repetition can amplify a weak foundation rather than strengthen it.
Rarity story
Anomalies require explanation. Transitional production or an undocumented variant is possible, but rarity is not the default explanation for inconsistency.
Age substitution
Old paper, metal, components and packaging can be reused. A material date does not automatically establish when, where or why the complete object was assembled.
Testing overreach
A test may show that one component is compatible with a period or that no tested substance is demonstrably modern. It may not establish the identity, assembly history or originality of the complete object.
Expert overreach
Pattern recognition can be valuable, but a responsible opinion should still be explainable through observable features, suitable references, scope and limitations.
Proportionate action
Not every object requires laboratory analysis, and not every high-stakes conclusion should rest on casual collector comparison. The method should expand with the value, uncertainty, consequences and technical difficulty of the claim.
Appropriate for: Inexpensive objects, obvious issue distinctions, cataloguing and initial screening.
Proportionate response: Use documented references, record key observations and retain clear photographs.
Appropriate for: Several legitimate variants, known counterfeits, meaningful value or incomplete provenance.
Proportionate response: Use multiple qualified references, a feature matrix and explicit alternative explanations.
Appropriate for: Material or process is decisive, restoration may be concealed or visible evidence conflicts.
Proportionate response: Consult a conservator or technical specialist and define the examination question before treatment or sampling.
Appropriate for: Composition, dating or microscopic source features matter and the financial or legal consequences are high.
Proportionate response: Use a qualified laboratory or forensic practitioner. Testing must answer a defined proposition rather than serve as a vague search for certainty.
Appropriate for: Sale, insurance, litigation, institutional acquisition, publication or high-value attribution.
Proportionate response: Require clear authorship, scope, evidence, methods, limitations, conclusion language and an auditable record.
Specialist threshold
Seek specialist assistance when financial stakes are significant, destructive sampling may be required, specialised imaging is needed, a signature is central, materials require chemical analysis, legal or insurance consequences may follow, cultural-property concerns exist or the conclusion is likely to be publicly challenged.
Authentication may require collaboration among subject specialists, conservators, materials scientists, archivists, provenance researchers, handwriting specialists and manufacturing historians. Escalation is not failure; it is a disciplined recognition of the limits of the current examination.
Documentation
A comparison conclusion without an audit trail is difficult to review, defend or improve. The record should allow a later collector, buyer, insurer, heir or specialist to see what was examined, which references were trusted and why the conclusion was proportionate.
“The examined characteristics are consistent with documented first-issue examples, and no material contradiction was identified.”
“The object appears period-manufactured, but the available evidence does not distinguish the original issue from the documented early reissue.”
“The printing method and paper structure are inconsistent with the claimed production date.”
“The principal object appears authentic, but the attachment is a later replacement.”
Review
A conclusion may change when a factory archive is discovered, a new production variant is documented, a reference object is reclassified, counterfeit methods evolve or an earlier test is shown to have been misinterpreted.
A responsible record retains the evidence originally available, the reasoning used, the date of the opinion, later amendments and the reason for any change. Revising a judgement after stronger evidence appears is not necessarily a failure of method. Refusing to revise it may be.
Final collector rule
Do not ask, “How many things look right?”
Ask whether the complete pattern of materials, manufacture, documented history, legitimate variation and ageing makes better sense under the claimed identity than under the plausible alternatives.
Return to judging when provenance independently supports an authentication claim and when it only repeats an inherited story.
Review the foundation pages covering authenticity, attribution, claims, confidence, burden of proof and evidence-led judgement.
Continue into physical examination and learn how material choice, fabrication and assembly can support or contradict a claim.
Connect comparative method to the wider principle that conclusions should follow evidence rather than desire, reputation or price.
Use proportionate language for strong support, weak support, contradiction, uncertainty and conclusions that remain open to review.
Examine how reference objects and records should be qualified before they are allowed to define what genuine looks like.
Understand how external opinions should be scoped, documented and weighed alongside the physical and historical evidence.
