Recognising unsourced claims: how to keep wonder and ask for evidence

An unsourced claim is not always false. It is a claim whose route back to evidence has not been shown. In tarot writing, such claims are common because images travel quickly, books quote books, and a confident sentence can feel older than it is. "This card has always meant..." "The ancients knew..." "The symbol proves..." "Readers agree..." These phrases may conceal a specific author, a later occult system, a personal association, a mistranslation, or no evidence at all.

The goal is not to become suspicious of every beautiful sentence. The goal is to learn what kind of sentence it is. A personal reflection can be valuable without a footnote. A historical claim needs historical evidence. A lineage claim needs a named tradition. A scientific claim needs more than symbolic resonance. Once the categories are clear, wonder does not disappear; it becomes more precise.

Listen for the claim word

Certain words should make a reader pause: always, never, ancient, secret, original, universal, proven, traditionally, scientifically, everyone, destined. None is automatically wrong. Each raises the standard of evidence.

"Traditionally, the Fool means..." invites questions: which tradition, which period, which deck, which author? "Ancient Egyptians used tarot..." requires a document or object linking ancient Egypt to tarot, not merely a later writer's assertion. "This frequency heals..." makes an empirical claim that needs appropriate scientific evidence, not an attractive card image.

The pause is not pedantry. It is a way of protecting readers from borrowed authority.

Identify the claim type

Use four labels.

Object claim: about a card, deck, maker, date, place, material, or image. Best supported by museum records, catalogues, publishers, or documented collections.

Historical claim: about origin, development, influence, or use over time. Best supported by scholarly history, primary documents, and object evidence.

Lineage claim: about what Waite, Crowley, Golden Dawn, Etteilla, a Marseille school, or another named method teaches. Best supported by primary texts or careful scholarship on that lineage.

Interpretive claim: about what an image may mean in a reading. Best supported by transparent method, deck context, position, and language that leaves room for alternatives.

An additional category, empirical claim, covers statements about medical, psychological, financial, legal, or causal effects. These need evidence appropriate to their stakes. Tarot correspondence alone is not enough.

Trace backward, not sideways

When a claim appears in several books or websites, do not count repetitions. Trace backward. Who is the earliest identifiable source? What evidence did that source use? Is it a contemporary witness, a later occult writer, a catalogue, an academic study, a deck guide, or an unattributed assertion?

Tracing backward often reveals a useful story. A modern keyword may lead to Waite; a correspondence chart may lead to Golden Dawn papers; an Egyptian origin claim may lead to Court de Gebelin; a card image may lead to a publisher reproduction rather than a museum object. The trail does not always end in proof, but it tells you what the claim actually is.

Record the trail in a note. A future reader should be able to see why you trust, qualify, or reject the statement.

The citation that does not support the sentence

A source can be present and still fail to support a claim. A book may cite a museum page that dates a card to the fifteenth century, then use that date to claim Egyptian origin. A source may discuss a myth and be cited as evidence that a tarot card derives from it. A modern teaching article may be cited as if it were a primary historical document.

Read the source itself. Ask whether it says what the sentence says. Does it discuss the same object, date, term, and conclusion? Is the quotation complete? Does the author of the source express uncertainty that the new writer has removed? Citation is not a charm. It is a relationship between a sentence and evidence.

The image without a provenance

An image can create false authority quickly. A faded card scan with ornate lettering looks historical; a modern digital redraw looks old; a collage claims an ancient source. Before using an image as evidence, find its provenance: collection, object number, maker or publisher, date range, and rights information. If the provenance is missing, label the image unverified.

Do not infer meaning from an image's apparent age. Vintage style is not a date. A sepia filter is not a source. The Archive should prefer fewer well-attributed images over a large gallery of contextless ones.

Evidence is not all the same weight

An artist statement can establish the artist's intention. A museum catalogue can establish an object's description and provenance. A primary occult text can establish what its author taught. A peer-reviewed or scholarly history can support an argument about development. A forum post may show contemporary reception. A personal journal can support a personal association.

Problems arise when one kind of evidence is asked to do another's job. A deck guide cannot prove a fifteenth-century practice; a museum object cannot prove a modern therapeutic interpretation; a personal experience cannot establish a medical effect. Matching source to claim is the heart of source literacy.

How to write a qualified sentence

Replace false certainty with precise attribution.

Instead of: "The High Priestess is the ancient goddess Isis."

Write: "Some modern occult and artistic interpretations connect the High Priestess with Isis-like imagery; this is a later symbolic association, not a documented origin of the tarot card."

Instead of: "The Tower predicts disaster."

Write: "In many modern reading traditions, the Tower can raise questions about disruption, collapse of an assumption, or sudden recognition; it should not be used as a prediction of literal catastrophe."

Instead of: "Mercury retrograde causes communication failures."

Write: "Mercury retrograde is a traditional astrological symbol often used to reflect on revision and communication; ordinary planning and fact checking remain the relevant practical tools."

Qualified language is not evasive. It tells the reader what they can actually rely on.

A claim review workflow for the Archive

Before publishing a sentence that sounds factual, follow this sequence:

1. Highlight the exact claim.
2. Label its type.
3. Find the strongest available source suited to that type.
4. Read the source, not only a quotation of it.
5. Record author, title, date, page or URL, and access date if online.
6. Write the claim in language that matches the evidence.
7. Add a note if credible sources disagree.

If no source can be found, choose one of three options: remove the claim, label it as personal or contemporary practice, or leave a research note for future work. Do not fill the gap with a confident tone.

A red-flag list for high-stakes claims

Reject or heavily qualify claims that promise diagnosis, treatment, cure, guaranteed income, investment outcomes, legal results, safety predictions, fertility outcomes, contact from the dead, or factual access to another person's secrets. Such statements can harm readers even when intended as spiritual comfort.

An Archive can discuss how tarot traditions have historically addressed illness, law, money, death, or relationships. It should not convert those histories into present-day advice. The distinction is crucial: describing a belief is not endorsing it as a decision tool.

Keep a living errata file

When you discover that a claim in your notes or Archive is weak, do not hide the correction. Keep an errata file with the original wording, why it was revised, the new source or wording, and the date. Public projects earn trust by correcting themselves visibly.

For private study, the same practice removes shame. Everyone inherits unsourced claims. The skill is not never repeating one; it is learning how to repair it.

The confidence ladder

Not every entry needs the same amount of evidence, but every entry should communicate how firmly it can be held. Use a confidence ladder.

High confidence: directly supported by a documented object, primary text, or multiple reliable sources that agree on the limited claim.

Moderate confidence: supported by a reasoned scholarly interpretation or well-documented lineage, with known limitations or disagreement.

Low confidence: plausible but incomplete; perhaps a visual comparison or a claim that requires further research.

Personal or creative: meaningful to a reader or artist, but not presented as historical or empirical fact.

The ladder does not turn an Archive into a spreadsheet. It teaches readers how to hold a sentence. A low-confidence claim may still be fascinating; it simply should not carry a high-stakes conclusion.

Search engines are indexes, not authorities

Search results reward popularity, repetition, technical optimisation, and recency. They do not automatically reward provenance. A result snippet can remove qualifying language; an AI-generated summary can invent a source; a social-media post can outrank a museum catalogue.

Use search to find possible sources, then leave the results page. Open the institution, book preview, catalogue, paper, or primary text. Check author, date, domain, and whether the page is actually describing the claim you want. If a source is inaccessible, do not cite a search snippet as though you read the source.

This habit matters especially for quotes. A quotation attached to a famous name may be paraphrased, mistranslated, or fabricated. Locate the work, edition, and context before using it in an educational entry.

The difference between plausible and documented

Some claims are plausible because they fit a pattern: a modern artist may have seen a famous painting; a card may resemble a mythic figure; a printer may have borrowed a popular motif. Plausibility can guide research. It is not the same as documentation.

Write the distinction explicitly. "The image resembles X" is an observation. "The artist was influenced by X" is a historical claim. The second needs a source: an artist statement, letter, sketchbook, interview, or scholarship that establishes the connection. If you cannot find one, retain the first sentence and leave the second open.

This practice protects both scholarship and imagination. It allows readers to notice resonances without turning every resonance into a secret genealogy.

What to do with a source you disagree with

Sometimes a source is clear, cited, and still wrong or unconvincing. Do not erase it. State the source's claim, identify the evidence it uses, and introduce a counter-source or reasoned limitation. The Archive should not manufacture consensus by omitting disagreement.

For example, an occult author may claim an Egyptian origin for tarot. That claim is historically important because it shaped later practice, even though modern historical scholarship does not accept it as an account of tarot's origin. The entry can preserve both facts: the claim mattered, and its historical evidence is inadequate.

Teaching readers to cite themselves

An Archive becomes a community resource when users learn to ask, "Where did I get this?" Encourage notes that include a source field, a deck field, and an interpretation field. A user who tags an association as personal has not made it less valuable. They have made it easier for another reader to understand.

The habit scales from a private notebook to a public encyclopedia: claim, source, scope, revision.

Sources and points of debate

• Museum collections such as the Metropolitan Museum of Art, the V&A, and the British Museum are important for object claims and image provenance.
• Michael Dummett and Decker, Depaulis, and Dummett are major historical sources for evaluating tarot origin claims; Waite, Crowley, Etteilla, and Golden Dawn materials are primary sources for their own systems.
• The Stanford Encyclopedia of Philosophy and scholarly encyclopaedias can provide orientation, but claims should still be checked against their cited sources where possible.
• The source hierarchy is contextual. A personal account may be the best source for personal experience, but it is not adequate evidence for a historical, scientific, or high-stakes factual claim.