Absolute Transparency Standard (ATS)
What this tool is for
A framework for designing or auditing communication that comes in installments: a sermon series, a newsletter run, a campaign, a long document. It applies seven tests (Roadmap, Standalone, Borrowed Trust, Jump-Ahead, Reversibility, Stacking and Keystone) to ask whether a reader is told where the argument is going, can judge each piece on its own, and can walk back out if they disagree. It separates honest sequencing from drip-fed persuasion by one measure: whether the plan is disclosed. Use it in design mode while writing or in audit mode on something already published. Current version: 1.2.
The tool
ABSOLUTE TRANSPARENCY STANDARD (ATS) — CORE
A Design and Audit Framework for Sequential and Extended Communication — v1.2
Domain-neutral core specification. Applies to any sequenced or extended communication — a marketing campaign, an educational course, a journalism series, a long-form article, a political messaging plan, a sermon series, a single persuasive essay. Project-specific applications (theological, organizational, or otherwise) should be built as separate application notes referencing this document, not by editing it.
ORIGIN, STATED GENERICALLY
This tool was developed after documenting a real case: a content creator's body of work, read one piece at a time, was individually reasonable in every instance. Read as a whole, it revealed a complete, coherent position the creator never stated outright, built from claims distributed one per piece across many separate pieces. No single piece contained the conclusion. The conclusion only existed as the sum.
v1.2 adds two tests, developed from a further finding in that same case. A content-accuracy tool run on the single most important piece in that body of work would very likely have caught its central error — the piece's own cited source explicitly contradicted its own claim. What no content-accuracy tool could catch, because it has no way of knowing other pieces exist, is that the erroneous piece was the keystone for two others: its source material stated an explicit logical dependency (this claim only follows if that one is already true) that the body of work never disclosed to its audience. Six tests told a reader whether they could see the whole. A seventh was needed to tell them whether they could see how the parts depended on each other.
That structure isn't specific to any one domain. It's the mechanism behind drip propaganda, salami-slicing negotiation tactics, cult recruitment escalation, predatory sales funnels, and — in its legitimate, disclosed form — ordinary good teaching and marketing sequencing. This tool exists to tell the legitimate form from the exploit, by construction, regardless of subject matter.
WHAT THIS TOOL DOES
The distinguishing question is not pacing. It is disclosure. Does the audience know a larger structure exists, can they see how its parts depend on one another, and can they ask for the whole of it at any point and get a straight answer? Legitimate sequencing says yes, structurally, before delivery begins. The exploit depends on the answer being no.
Two modes:
- DESIGN MODE — building a new sequence or document from scratch, so the tests are satisfied by the structure itself.
- AUDIT MODE — checking an existing sequence or document, applying the same tests retroactively.
Two scopes:
- MULTI-UNIT SCOPE — the "pieces" are separate published items: posts, episodes, emails, chapters, sessions.
- SINGLE-DOCUMENT SCOPE — the "pieces" are internal units within one document: sections, paragraphs, or the individual steps of a single argument. All seven tests apply identically in both scopes; only the definition of "a piece" changes.
THE CORE PRINCIPLE
Trust earned through early, easier material is not the problem. The problem is narrow: later material must never depend on earlier-earned trust as a substitute for its own evidence, and claims that are logically dependent on one another must never be presented as though they stood independently. Earned trust may accompany independent support. It cannot replace it. A dependent claim may be introduced before its foundation is fully argued. It cannot be presented as though it had no foundation to depend on at all.
THE SEVEN TESTS
Test 1 — The Roadmap Test
Question: Is the full scope of the communication stated and findable, before or at any point during delivery?
Pass: A stated table of contents, a declared thesis, an opening line naming the destination.
Fail: Each unit presented as though it were the whole, with no signal that more exists or where it's headed.
Test 2 — The Standalone Test
Question: Would each unit remain fair and defensible if the full roadmap were attached to it?
Pass: Persuasive force comes from the unit's own evidence.
Fail: The unit only lands because the audience doesn't yet know where it's going.
Test 3 — The Borrowed Trust Test
Question: Does the hardest claim rely on credibility earned by earlier, easier material — or does it carry independent support?
Pass: Every claim, including the hardest, is supported on its own terms.
Fail: The hardest claims are the least independently supported, arriving only after trust has been banked elsewhere.
Test 4 — The Jump-Ahead Test
Question: Can the audience ask "what does this add up to" at any point and get a complete, undeflected answer?
Pass: The question is answerable and answered.
Fail: Deflected, delayed, or unanswerable because no one — including the author — has stated the whole.
Test 5 — The Reversibility Test
Question: If the audience encountered the LAST unit first, would it read as a fair preview or as exposure?
Pass: Any unit, read first, gives an honest preview of the whole.
Fail: Order is load-bearing for concealment, not just pacing — reading backward reveals what earlier units were withholding.
Test 6 — The Stacking Test (New in v1.2)
Question: If every distributed, individually reasonable claim across the sequence is gathered and stated together, in one breath, does the resulting cumulative claim match what the audience was told, or reasonably assumed, they were being offered at the outset?
Pass: The aggregated sum is recognizable as the thing the audience signed up for.
Fail: The aggregated sum reveals a conclusion — often more specific, more total, or more extreme — that no single unit disclosed, and that the audience would likely have hesitated over if it had been offered whole at the start.
Method: Do the aggregation as an actual exercise, not a judgment call — write out each unit's core claim in one plain sentence, then read all of them together as a single paragraph, then compare that paragraph against the sequence's stated (or implied) opening premise.
Test 7 — The Keystone Test (New in v1.2)
Question: When a sequence makes several claims, does it disclose which claims are logically dependent on others — so the audience can see where the argument's actual weight rests — or are dependent claims presented as separate, parallel assertions of equal and independent standing?
Pass: The sequence identifies, or its own logical structure makes plain, which claim is foundational and which claims are built on it — a reader who doubts the foundation knows to doubt what depends on it.
Fail: Claims that are genuinely sequential or dependent — where doubting one would rationally require doubting others — are instead presented as though a reader could accept or reject each independently, with no signal that rejecting one unravels more than itself.
Distinct from Test 3: Borrowed Trust is about psychological sequencing — a claim leaning on earned confidence built up over time. Keystone is about logical structure — a claim leaning on another specific claim's truth, a dependency that can exist even within a single moment, independent of any trust-building sequence at all, and that a source's own material may already state explicitly without the communication itself ever disclosing it.
Note on scope: This test asks only whether the dependency is disclosed, never whether the foundational claim is actually true. Whether a claim is true or false, contradicted or supported, is a content-accuracy question for a different tool. ATS asks only whether the audience was shown the load-bearing structure well enough to know where to direct their own scrutiny.
SINGLE-DOCUMENT MODE
A single extended piece — an article, an essay, a long pitch, a chapter, a speech — can run the identical mechanism internally, at the level of sections or claims rather than separate published items. The tests apply unchanged; only the unit of analysis shrinks.
Defining "a piece" inside one document: typically a section, a numbered claim, or a stage in an argument's progression from premise to conclusion.
Running the seven tests at document scale:
- Roadmap — Does the document state its own conclusion or scope near the start, or does the real point only surface deep in the piece with no early signal?
- Standalone — Would the document's final section survive being read with the actual thesis stated at the top?
- Borrowed Trust — Does the hardest claim have its own independent evidence, or is it resting on accumulated goodwill from earlier sections?
- Jump-Ahead — Could a reader, a few paragraphs in, accurately state where the document is heading?
- Reversibility — Read the final paragraph first, cold. Fair preview, or a different document than the opening promised?
- Stacking — Combine the document's section-by-section claims into one summary paragraph. Does it match the introduction?
- Keystone — Identify the document's foundational claim. Does the document say, anywhere, which of its other claims depend on that foundation holding — including cases where the document's own cited sources already state the dependency, and the document simply never carries that structure forward into its own argument?
DESIGN MODE
- State the full scope before writing unit one — a real roadmap, not a private outline.
- List every planned unit and its core claim.
- Identify the hardest claim, and give it independent support deliberately, not support borrowed from surrounding units.
- Map the actual dependency structure between your claims — which ones would fall if another one fell — and state that structure to your audience rather than presenting dependent claims as though they stood alone.
- Write a one-paragraph "what this adds up to" answer, available from unit one onward.
- Do the Test 6 aggregation exercise on your own planned sequence before publishing, and check the result against what you told your audience they were getting.
- Read the planned final unit as if it were the first thing anyone encountered. If it reads as a betrayal of the opening, redesign before publishing anything.
AUDIT MODE
Run all seven tests in order (adjusting unit definition per Single-Document Mode above if applicable). Record Pass / Fail / Partial per test, with specific cited evidence — not an aggregate score. Any Fail needs correction before the material reaches its audience.
SCOPE AND LIMITS
ATS checks structural disclosure only — can the audience see the whole shape of what's being delivered, and can they see how its parts depend on one another. It does not check, and was deliberately not extended in v1.2 to check, whether any individual claim is factually accurate, whether a citation actually supports what it's attached to, or whether a premise is contradicted by evidence elsewhere in the piece. Those are content-accuracy questions, and belong to tools built for content accuracy, not to this one. Two tests considered for v1.2 were rejected specifically because they crossed this line — a validity-framed load-bearing-assumption check and a citation-verification check — both of which substantially duplicated existing content-accuracy mechanisms in this project's other tools, at weaker calibration, and both of which would have blurred the boundary that keeps ATS distinct rather than becoming a second, softer copy of a tool that already exists. Test 7, Keystone, was kept only after being rebuilt around disclosure rather than validity — it asks whether a dependency was shown, never whether the thing depended on was true. ATS answers exactly one family of questions: is the delivery method hiding the shape of what's being delivered, including the shape of how its own claims depend on each other.
ATS Core v1.2. No domain-specific branding, examples, or application. For a worked application of this framework to a specific project or content domain, see the relevant application note.
# ABSOLUTE TRANSPARENCY STANDARD (ATS) — CORE ### A Design and Audit Framework for Sequential and Extended Communication — v1.2 *Domain-neutral core specification. Applies to any sequenced or extended communication — a marketing campaign, an educational course, a journalism series, a long-form article, a political messaging plan, a sermon series, a single persuasive essay. Project-specific applications (theological, organizational, or otherwise) should be built as separate application notes referencing this document, not by editing it.* --- ## ORIGIN, STATED GENERICALLY This tool was developed after documenting a real case: a content creator's body of work, read one piece at a time, was individually reasonable in every instance. Read as a whole, it revealed a complete, coherent position the creator never stated outright, built from claims distributed one per piece across many separate pieces. No single piece contained the conclusion. The conclusion only existed as the sum. v1.2 adds two tests, developed from a further finding in that same case. A content-accuracy tool run on the single most important piece in that body of work would very likely have caught its central error — the piece's own cited source explicitly contradicted its own claim. What no content-accuracy tool could catch, because it has no way of knowing other pieces exist, is that the erroneous piece was the *keystone* for two others: its source material stated an explicit logical dependency (this claim only follows if that one is already true) that the body of work never disclosed to its audience. Six tests told a reader whether they could see the whole. A seventh was needed to tell them whether they could see how the parts depended on each other. That structure isn't specific to any one domain. It's the mechanism behind drip propaganda, salami-slicing negotiation tactics, cult recruitment escalation, predatory sales funnels, and — in its legitimate, disclosed form — ordinary good teaching and marketing sequencing. This tool exists to tell the legitimate form from the exploit, by construction, regardless of subject matter. --- ## WHAT THIS TOOL DOES **The distinguishing question is not pacing. It is disclosure.** Does the audience know a larger structure exists, can they see how its parts depend on one another, and can they ask for the whole of it at any point and get a straight answer? Legitimate sequencing says yes, structurally, before delivery begins. The exploit depends on the answer being no. **Two modes:** - **DESIGN MODE** — building a new sequence or document from scratch, so the tests are satisfied by the structure itself. - **AUDIT MODE** — checking an existing sequence or document, applying the same tests retroactively. **Two scopes:** - **MULTI-UNIT SCOPE** — the "pieces" are separate published items: posts, episodes, emails, chapters, sessions. - **SINGLE-DOCUMENT SCOPE** — the "pieces" are internal units within one document: sections, paragraphs, or the individual steps of a single argument. All seven tests apply identically in both scopes; only the definition of "a piece" changes. --- ## THE CORE PRINCIPLE Trust earned through early, easier material is not the problem. The problem is narrow: **later material must never depend on earlier-earned trust as a substitute for its own evidence, and claims that are logically dependent on one another must never be presented as though they stood independently.** Earned trust may accompany independent support. It cannot replace it. A dependent claim may be introduced before its foundation is fully argued. It cannot be presented as though it had no foundation to depend on at all. --- ## THE SEVEN TESTS ### Test 1 — The Roadmap Test **Question:** Is the full scope of the communication stated and findable, before or at any point during delivery? **Pass:** A stated table of contents, a declared thesis, an opening line naming the destination. **Fail:** Each unit presented as though it were the whole, with no signal that more exists or where it's headed. ### Test 2 — The Standalone Test **Question:** Would each unit remain fair and defensible if the full roadmap were attached to it? **Pass:** Persuasive force comes from the unit's own evidence. **Fail:** The unit only lands because the audience doesn't yet know where it's going. ### Test 3 — The Borrowed Trust Test **Question:** Does the hardest claim rely on credibility earned by earlier, easier material — or does it carry independent support? **Pass:** Every claim, including the hardest, is supported on its own terms. **Fail:** The hardest claims are the least independently supported, arriving only after trust has been banked elsewhere. ### Test 4 — The Jump-Ahead Test **Question:** Can the audience ask "what does this add up to" at any point and get a complete, undeflected answer? **Pass:** The question is answerable and answered. **Fail:** Deflected, delayed, or unanswerable because no one — including the author — has stated the whole. ### Test 5 — The Reversibility Test **Question:** If the audience encountered the LAST unit first, would it read as a fair preview or as exposure? **Pass:** Any unit, read first, gives an honest preview of the whole. **Fail:** Order is load-bearing for concealment, not just pacing — reading backward reveals what earlier units were withholding. ### Test 6 — The Stacking Test *(New in v1.2)* **Question:** If every distributed, individually reasonable claim across the sequence is gathered and stated together, in one breath, does the resulting cumulative claim match what the audience was told, or reasonably assumed, they were being offered at the outset? **Pass:** The aggregated sum is recognizable as the thing the audience signed up for. **Fail:** The aggregated sum reveals a conclusion — often more specific, more total, or more extreme — that no single unit disclosed, and that the audience would likely have hesitated over if it had been offered whole at the start. **Method:** Do the aggregation as an actual exercise, not a judgment call — write out each unit's core claim in one plain sentence, then read all of them together as a single paragraph, then compare that paragraph against the sequence's stated (or implied) opening premise. ### Test 7 — The Keystone Test *(New in v1.2)* **Question:** When a sequence makes several claims, does it disclose which claims are logically dependent on others — so the audience can see where the argument's actual weight rests — or are dependent claims presented as separate, parallel assertions of equal and independent standing? **Pass:** The sequence identifies, or its own logical structure makes plain, which claim is foundational and which claims are built on it — a reader who doubts the foundation knows to doubt what depends on it. **Fail:** Claims that are genuinely sequential or dependent — where doubting one would rationally require doubting others — are instead presented as though a reader could accept or reject each independently, with no signal that rejecting one unravels more than itself. **Distinct from Test 3:** Borrowed Trust is about psychological sequencing — a claim leaning on *earned confidence* built up over time. Keystone is about logical structure — a claim leaning on *another specific claim's truth*, a dependency that can exist even within a single moment, independent of any trust-building sequence at all, and that a source's own material may already state explicitly without the communication itself ever disclosing it. **Note on scope:** This test asks only whether the dependency is disclosed, never whether the foundational claim is actually true. Whether a claim is true or false, contradicted or supported, is a content-accuracy question for a different tool. ATS asks only whether the audience was shown the load-bearing structure well enough to know where to direct their own scrutiny. --- ## SINGLE-DOCUMENT MODE A single extended piece — an article, an essay, a long pitch, a chapter, a speech — can run the identical mechanism internally, at the level of sections or claims rather than separate published items. The tests apply unchanged; only the unit of analysis shrinks. **Defining "a piece" inside one document:** typically a section, a numbered claim, or a stage in an argument's progression from premise to conclusion. **Running the seven tests at document scale:** 1. **Roadmap** — Does the document state its own conclusion or scope near the start, or does the real point only surface deep in the piece with no early signal? 2. **Standalone** — Would the document's final section survive being read with the actual thesis stated at the top? 3. **Borrowed Trust** — Does the hardest claim have its own independent evidence, or is it resting on accumulated goodwill from earlier sections? 4. **Jump-Ahead** — Could a reader, a few paragraphs in, accurately state where the document is heading? 5. **Reversibility** — Read the final paragraph first, cold. Fair preview, or a different document than the opening promised? 6. **Stacking** — Combine the document's section-by-section claims into one summary paragraph. Does it match the introduction? 7. **Keystone** — Identify the document's foundational claim. Does the document say, anywhere, which of its other claims depend on that foundation holding — including cases where the document's own cited sources already state the dependency, and the document simply never carries that structure forward into its own argument? --- ## DESIGN MODE 1. State the full scope before writing unit one — a real roadmap, not a private outline. 2. List every planned unit and its core claim. 3. Identify the hardest claim, and give it independent support deliberately, not support borrowed from surrounding units. 4. Map the actual dependency structure between your claims — which ones would fall if another one fell — and state that structure to your audience rather than presenting dependent claims as though they stood alone. 5. Write a one-paragraph "what this adds up to" answer, available from unit one onward. 6. Do the Test 6 aggregation exercise on your own planned sequence before publishing, and check the result against what you told your audience they were getting. 7. Read the planned final unit as if it were the first thing anyone encountered. If it reads as a betrayal of the opening, redesign before publishing anything. --- ## AUDIT MODE Run all seven tests in order (adjusting unit definition per Single-Document Mode above if applicable). Record Pass / Fail / Partial per test, with specific cited evidence — not an aggregate score. Any Fail needs correction before the material reaches its audience. --- ## SCOPE AND LIMITS ATS checks structural disclosure only — can the audience see the whole shape of what's being delivered, and can they see how its parts depend on one another. It does not check, and was deliberately not extended in v1.2 to check, whether any individual claim is factually accurate, whether a citation actually supports what it's attached to, or whether a premise is contradicted by evidence elsewhere in the piece. Those are content-accuracy questions, and belong to tools built for content accuracy, not to this one. Two tests considered for v1.2 were rejected specifically because they crossed this line — a validity-framed load-bearing-assumption check and a citation-verification check — both of which substantially duplicated existing content-accuracy mechanisms in this project's other tools, at weaker calibration, and both of which would have blurred the boundary that keeps ATS distinct rather than becoming a second, softer copy of a tool that already exists. Test 7, Keystone, was kept only after being rebuilt around disclosure rather than validity — it asks whether a dependency was shown, never whether the thing depended on was true. ATS answers exactly one family of questions: is the delivery method hiding the shape of what's being delivered, including the shape of how its own claims depend on each other. --- *ATS Core v1.2. No domain-specific branding, examples, or application. For a worked application of this framework to a specific project or content domain, see the relevant application note.*