Method

The authority given to Scripture

This is an evangelical project. It treats the Old and New Testaments as the word of God written, true in what they affirm, and the final court of appeal on the fate of the wicked. No tradition, no council, no scholar and no argument on this site outranks a biblical text read in its own context.

That commitment cuts in a direction some readers will not expect. It means the question is not which view is more comfortable, or more traditional, or better represented in the churches the author grew up in. It means the question is what the texts say. Where the texts are difficult, the difficulty is stated rather than resolved by assertion.

It also means the case can be defeated the way any biblical argument can be defeated: by showing that a text has been read against its context, that a word has been given a sense it does not carry, or that a pattern claimed across Scripture is not actually there.

Why the case is cumulative

No single verse settles this question, and the site does not pretend otherwise. The argument is cumulative: a large number of passages, read in their own contexts, point in the same direction, and the strength of the case lies in the convergence rather than in any one page.

A cumulative case carries an obvious risk. Weak arguments can be piled up until the heap looks impressive, and a reader has no way to tell the load-bearing pieces from the decoration. Two devices guard against that here. First, every page carries an evidence-role label, described below, which says plainly what kind of weight the argument is meant to bear. Second, the case map names the six principal claims and shows which pages support each of them, so a reader can attack the structure directly instead of arguing with the pile.

If a page is removed, the case map shows what the removal costs. Some pages could be deleted with no effect on the conclusion. Others could not, and those are labelled as core biblical arguments.

Direct evidence and inference

A statement in a text and a conclusion drawn from several texts are different things, and the site keeps them apart. When Paul writes that the wages of sin is death, that is direct evidence about the stated penalty. When the site argues that the death in view must be the second death because believers and unbelievers alike die physically, that is an inference: a conclusion assembled from more than one claim.

Inferences are not weaker by definition, and Scripture itself reasons this way. But an inference can fail in ways a quotation cannot, because it depends on every step holding. So inferences are marked as inferences, both in the page label and in the prose, and each page states what its argument establishes and what it does not establish by itself.

Where the author is offering his own reading rather than reporting a consensus, the page says so and the review status records it. That distinction is not a hedge. It is information a critic needs in order to aim at the right target.

How the opposing view is presented

Every page that touches a disputed passage states the eternal conscious torment reading before answering it, in the strongest form its own defenders would recognise. Where possible the traditional reading is drawn from writers who hold it, not from conditionalist summaries of it.

Three things are ruled out. The site does not caricature the traditional view, does not suggest that those who hold it are uncaring, dishonest or deceived, and does not claim that the matter is obvious. Millions of careful readers of Scripture have held the traditional view, including most of the people who taught the author to read the Bible at all.

Agreement is also recorded. Each passage page lists what both readings accept before it states where they part, because a disagreement is only interesting once the shared ground is clear.

How Bible translations are handled

Full passage displays use the World English Bible, which is in the public domain. Every displayed verse comes from a verified corpus held in the codebase, and no author can type Scripture text by hand into a page. A page names a reference and the text is rendered from the corpus, which removes the possibility of a misquoted verse and makes the translation visible on the page.

Where an argument turns on a specific modern rendering, the few words at issue are quoted, the translation is named, and the difference is explained in the site’s own prose rather than by reproducing the passage. That is a deliberate limit. A reference work of this size quotes Scripture far beyond what the incidental quotation allowances of the copyrighted translations cover, and a project asking to be judged on its honesty cannot help itself to text it has no licence to publish.

The source document quoted modern translations at length. The decision to move to a public-domain base text is recorded in the changelog along with its reasoning.

Two consequences are worth stating plainly. The World English Bible is a revision of the American Standard Version and its English is occasionally stiffer than a modern reader expects. And no translation is neutral on this subject, this one included, which is exactly why arguments that turn on a rendering are examined at the level of the underlying word rather than settled by citing a favourite version.

How original languages are handled

Greek and Hebrew appear on this site in their own script, with a transliteration and a gloss, marked up with the right language attribute so that assistive technology pronounces them correctly. Hebrew is set right to left. Nothing depends on a reader being able to read either script.

The standard for a language claim is higher than for an English observation. A word study is not an argument on its own: a lexicon entry lists a range of senses, and the question is always which sense the context requires. Where the site makes a claim about a word, it names the lexical range, states the reading it is arguing for, and states the competing readings alongside it. The lexicons and grammars used are cited by name in the source library.

The author of the source document does not read Greek or Hebrew formally. Any page whose argument depends on a technical claim about either language carries the specialist review pending status, and the specific claim in question is stated so that someone with the training can check it. Several such pages exist, and they are listed as such rather than quietly presented as settled.

How historical claims are reviewed

Historical claims are the easiest place for a theological argument to go wrong, because a striking fact travels faster than its evidence. Three rules apply.

1. A claim about what an early writer believed is tied to a passage in that writer, cited with a locator that a reader can look up, not to a secondary source reporting it.
2. Where the surviving evidence does not support a strong claim, the site states the narrower claim the evidence does support, and says what would be needed to settle the stronger one.
3. Where a familiar claim turns out to have no primary support, it is withdrawn in public rather than quietly dropped, with the reason recorded in the changelog.

One claim in the source document was narrowed under this rule and two were withdrawn entirely. Those decisions are in the changelog, each with the original claim, the problem, and what replaced it.

The evidence-role labels

Every page in the case carries exactly one evidence role. It is shown as a labelled badge at the top of the page and repeated in the page metadata. The label says what kind of weight the page is meant to bear, so that a reader can tell a load-bearing exegetical argument from a pastoral reflection without having to guess.

◆Core biblical argument

The argument rests directly on what a biblical text says, read in its own literary context. If this argument fails, the cumulative case is materially weaker.

◆Supporting biblical argument

The argument draws on biblical material, but it reinforces a conclusion established elsewhere rather than carrying weight on its own.

◆Historical or linguistic argument

The argument depends on evidence about history, or about how a Greek or Hebrew word is used. It is only as strong as the sources cited, which are named in full.

◆Theological inference

The conclusion is inferred from several biblical claims taken together. Scripture does not state it in so many words, and the inference is identified as an inference.

◆Further consideration

An observation, illustration or pastoral reflection. It is offered as something worth thinking about, not as evidence that the position is true.

The review-status labels

The second badge on every page records how far that page has been checked. It is deliberately visible rather than hidden in an editorial system, because a reader is entitled to know whether a page has been read against its own citations, and whether a known problem with it is outstanding.

✓Source checked

Every source cited on this page has been retrieved and checked against the claim it supports.

!Specialist review pending

The page makes a claim about Greek, Hebrew or primary historical texts that would benefit from review by someone with formal training in that field.

✓Reviewed

The page has been read end to end against the source document and its own citations.

!Revision needed

A known problem with this page is recorded and a revision is outstanding.

✓Author's present interpretation

This is the author's current reading, offered as his own judgement rather than as a settled conclusion. Other conditionalists read it differently.

✓Alternative conditionalist reading

The page sets out a reading that some conditionalists hold and others do not. It is not required by the wider case.

Nothing here is conveyed by colour alone. Each badge carries its own words and a text glyph, so the distinction survives a greyscale print, a screen reader and a colour vision deficiency alike.

The source hierarchy

Not all sources count the same, and the site ranks them in this order.

1. The biblical text, read in context. Everything else is subordinate to it.
2. Primary historical sources: the early Christian writers, Josephus, the classical authors, quoted from the work itself with a locator.
3. Lexicons, grammars and reference works, for what a word can mean, never for what it must mean in a given verse.
4. Academic scholarship and serious commentary, including scholarship that argues against this site’s conclusion.
5. Books, articles and sermons by advocates on either side, useful for stating a position accurately, not for establishing a fact.
6. Popular websites and videos, which are cited only where they are the thing being discussed, never as evidence for a historical or linguistic claim.

Every source used anywhere on the site has a record in the source library with its type, its perspective where that is relevant, its rights status, and when its link was last checked. A claim supported only by a source low in this list is either qualified on the page or not made.

Corrections and revisions

Anyone can submit a correction, a better source, an accessibility problem or a counterargument through the corrections form. There is no account and no CAPTCHA. The process after that is fixed.

1. The submission is recorded and read.
2. The claim is checked against the source it concerns, not against whether it helps or hurts the case.
3. If it is accepted, the affected page is changed and a revision record is written with the date, the section, the issue, and the decision taken.
4. The revision appears in the public changelog and on the revised page itself.
5. Nobody is named without asking. A submitter is credited only where they chose that option on the form.

Revisions are typed, so the record distinguishes a factual correction from a clarification, a source update, a translation change, a substantive revision and an accessibility fix. Nothing is silently edited: where a claim from the source document was narrowed or withdrawn, the original wording is preserved in the migration ledger described on the original document page.

How unresolved issues are shown

Open questions are published, not parked. Where the site does not know something, the page says so in the main flow of the argument rather than in a footnote, and the page carries a status that reflects it.

• A page with a known outstanding problem carries the revision needed status and states what the problem is.
• A page whose argument depends on a technical language or history claim carries the specialist review pending status and states the specific claim that needs checking.
• A reading held loosely by the author is labelled as his present interpretation, with his own hesitancy preserved rather than tidied away.
• A reading that some conditionalists hold and others do not is labelled as such, and the page states that the wider case does not depend on it.

A cumulative case that hides its weak points is not making an argument, it is running a campaign. The labels exist so that a critic can find the soft ground quickly.

Why philosophy is not treated as proof

Arguments about what a just God would or would not do appear on this site, because they appear in the discussion and because the source document raises them. They are never treated as biblical evidence, and they are labelled as further consideration or theological inference rather than as core biblical argument.

The reason is simple and it applies against this site as much as for it. Our sense of what a proportionate punishment looks like is formed by our own time and place, and it is not a reliable guide to what God has said. An argument that endless torment feels excessive is not evidence that Scripture teaches otherwise. An argument that sin against an infinite God must warrant infinite punishment is not evidence either, and the site says so on the page where that argument is examined.

This also rules out a particular kind of rhetorical shortcut. The site does not argue that conditional immortality must be true because the alternative is unbearable. Final judgment on this account is real, conscious, just, deserved and permanent, and nothing here is offered as a softening of it.

Why this stays open to correction

The author held the traditional view for about thirty-five years and now holds a different one. Whatever else that shows, it shows that at least one of those two positions was wrong, and that he is capable of being wrong about it for a long time while being entirely sincere.

So the project is built to be corrected. The evidence labels, the review statuses, the public changelog, the migration ledger and the source library all exist for the same reason: to make it possible for a reader who thinks this is mistaken to show exactly where, and to make it costly to quietly ignore them.

If you can show that a page has misread a text, misused a word, misreported a historical source or misstated the traditional view, please send it in. Accepted corrections are published with the reasoning, including the ones that make the case weaker.