Our sources

We only publish numbers we can trace to a primary source: the original study, official statistics, or the organization's own reporting. Secondary retellings, influencer polls, and unattributed "studies show" claims are rejected at the fact-gate.

Primary sources currently cited

No primary sources are cited yet — the fact gate is empty. Every source we rely on will be listed here with its year and method.

What we rejected, and why

This list matters as much as the one above. These are figures that circulate widely in writing about widowhood and that we could not trace to a primary source, so they do not appear anywhere on this site.

Subjects that normally take a year of friendship arrive in the first evening — Quantified empirical claim ('a year') presented as fact; no primary source. content-spec §0.5.
Closeness that would take months of ordinary small talk can be reached in one long night ... This is one of the more predictable things about human closeness. — Implicit research claim ('one of the more predictable things') with an unsourced duration; reads as 'studies show' without a study.
High-arousal situations produce fast attachment and are frequently remembered afterwards as fated meetings. — States a psychological finding as established fact with no source; the site is number-free/claim-free until a primary source is verified.
The reports are remarkably consistent ... Three elements come up almost every time. — Frequency claim about a body of reports, no survey or study behind it.
People frequently place this at the top of their list. — Unsourced frequency/ranking claim; sounds like poll data.
A conversation that went four hours. ... the thing that happened when you were nine. ... running into each other three times in a month. — Illustrative specifics, not statistics, but the site standard is number-free prose; concrete figures in a factual register invite misreading as data.
Conversations that pick up mid-sentence after six months of silence. — Concrete figure in a factual register describing what these friendships do; no source, and the site is number-free (content-spec §0.5). Same precedent as the 'four hours / three times a month' rejects above.
Not fate distributing rare matches, but a period that supplied the conditions. — Not a statistic — a verdict on how fate does or does not operate, banned by content-spec §0 HARD rule 4 ('Твердження про існування або неіснування долі. Ми описуємо позиції, не виносимо вирок'). The FAQ line 'not a cosmic allocation' has the same problem and also clips the §0 tone rule.
Three of these words are ordinary description; one of them arrived carrying a myth. — Count contradicts the table directly above it: that table attributes 'soulmate' in three of the four rows (platonic soulmate, soulmate sister, romantic soulmate) to Plato's Symposium. Only 'best friend' is listed as myth-free.
In learning psychology, a reward that arrives unpredictably produces more persistent behaviour than a reward that arrives every time. Relationships that alternate warmth with withdrawal reproduce that pattern exactly, and people in them often describe the strongest pull. — Research finding asserted as established fact with no citation — a 'studies show' in disguise — plus an unsourced extrapolation from laboratory schedules of reinforcement to human romantic relationships ('reproduce that pattern exactly'). The partial-reinforcement extinction effect is a real experimental literature but is contested across domains, and no primary source is cited or approved for this site. content-spec §0.5.
This is the part worth understanding, because it means intensity can be evidence of inconsistency rather than of depth. — Conclusion drawn from the unsourced reinforcement claim above; presents a causal inference about the reader's relationship as established fact, and edges into relationship diagnosis (content-spec §0.6: 'ознаки' не є тестом на здоров'я стосунків).
Fast closeness is not mysterious. Telling someone things you rarely say, in a stretch of time that mattered, builds a bond quickly. Crisis, travel and isolation all compress the process. — Psychological mechanism stated as established fact with no source — same shape as the already-rejected 'High-arousal situations produce fast attachment' on /meanings/soul-connection/. 'Fast closeness is not mysterious' additionally adjudicates the mystical reading, against §0 rule 4 and the tone rule.
That is normal, and it is not a sign the bond is supernatural. — Verdict on the metaphysical status of the reader's experience. content-spec §0 HARD rule 4 — the site describes positions and does not rule on them; same precedent as the /meanings/platonic-soulmate/ 'not fate distributing rare matches' reject.
that usage is now the more common one among people who have never heard it in a church — Comparative prevalence claim about language use, with no survey, corpus study or usage data behind it. Reads as data.
nobody in psychology uses it as a technical word — Universal negative asserted as fact; unverifiable as stated.
Twin flame vs soulmate is one of the most searched comparisons in this subject — Search-volume ranking stated as fact in the opening line — a statistic in words, with no published source. Keyword-tool data is internal research, not a citable primary source. content-spec §0.5.
The descriptions repeat themselves with striking consistency — Consistency/frequency claim about a body of accounts with no survey behind it. Direct precedent: 'The reports are remarkably consistent' rejected on /meanings/soul-connection/.
one shared experience underneath (intro) / "Same underlying experience, two ways of naming it." (closing line inside the SVG figure) — Asserts as fact that the two vocabularies name the same underlying experience — an empirical claim about inner states with no source, and it contradicts the page's own body, which contrasts a low-friction experience (soulmate) with a high-friction one (twin flame). Same class as the platonic-soulmate count that contradicted its own table.
a one-sided sense of recognition is common (FAQ 'Do twin flames recognise each other?') — Prevalence claim ('is common') presented as fact, no survey. Precedent: 'People frequently place this at the top of their list' rejected on /meanings/soul-connection/.
No, though the search suggestions insist otherwise ("what's higher than a soulmate?") — Unsourced claim about what search autocomplete shows: unverifiable, unstable over time, and reads as data about user behaviour.
Only one of them keeps producing new information after the second year. — Numeric threshold attached to an empirical claim about relationships; reads as a research finding, no source. content-spec §0.5.
FAQ 'Which is better, a soulmate or a life partner?': ... the second question is the one that keeps mattering after the first year. — Same numeric threshold in a factual register; implies a measured turning point that no source establishes.
no shared plans past next spring — sustained for a decade by choice — Concrete duration in a factual register with no source; same precedent as 'four hours' / 'three times a month' / 'six months of silence'.
Couples who merge lives discover this early, often during a first shared expense. — Unsourced frequency claim about couples in general; reads as survey data. Precedent: 'People frequently place this at the top of their list' rejected on /meanings/soul-connection/.
FAQ 'Do soulmates break up and get back together?': That pattern is common in relationships generally, with or without the word attached. — Prevalence claim ('common ... generally') about relationships at large with no study behind it.
a friend of twenty years — Concrete figure in a factual register with no source; the site is number-free (content-spec §0.5). Same class as the rejected 'six months of silence' (/meanings/platonic-soulmate/) and 'four hours' (/meanings/soul-connection/).
soulmates they left after two years — Unsourced concrete duration inside a sentence describing what people report; number-free standard, same precedent as 'sustained for a decade'.
Most lists name four — romantic, platonic, karmic, twin flame — but others give three, seven or twelve, drawn from different books with different lineages. — The folklore counts themselves are permitted here (they are explicitly framed as vocabulary, per §0.5 'фольклорні числа називаємо тим, чим вони є'), but 'Most lists' is an unsourced prevalence claim about a body of publications — no census of that literature exists.
FAQ 'What are the four types of soulmates?': 'Romantic, platonic, karmic and twin flame are the four that most lists use.' — Same unsourced prevalence claim ('most lists') in an FAQ answer — the position AI Overviews are most likely to lift verbatim, which makes it the highest-risk placement on the page.
For most people asking this question, the answer is more than one, which is usually why they are asking. — Population-level frequency claim about readers, stated as fact; no survey or published data behind it, and none could be cited.
almost every one of these lists is plural by design — Frequency claim over an uncounted body of lists. Precedent: 'The descriptions repeat themselves with striking consistency' (/meanings/twin-flame-vs-soulmate/).
'there is only one, and this is them' is one of the most effective reasons anyone ever gave themselves for staying somewhere bad — Superlative empirical claim about why people stay in harmful relationships, presented as fact — and placed inside the safety paragraph, where an unsourced assertion does the most damage. content-spec §0.5.
including the destined one, which many believers hold alongside the observation that timing, geography and circumstance intervene — 'many believers hold' is an unsourced claim about the distribution of a belief; no poll cited and none approved.
a word that quietly carries at least three meanings ... one word is doing three jobs — Internal contradiction rather than a sourcing failure: the page's own AnswerTable lists four senses, and the paragraph under it says 'the other three' out of four. A count that disagrees with the artefact beside it is the same defect as the rejected 'Three of these words are ordinary description' on /meanings/platonic-soulmate/.
Synchronicity — running into each other three times in a month, the same rare book, the same small town. — Verbatim repeat of a string already rejected on /meanings/soul-connection/ ('running into each other three times in a month'). Concrete figure in a factual register on a number-free site (content-spec §0.5).
know the same three people / the two weeks of not replying / whether you are calmer or more anxious two months in than you were in the first week — Further concrete figures in a factual register; same precedent as 'four hours', 'six months of silence', 'a friend of twenty years', 'past the second year'. Numbers used illustratively still read as data on a site that carries none.
This is one of the most reliable things about human attention, and it does not spare sceptics. / FAQ: That is how attention works, and it happens to believers and sceptics equally. — 'Studies show' in disguise — a psychological finding plus a universal claim about who it affects, with no source. Direct precedent: 'This is one of the more predictable things about human closeness' rejected on /meanings/soul-connection/.
Four kinds come up over and over. / directAnswer: 'they usually mean four things: synchronicity, repeating numbers, vivid dreams and a sense of having met before.' — The count of an explicitly given list is fine (see the folklore-count precedent on /signs/more-than-one-soulmate/); the frequency frame is not. 'Come up over and over' and 'usually mean' are prevalence claims about an uncounted body of reports. Precedent: 'The reports are remarkably consistent ... Three elements come up almost every time' (/meanings/soul-connection/).
Memory keeps what fits the current ending and quietly drops what does not ... Everyone does this, about everything. — Memory-research finding asserted as established fact and closed with a universal claim about all people. Same class as the rejected 'nobody in psychology uses it as a technical word' (/meanings/soul-tie/) and the mechanism-as-fact rejects on that page.
Dreams run on daytime material. Whatever occupies your attention shows up at night ... / FAQ: 'dreams draw on whatever occupied you during the day, so a person you think about often is likely to appear.' — Dream-continuity research finding stated as established fact with no citation, plus a likelihood claim ('is likely to appear'). content-spec §0.5 and the 'no studies show without a study' standard.
Shared context manufactures coincidence. Two people who move in the same city, industry, hobby or age bracket will run into each other ... / 'The mechanisms behind every experience described above run on attention, context and adrenaline.' — Mechanism asserted as an established general law ('will run into each other', 'every experience ... run on'). Same shape as the rejected 'Fast closeness is not mysterious ... Crisis, travel and isolation all compress the process' (/meanings/soul-tie/).
most of them are followed by an unremarkable spring / FAQ: 'Restless months are common; most are followed by ordinary weeks, which nobody writes up.' — Prevalence claims ('common', 'most') about outcomes with no survey behind them. Precedent: 'That pattern is common in relationships generally' (/meanings/soulmate-vs-life-partner/) and 'For most people asking this question...' (/signs/more-than-one-soulmate/).
Recognising a stranger's face is strange and lovely without any help. — Reads as a ruling that no supernatural agency is involved — content-spec §0 HARD rule 4 (the site describes positions, does not adjudicate), and it undercuts the paragraph's own promise two sentences earlier that 'this site does not rule on it'. Direct precedent: 'That is normal, and it is not a sign the bond is supernatural' (/meanings/soul-tie/).
FAQ 'What happens right before you meet your soulmate?': 'Nothing detectable, and no page can tell you otherwise.' / body: 'Usually nothing that could have been spotted at the time.' — Absolute negative asserted as fact about the world; it adjudicates the fate question from the sceptical side — the mirror image of a prediction, and equally banned by §0 rule 4. The refusal half ('no page can tell you') is the safe part.
The signs you met your twin flame are remarkably consistent across the writing that uses the phrase ... Four items, repeated almost everywhere the term appears. — Corpus-frequency claim stated as fact, with no content analysis or survey behind it. Verbatim precedent: 'The reports are remarkably consistent' (rejected, /meanings/soul-connection/) and 'The descriptions repeat themselves with striking consistency' (rejected, /meanings/twin-flame-vs-soulmate/). content-spec §0.5.
Across books, videos and forums, the list barely varies. — Claim about how little a body of published material varies — data-shaped, no source, no census of that literature exists. Same class as the two rejects above.
FAQ 'What are the signs of twin flames?': 'Twin flame writing lists four things almost every time' — 'almost every time' is an unsourced prevalence claim about a corpus, in the FAQ slot most likely to be lifted verbatim by an AI Overview. content-spec §0.5.
Mutual recognition is common in any strong mutual attraction (FAQ) / 'It is also a genuine feature of strong mutual attraction generally' (body) — Prevalence claim about attraction in general asserted as fact, no study. Direct precedent: 'a one-sided sense of recognition is common' rejected on /meanings/twin-flame-vs-soulmate/.
In the stories people tell, usually both sides report the same thing / FAQ: 'both sides often report the same jolt of familiarity' — 'usually' / 'often' quantifies a body of self-selected accounts nobody has counted, and presents it as a rate. Softer than the rejects above, same shape.
strong feeling arrives just as readily in relationships that fail as in ones that last — Comparative empirical claim about relationship outcomes ('just as readily') — a statistic in words, no source. content-spec §0.5.
Ordinary mechanisms produce recognition and intensity without making either less genuine: shared references and background, mutual vulnerability offered early, meeting during upheaval, and a nervous system that reads high arousal as significance. — Psychological mechanisms asserted as established findings with no citation — 'studies show' in disguise; the arousal clause is the misattribution-of-arousal literature, unnamed and unsourced. Verbatim precedent: 'High-arousal situations produce fast attachment' (rejected, /meanings/soul-connection/) and 'Fast closeness is not mysterious…' (rejected, /meanings/soul-tie/).
the label offers no protection, and it is the most common reason people give for staying somewhere they have already decided to leave — Superlative prevalence claim ('the most common reason people give') with no survey or clinical source — and it sits inside the mandatory SAFETY paragraph, where an invented statistic does the most damage. content-spec §0.5 plus the §0 safety rule ('Без конкретних номерів до fact-checker'а'). Direct precedent: the identical superlative rejected on /signs/more-than-one-soulmate/.
Nearly everything that determines how a relationship goes lives in the unremarkable middle — Quantified claim ('nearly everything') about the determinants of relationship outcomes, presented as fact; reads as a research finding, no source.
the sense of having been known within an hour of meeting — Concrete duration in a factual register describing what people experience; unsourced. Same class as the rejected 'four hours' (/meanings/soul-connection/), 'six months of silence' (/meanings/platonic-soulmate/) and 'the two weeks of not replying' (/signs/spiritual-signs/). Site is number-free per content-spec §0.5.
people describe this consistently, across cultures and generations — Cross-cultural consistency claim — the shape of an anthropological finding, with no study behind it. Direct precedent: 'The reports are remarkably consistent' (/meanings/soul-connection/) and 'The descriptions repeat themselves with striking consistency' (/meanings/twin-flame-vs-soulmate/).
Yes, and the description is remarkably stable. ... The description is the most reliable part of this whole subject. — Two unsourced claims about a body of accounts ('remarkably stable', 'the most reliable part') presented as established. No survey, corpus or study. content-spec §0.5.
Fast closeness has ordinary drivers that are not remotely disappointing: overlapping references and background, meeting in a context that already selects for similar people, honesty offered early and reciprocated, and shared stress, which compresses months of acquaintance into days. — Causal mechanism asserted as established fact, plus an unsourced quantified compression ('months ... into days'). Verbatim precedent already rejected on this site: 'Fast closeness is not mysterious ... Crisis, travel and isolation all compress the process' (/meanings/soul-tie/), 'High-arousal situations produce fast attachment' (/meanings/soul-connection/) and 'Ordinary mechanisms produce recognition and intensity…' (/signs/twin-flame-signs/). 'Ordinary drivers' additionally leans on adjudicating the mystical reading (§0 rule 4).
That is not dishonesty; it is how memory organises anything important. — A claim about how human memory works, stated as settled fact — a 'studies show' with no study. Direct precedent: 'Memory keeps what fits the current ending ... Everyone does this, about everything' (/signs/spiritual-signs/).
then the first genuinely hard year reads as proof that you picked wrong — Numeric threshold in a factual register about relationships. Direct precedent on this site: 'after the second year' and 'after the first year' both rejected on /meanings/soulmate-vs-life-partner/.
belief in destiny changes behaviour, in both directions ... treating a relationship as significant tends to make people more patient during a bad stretch — Behavioural effect of a belief asserted as an established finding. There is a real literature on destiny vs growth relationship beliefs, but nothing is cited here and nothing is approved for this site — an uncited finding is exactly the 'studies show' pattern §0.5 blocks. The spec does require a 'why the question matters practically' section, but it must be written as description of what people say the belief does for them, not as a result.
The sentence "we are soulmates, so I put up with it" is produced by this niche constantly. — Frequency claim about a body of discourse, unsourced, and sitting in the paragraph that carries the page's safety argument. Precedent: 'is one of the most effective reasons anyone ever gave themselves for staying somewhere bad' (/signs/more-than-one-soulmate/) and 'the most common reason people give for staying' (/signs/twin-flame-signs/).
It predicts intensity, not compatibility — Predictive claim about a feeling stated as an established finding. The site's approved framing is a stated limit, not a positive prediction — see the accepted fix on /meanings/soul-tie/: 'Intensity is not evidence of compatibility.'
FAQ 'Does everyone have a soulmate?': ... most people have several over a lifetime, including friends and family members. — Population-level frequency claim stated as fact in an FAQ answer — the slot AI Overviews lift verbatim, so the highest-risk placement on the page. Precedent: 'For most people asking this question, the answer is more than one' (/signs/more-than-one-soulmate/).
Yes, and most people who use the word already manage it. — Unsourced prevalence claim about how people use the word. Same class as 'that usage is now the more common one' (/meanings/soul-tie/).
FAQ 'Do soulmates end up together?': People routinely describe someone as a soulmate precisely because the connection mattered and the relationship still ended. — 'Routinely' is a frequency claim about how people speak, with nothing behind it. Precedent: 'a one-sided sense of recognition is common' (/meanings/twin-flame-vs-soulmate/).
Structural, not sourcing: the FAQ block carries 7 questions and internalLinks carries 5 in-site links. — content-spec §0 GEO block requires FAQ 5–6; the internal-linking rule is 2–4 in-site links (hub + neighbours). Recorded here so it is not lost, though it is a spec deviation rather than a fact-gate failure.
Those reports are consistent and there is no reason to doubt them. — Consistency claim about an uncounted body of reports, stated as fact. Verbatim-class precedent: 'The reports are remarkably consistent' (/meanings/soul-connection/), 'The descriptions repeat themselves with striking consistency' (/meanings/twin-flame-vs-soulmate/), 'people describe this consistently, across cultures and generations' (/the-idea/do-soulmates-exist/).
There are ordinary things that produce that combination: shared references and context, early mutual vulnerability, meeting during upheaval, and a body that reads high arousal as significance. — Psychological mechanisms asserted as established causes with no source. This is the fourth appearance of the same unsourced paragraph on this site — already rejected on /meanings/soul-connection/, /meanings/soul-tie/, /signs/twin-flame-signs/ and /the-idea/do-soulmates-exist/. content-spec §0.5.
For the experiential claim, the evidence is ordinary and already in hand: consistent descriptions from very large numbers of people across decades. That is how we establish that any subjective experience exists. — The page's evidential core is an unsourced quantity-plus-consistency claim ('very large numbers', 'across decades') presented as evidence. No survey, corpus or study is cited, and none is approved for this site. It sits in an H2 answering 'What would count as evidence' — the sentence an AI Overview is most likely to lift as data.
The language of destined pairs is among the most common reasons people give for staying somewhere they have already decided to leave. — Superlative/prevalence empirical claim about why people stay in harmful relationships, presented as fact and placed INSIDE the mandatory safety paragraph, where an unsourced assertion does the most damage. content-spec §0 safety rule ('Без конкретних номерів до fact-checker'а') plus §0.5. Identical construction already rejected on /signs/more-than-one-soulmate/ and /signs/twin-flame-signs/.
Search the term and much of what returns is not description — it is a product. — Claim about the composition of search results: unverifiable, unstable over time, reads as data about the SERP. Direct precedent: 'No, though the search suggestions insist otherwise' (/meanings/twin-flame-vs-soulmate/) and 'one of the most searched comparisons' (same page). Internal keyword/SERP research is not a citable primary source.
A commercial layer grew around this particular phrase in a way it did not grow around ordinary words for love. / The most valuable customer for that layer is someone waiting for a person who is not currently in their life. / reassurance is sold by the session. — Market claims stated as established fact — a comparative growth claim across vocabularies and a superlative about customer value — with no industry data, filings or study behind them. Same class as the rejected superlative on /signs/more-than-one-soulmate/. content-spec §2 authorises DESCRIBING the industry ('чесно називаємо друге — без інструкцій'), not asserting its economics as measured fact.
Notice how rarely anyone selling one invites you to check it afterwards. — Frequency claim ('how rarely') about the behaviour of an uncounted group, asserted as fact.
H2 'Why does the framework never tell anyone to leave?' / 'Every disappointing outcome converts into evidence that the process is on track.' — Universal claims ('never', 'every') over a body of material nobody has surveyed. Precedent: 'nobody in psychology uses it as a technical word' (/meanings/soul-tie/) rejected as an unverifiable universal negative. As an H2 it is also among the most quotable lines on the page.
FAQ 'Do twin flames recognise each other?': Mutual recognition is common in any strong mutual attraction, so it does not mark a connection as unique. — Prevalence claim ('is common') presented as fact in an FAQ answer — the slot AI Overviews lift verbatim. Verbatim string already rejected twice: /meanings/twin-flame-vs-soulmate/ and /signs/twin-flame-signs/.
FAQ 'Is a soul connection the same as a twin flame?': People often swap them mid-conversation, which is where most disagreements about the terms begin. — 'most disagreements' is a population-level claim about how arguments arise, with nothing behind it. Precedent: 'That pattern is common in relationships generally' (/meanings/soulmate-vs-life-partner/).
FAQ 'What are the 7 stages of twin flame?': 'They are folklore in numbered form: meeting, turbulence, separation, return.' — Two defects. (a) Naming the folklore number as folklore is permitted (§0.5), but the answer then partially ENUMERATES the stage sequence — content-spec §0 HARD rule 3 bars стадії / runner-chaser / reunion, and the sibling page /signs/twin-flame-signs/ carries an explicit 'НЕ пишемо: стадії'. Reproducing an abridged stage list in an FAQ is exactly the fragment an AI Overview would republish as this site's teaching. (b) Four items answer a question about seven — an internal count mismatch of the same class as 'Three of these words are ordinary description' (/meanings/platonic-soulmate/).
That myth is the ancestor of the English word *soulmate*. — Etymological claim stated as fact. The Symposium myth is an ancestor of the IDEA; the English compound has its own separate attestation history, and no lexicographic or philological source is cited. content-spec §2 (/the-idea/where-the-idea-came-from/) explicitly warns against inventing years, translations and derivations — naming work and author is allowed, deriving a word is not.
English invents words for feelings constantly, and a term that spreads this fast is generally naming something people struggled to name before. — Linguistic generalisation ('generally') plus an unsourced diffusion claim ('spreads this fast') asserted as fact. Same 'studies show without a study' shape as 'This is one of the most reliable things about human attention' (/signs/spiritual-signs/).
the parts that arrive in the first hour … None of that is visible in the first hour, which is exactly when recognition arrives. — Concrete interval in a factual register on a number-free site, used twice as if measured. Precedent chain: 'four hours' (/meanings/soul-connection/), 'six months of silence' (/meanings/platonic-soulmate/), 'a friend of twenty years' (/signs/more-than-one-soulmate/), 'within an hour of meeting' (/the-idea/do-soulmates-exist/).
SVG figure, centre column: 'Reported constantly.' — Frequency claim inside the diagram — a quotable artefact, and the hardest place to qualify a claim. No survey behind it.
Passed on this page, recorded for the record: the mandatory SAFETY paragraph ('Where this page stops') is present and correctly routed (therapist/counsellor, domestic-abuse service, helpline in your country) with NO specific phone numbers; no astrology/numerology; no predictions; no verdict for or against fate (§0 rule 4 respected — 'this page does not rule on it'); no relationship diagnosis; no advice on changing another person; 4 internal links, all in-site; directAnswer within 40–60 words; FAQ count 7 exceeds the §0 GEO range of 5–6. — Not a rejection. Recorded so the writers keep these properties through the rewrite, and so the FAQ-count deviation (same defect logged on /the-idea/do-soulmates-exist/) is not lost.
Stories that begin "I knew immediately" are almost always assembled afterwards, with the ending already known. That is not dishonesty; it is how memory organises anything important. — Two unsourced empirical claims: a frequency quantifier ('almost always') and a memory-science mechanism ('how memory organises anything important') stated as established fact. Same shape as the already-rejected 'High-arousal situations produce fast attachment' on /meanings/soul-connection/. content-spec §0.5.
A shared context, a stretch of stress endured together, and a willingness to be quickly honest will reliably produce the feeling people call recognition — and the feeling is not counterfeit because it has a mechanism. — Causal claim with a reliability guarantee ('will reliably produce') presented as an established finding, no study behind it; 'it has a mechanism' asserts the explanation is settled. Precedent: the identical mechanism claim was rejected on /meanings/soul-connection/ and /meanings/soul-tie/.
the other is more predictive of a workable life together — Predictive-validity claim = 'studies show' without a study. §2 commissions the behavioural angle on /meanings/soulmate-vs-life-partner/, but predictiveness may not be asserted without a cited primary source.
The everyday sense is quieter and probably more common in ordinary speech. — Unsourced comparative prevalence claim about language use; hedged with 'probably' but still reads as a corpus finding. Same precedent as the /meanings/soul-tie/ 'now the more common one' reject.
What exists is a set of experiences people consistently describe — 'Consistently' asserts a pattern across reports without a survey; same shape as the rejected 'the reports are remarkably consistent'.
intensity and compatibility are separate properties that sometimes travel together and often do not — The thesis itself is §0-sanctioned, but 'sometimes ... often do not' adds an unsourced frequency distribution.
In current English, "soulmate" is used in four distinguishable ways. — Weakest of the rejects — §2 commissions the four-use map and the FAQ correctly labels 'four types' as folklore from books and blogs. But as written the body asserts a fact about English rather than presenting the page's own grouping, which is what §0.5 requires of folklore-shaped counts.
directAnswer: 'real experiences that predict little. The parts that do carry information are quieter' / H2 'Which of these signs predict anything?': 'Two of them, roughly ... Safety and respect are behavioural: they show up in what people do over months, and they keep predicting something long after the first impression has faded.' — The page's central move is a claim of predictive validity — that some signs forecast relationship outcomes and others do not. That is a research finding about relationships asserted as established fact, with no source, and it sits in the directAnswer (the slot AI Overviews lift verbatim). content-spec §0.5 and the 'no studies show without a study' standard; it also edges into §0.6 (ознаки не є тестом на здоров'я стосунків) by ranking signs as more or less diagnostic.
FAQ 'What signs are soulmates?': 'None of them has been shown to forecast whether a relationship lasts, works or stays kind.' — Negative claim about the state of the research literature ('has been shown'), asserted without any literature cited. Same class as the rejected universal negative 'nobody in psychology uses it as a technical word' (/meanings/soul-tie/).
AnswerTable row: 'Being treated with respect under strain — The most informative thing on this list' — Superlative empirical ranking presented as fact inside a structured artefact (highest AIO-lift risk). Precedent: 'one of the most effective reasons anyone ever gave themselves for staying somewhere bad' rejected on /signs/more-than-one-soulmate/.
Intro: 'a large, consistent body of description — people reporting what the experience felt like' — Consistency/volume claim about an uncounted body of reports. Verbatim precedent: 'The reports are remarkably consistent' (/meanings/soul-connection/) and 'The descriptions repeat themselves with striking consistency' (/meanings/twin-flame-vs-soulmate/).
'Four things recur, almost regardless of who is speaking.' / 'In accounts of long friendships this is the dominant note, and it is the one most people mean by the word in daily use.' — The count of an explicitly given list is permitted (folklore-count precedent), but 'almost regardless of who is speaking', 'the dominant note' and 'the one most people mean by the word in daily use' are prevalence claims about speakers and usage with no survey or corpus data. Precedent: 'that usage is now the more common one' (/meanings/soul-tie/); 'Four kinds come up over and over' (/signs/spiritual-signs/).
FAQ 'What are the first signs of a soulmate?': 'They are also the least informative stage of any relationship, because nothing has yet been tested by disagreement, strain or time.' — Superlative empirical claim about relationship stages stated as fact, no source.
'Any of them can be true of a relationship that ends badly, and false of one that lasts fifty years.' — Concrete figure in a factual register on a number-free site (content-spec §0.5). Direct precedent: 'a friend of twenty years' (/signs/more-than-one-soulmate/), 'six months of silence' (/meanings/platonic-soulmate/), 'sustained for a decade' (/meanings/soulmate-vs-life-partner/).
'Intense situations produce intense feelings; so do uncertainty, intermittent attention and fear of loss.' / 'Fear and intensity produce overlapping sensations.' — Psychological mechanisms asserted as established fact with no citation — 'studies show' in disguise. 'Intermittent attention' is the reinforcement literature already rejected once on /meanings/soul-tie/; the arousal claim mirrors 'High-arousal situations produce fast attachment' (/meanings/soul-connection/).
'The "I knew immediately" account is usually assembled afterwards ... That is normal memory, not dishonesty — but it means the evidence for immediate knowing is collected almost entirely from the winners.' — Memory-research finding stated as established fact ('normal memory') plus unsourced proportions about where such accounts come from ('usually', 'almost entirely'). Precedent: 'Memory keeps what fits the current ending ... Everyone does this, about everything' (/signs/spiritual-signs/) and 'it is how memory organises anything important' (/the-idea/do-soulmates-exist/).
'There is no feeling that only occurs with the right person.' — Universal negative about inner experience asserted as fact, and it quietly adjudicates the destiny question from the sceptical side — the same defect as 'Nothing detectable' (/signs/spiritual-signs/) and 'not a sign the bond is supernatural' (/meanings/soul-tie/), banned by §0 HARD rule 4.
'The belief that the right person must therefore stay is where the word does the most damage' / 'That reasoning has kept people in relationships they had already decided to leave.' — Superlative causal claim about harm plus an unsourced outcome claim about what has happened to people — sitting in the safety material, where an unsourced assertion does the most damage. Direct precedent: the identical superlative rejected on /signs/more-than-one-soulmate/ and /signs/twin-flame-signs/.
Structural, not sourcing: the FAQ block carries 7 questions. — content-spec §0 GEO block specifies FAQ 5–6 drawn from real PAA. Recorded so it is not lost, though it is a spec deviation rather than a fact-gate failure.
Type "soulmates Plato" into anything and you get one image: two halves of a single being, wandering the world, looking for each other. — Claim about what search engines return: unverifiable, unstable over time, and data-shaped. Direct precedent: 'No, though the search suggestions insist otherwise' rejected on /meanings/twin-flame-vs-soulmate/.
Two details get lost every time this is retold. — Frequency claim ('every time') about an uncounted body of retellings, stated as fact. Precedent: 'The descriptions repeat themselves with striking consistency' (/meanings/twin-flame-vs-soulmate/) and 'Four kinds come up over and over' (/signs/spiritual-signs/). The count 'two' itself is fine — it enumerates the page's own list.
The twentieth-century layer is the one most often misdated by the internet. — Superlative frequency claim about a body of online content ('most often misdated') with no corpus, survey or published source. Same class as the rejected search-volume ranking on /meanings/twin-flame-vs-soulmate/.
the vocabulary is younger than the people quoting it usually assume — Unsourced claim about what readers assume ('usually assume') — a prevalence statement about beliefs with no survey behind it. Precedent: 'many believers hold' (/signs/more-than-one-soulmate/).
FAQ 'Is there a scientific explanation of soulmates?': 'It does describe the ingredients of the experience: repeated proximity, similarity, fast mutual disclosure, high-arousal circumstances, and memory that tidies a story afterwards. Those describe how recognition arises.' — Attributes a body of findings to science with no study named — 'studies show' in disguise — and asserts a causal mechanism ('how recognition arises'). Direct precedents already rejected on this site: 'High-arousal situations produce fast attachment' (/meanings/soul-connection/), 'The mechanisms behind every experience described above run on attention, context and adrenaline' (/signs/spiritual-signs/), 'Ordinary mechanisms produce recognition and intensity…' (/signs/twin-flame-signs/), 'Fast closeness has ordinary drivers…' (/the-idea/do-soulmates-exist/). Highest-risk placement: FAQ answers are what AI Overviews lift verbatim.
Body, 'Is there a scientific explanation of soulmates?': 'What has been described in ordinary terms is the machinery of the experience: why recognition arrives so fast and feels like evidence.' + the five labelled bullets (Proximity and repetition / Similarity and shared context / Fast mutual disclosure / High-arousal circumstances / Retrospective assembly). — The section presents five research-flavoured mechanisms as established description ('the machinery of the experience', 'why recognition arrives so fast') under a heading promising a scientific explanation, with zero sources. The bullets are bare labels in this draft, but the frame licenses whatever prose fills them, and not one of the five has an approved primary source on this site.
FAQ 'Why did the soulmate idea become so popular?': 'Once marriage became a matter of choice rather than arrangement, people needed a way to explain why this person and not another.' — Social-history claim about marriage across societies, asserted as established fact with no historian or primary source cited. Naming a work and its author is permitted by the spec ('це не статистика'); asserting an uncited historical transition as the cause of a modern belief is not.
Structural, not sourcing: the FAQ block carries 7 questions and internalLinks carries 5 in-site links. — content-spec §0 GEO block requires FAQ 5–6; the internal-linking rule is 2–4 in-site links (hub + neighbours). Same deviation already recorded on /the-idea/do-soulmates-exist/.
The reports cluster around three things, and they are the same three that appear across [the signs hub](/signs/): recognition, ease and speed. — Frequency/consistency claim about a body of reports ('the reports cluster'), no survey behind it. Direct precedent: 'The reports are remarkably consistent ... Three elements come up almost every time' rejected on /meanings/soul-connection/, and 'Four kinds come up over and over' on /signs/spiritual-signs/.
The dozen equally pleasant first meetings that led nowhere are not retrieved at all — Illustrative figure in a factual register; the site is number-free (content-spec §0.5). Precedent: 'four hours', 'three times a month', 'six months of silence', 'a friend of twenty years'.
Someone who met their partner in a dull way and stayed forty years will still, eventually, have a good version of the story. That does not make the forty years less true. — Concrete duration in a factual register with no source; same class as the rejected 'sustained for a decade by choice' (/meanings/soulmate-vs-life-partner/).
SVG figure labels: 'Eight unremarkable meetings. One of them later mattered.' / 'The other seven are not remembered, because nothing later made them worth keeping.' — Numbers inside a factual diagram. Even labelled Illustration, an eight-of-which-one ratio reads as a base rate for how often a meeting turns into something; no source supports any ratio. content-spec §0.5.
FAQ 'What age do most soulmates meet?': Meetings that people later call fated happen at school, at forty, after divorce, in a hospital waiting room. — Names an age one sentence after promising 'There is no age this site will name' — self-contradiction plus a figure in a factual register on an age question. content-spec §0.5.
it turns an unlucky decade into a personal verdict — Duration figure in a factual register on a number-free site; low severity, same rule.
Meetings that turn into long relationships tend to come out of contexts with three ordinary properties. — Empirical claim about what long relationships come out of, asserted as established fact with no study — a 'studies show' in disguise, and it is the load-bearing claim of the page's practical section.
Repeated low-stakes exposure lets familiarity build on its own, which is how most reported 'we just clicked' moments actually get their head start. — Psychological mechanism (mere exposure) asserted as fact, plus a 'most' frequency claim about accounts. Same shape as the rejected 'High-arousal situations produce fast attachment' (/meanings/soul-connection/).
People show their temperament while cooking, rehearsing, canvassing or fixing something, which is far more informative than an interview across a table. — Comparative empirical claim ('far more informative') about how well different contexts reveal character; no source.
Places that gather people around what they care about ... tend to produce compatibility that survives contact with real life. — Predictive claim about what produces durable compatibility, asserted as fact with no source.
they are good at introductions and poor at repetition ... skips the part that usually does the work, which is seeing the same person behave normally across weeks — Asserts what 'usually does the work' in forming relationships — an unsourced causal finding comparing apps with offline contexts.
Reading intensity as identification is the single most common mistake in this subject. — Superlative frequency claim with no data. Precedent: 'one of the most effective reasons anyone ever gave themselves for staying somewhere bad' (/signs/more-than-one-soulmate/).
The same physical state accompanies attraction, nerves, novelty and fear, and it appears in meetings that end well and meetings that end badly. — Physiological finding stated as established fact ('the same physical state'), no source.
That is logistics, not destiny, and this site does not rule on whether destiny is behind the logistics. — Rules on the metaphysical reading in the first clause and retracts it in the second. content-spec §0 HARD rule 4. Precedent: 'not fate distributing rare matches' (/meanings/platonic-soulmate/), 'not a sign the bond is supernatural' (/meanings/soul-tie/), 'strange and lovely without any help' (/signs/spiritual-signs/).
FAQ: 'Nothing reliable, and anyone listing omens is describing hindsight.' / 'There is no signal to read.' / body: 'Nothing that can be checked in advance.' — Absolute negatives asserted as fact about the world — the sceptical mirror image of a prediction, banned by §0 rule 4 alongside the predictions themselves. Exact precedent set the same day on /signs/spiritual-signs/ ('Nothing detectable, and no page can tell you otherwise'). 'Anyone listing omens' also clips the §0 tone rule (a believing reader should finish the page without feeling mocked).
The idea that a person was meant for you is the most common reason people give for staying somewhere they are being hurt. — Superlative prevalence claim ('the most common reason') about why people stay in harmful relationships, asserted as fact with no survey or study. Worst possible placement: inside the safety paragraph, where an invented ranking does the most damage. Direct precedent: the near-identical 'one of the most effective reasons anyone ever gave themselves' rejected on /signs/more-than-one-soulmate/. content-spec §0.5.
the sort of openness that is easier at nineteen than at forty-one — Two concrete ages attached to a developmental claim about openness; reads as a measured finding. Precedent: 'four hours' / 'three times a month' / 'six months of silence' / 'after the second year' / 'a friend of twenty years' all rejected as figures in a factual register.
FAQ 'Is it true that soulmates are real?': People live it, describe it consistently, and remember it for decades. — Two unsourced empirical claims in one sentence: consistency across accounts (precedent — 'The reports are remarkably consistent', /meanings/soul-connection/) and a duration ('for decades') with no source. Sits in an FAQ answer, the text most likely to be lifted verbatim by an AI Overview.
FAQ 'Can you have multiple soulmates?': most people who use the word that way name several — Prevalence claim about word users ('most people') presented as fact; no survey, corpus or usage data. Same class as 'that usage is now the more common one' (/meanings/soul-tie/) and 'For most people asking this question' (/signs/more-than-one-soulmate/).
FAQ 'What if I have never felt it': 'Many people report it first appearing late, and many describe it about a friend rather than a partner.' / body: 'Many accounts of this feeling arrive late, and many arrive about a friend rather than a partner.' — 'Many people report' is survey-shaped language with no survey behind it, and it appears twice (FAQ + body). Precedent: 'People frequently place this at the top of their list' (/meanings/soul-connection/).
The plural version tends to be the one people arrive at with age, as the word slowly shifts from 'the one' to 'the ones who fit.' — Claimed trend in language use across the lifespan, stated as fact; no longitudinal or usage data exists to support it.
AnswerTable 'Same question, four dictionaries', cells: 'Yes, and almost certainly several' / 'Common, though not universal' / 'Usually yes, looking backwards' — Frequency verdicts sitting in a table stamped checkedOn="August 26, 2026" — the stamp implies these prevalences were checked, and none were. The table is the page's most citable element (AIO magnet), which makes an implied-data cell the highest-risk text on the page.
Most arguments about this question are people answering from different dictionaries and assuming the other side is being difficult. Neither side usually notices. — Lower severity, but still a quantified claim about how arguments generally go ('Most' / 'usually'), asserted as fact.
FAQ 'Do soulmates usually marry?': No usual pattern can be quoted here, because no figure worth trusting exists on this page. — Not a statistic, but an overreaching universal ('no figure worth trusting exists') where the truthful statement is narrower: none has been verified for this site.
FAQ 2: "real experiences, produced by attraction and nervous systems, not by cosmic matching" — Verdict on the metaphysical cause of the reader's experience — rules out a cosmic mechanism as fact. content-spec §0 HARD rule 4 (описуємо позиції, не виносимо вирок). Direct precedent: /meanings/soul-tie/ 'it is not a sign the bond is supernatural' — rejected.
FAQ 1: "Whether faces themselves converge is disputed" — Asserts the state of a research literature (that a scientific dispute exists) with no study named, no year and no URL — a 'studies show' in disguise. Nothing on facial convergence is approved for this site.
"There is a competing observation, equally common, that people are drawn to faces unlike anything in their family." — Comparative prevalence claim ('equally common') with no survey or corpus behind it. Precedent: /meanings/soul-tie/ 'that usage is now the more common one' — rejected.
"Copied expressions. This is the strongest everyday explanation for the 'they look alike' impression." — Ranks explanatory power ('the strongest') as if it had been measured; no source compares these mechanisms. Same class as the unsourced ranking claims rejected on /meanings/twin-flame-vs-soulmate/.
"That framing manufactures resemblance in strangers too." — Empirical claim about a photographic effect on unrelated faces, asserted as fact with no source.
FAQ 6 ('Is there a scientific explanation for couples looking similar?'): "familiar faces tend to feel more likeable" — A research finding (mere exposure) stated as a general fact inside an answer explicitly headed 'scientific explanation' — invites the reader and AI Overviews to take it as sourced science. No primary source cited or approved.
FAQ 4: "Strangers resemble each other constantly, which is why doppelganger stories exist at all." — Frequency claim ('constantly') presented as fact about unrelated faces; no source. Precedent: 'People frequently place this at the top of their list' — rejected on /meanings/soul-connection/.
it is consistent enough across accounts to be treated as a genuine category of experience — Consistency claim over an uncounted body of accounts, used as the warrant for calling it a genuine category. Verbatim precedent: 'The reports are remarkably consistent' (rejected, /meanings/soul-connection/) and 'The descriptions repeat themselves with striking consistency' (rejected, /meanings/twin-flame-vs-soulmate/). content-spec §0.5.
Three elements show up again and again. — Frequency frame over reports nobody has counted. Counting the page's own list is permitted; 'again and again' is not. Direct precedent: 'Three elements come up almost every time' (rejected, /meanings/soul-connection/) and 'Four kinds come up over and over' (rejected, /signs/spiritual-signs/).
The reports are too common and too similar to wave away. — Double prevalence claim ('too common', 'too similar') presented as the evidence base for the page's central assertion. No survey, no corpus study. content-spec §0.5.
That is not seriously disputed by anyone, including people who consider the romantic reading of it naive. — Universal negative about the state of debate, asserted as fact. Direct precedent: 'nobody in psychology uses it as a technical word' (rejected, /meanings/soul-tie/).
Attraction to a face, which registers faster than any deliberate judgment. Novelty, which makes the whole scene vivid and easy to remember. Alertness, the body responding as it does to anything that matters. And familiarity: ... which is one of the plainer explanations for the "I've met you before" sensation. — A list of psychological mechanisms asserted as established findings with no citation — 'studies show' in disguise. The face clause is a speed claim from the face-processing literature (unnamed, unsourced); the novelty clause is a memory-encoding finding. Verbatim precedent: 'Ordinary mechanisms produce recognition and intensity...' (rejected, /signs/twin-flame-signs/) and 'High-arousal situations produce fast attachment' (rejected, /meanings/soul-connection/).
certainty is what the mind produces when a strong signal has no competing detail to argue with / FAQ 3: 'Intensity without information feels like certainty, which is exactly what people describe afterwards.' — General law of cognition stated as fact, no source. Direct precedent: 'This is one of the more predictable things about human closeness' (rejected, /meanings/soul-connection/) and 'This is one of the most reliable things about human attention' (rejected, /signs/spiritual-signs/).
A stranger has not yet disappointed anyone. Early on, the gaps get filled with the most generous available material, and the person is briefly a very good version of themselves assembled by somebody else. — Idealisation mechanism asserted as a general fact about all people, no source. Same class as the memory-mechanism reject on /signs/spiritual-signs/ ('Memory keeps what fits the current ending ... Everyone does this, about everything').
Most disagreement about whether love at first sight is real is two people using those two definitions and assuming they are talking about the same subject. — Population-level prevalence claim ('most disagreement') stated as fact; no study of how these arguments actually run. Direct precedent: 'For most people asking this question, the answer is more than one' (rejected, /signs/more-than-one-soulmate/).
Consider what compatibility actually turns on: how someone behaves during a long illness, what they do with money, whether they can apologise ... whether their plans for the next ten years fit in the same house. — Two defects. (a) 'what compatibility actually turns on' asserts the determinants of relationship outcomes as established fact — precedent: 'Nearly everything that determines how a relationship goes lives in the unremarkable middle' (rejected, /signs/twin-flame-signs/). (b) 'the next ten years' is a concrete figure in a factual register on a number-free site — precedent: 'sustained for a decade', 'past the second year'.
Plenty of durable partnerships began with someone thinking their future partner was fine, and nothing more. — Prevalence claim ('plenty of durable partnerships') about relationship histories at large, no survey behind it. Direct precedent: 'That pattern is common in relationships generally' (rejected, /meanings/soulmate-vs-life-partner/).
Couples who met that way and are still together do not describe a first impression that carried them. They describe the same unglamorous work everyone else does. / FAQ 5: 'Couples who met that way and stayed together describe the same ordinary work as everyone else.' — Universal claim about what an entire population of couples does and does not say, asserted as fact and used as the page's answer to 'can it last'. No survey, no interview study. Same shape as the corpus-frequency rejects on /signs/twin-flame-signs/; the FAQ placement is the slot an AI Overview lifts verbatim, so it is the highest-risk instance on the page.
What happens after that is decided by ordinary compatibility, and the strong start neither helps nor hinders it. — A null-effect finding stated as fact — a statistic in words. Nobody has measured whether a dramatic beginning affects later outcomes; the clause also contradicts the two sentences before it, which credit the strong start with buying the second and third meeting. Direct precedent: 'strong feeling arrives just as readily in relationships that fail as in ones that last' (rejected, /signs/twin-flame-signs/).
Once a relationship becomes important, its beginning is revisited and edited. ... The result is a genre with a reliable shape. — Memory-reconstruction finding asserted as established fact, closed with a corpus-consistency verdict. Verbatim precedent: 'Memory keeps what fits the current ending and quietly drops what does not' (rejected, /signs/spiritual-signs/) and 'That is not dishonesty; it is how memory organises anything important' (rejected, /the-idea/do-soulmates-exist/). The surrounding observation — that accounts are told by people who already know how it ended — is fine.
a fair share of people searching this phrase want the film rather than the idea — Search-intent distribution stated as fact. Keyword-tool data is internal research, not a citable primary source. Direct precedent: 'Twin flame vs soulmate is one of the most searched comparisons' and 'though the search suggestions insist otherwise' (both rejected, /meanings/twin-flame-vs-soulmate/). content-spec §0.5.
the script confirms the first impression within ninety minutes — Concrete figure in a factual register on a number-free site. Precedent chain: 'four hours', 'six months of silence', 'a friend of twenty years', 'three times a month', 'within an hour of meeting'. content-spec §0.5.
FAQ 2: 'The experience is real and widely reported, and this page treats it as such.' — 'Widely reported' is a prevalence claim, placed in the FAQ slot most likely to be lifted verbatim by an AI Overview. Precedent: 'Mutual recognition is common in any strong mutual attraction' (rejected, /signs/twin-flame-signs/).
one of the few phrases in this subject that almost everyone claims to understand and almost nobody defines the same way (opening line) — Lowest-severity item on the page, recorded for consistency: 'almost everyone / almost nobody' is a rhetorical quantifier, but on a site where every other prevalence quantifier has been struck it reads as data in the first sentence a reader and an AI Overview see.
PASSED (recorded for precedent, no action required): Christopher Marlowe, Hero and Leander — 'who ever loved that loved not at first sight' — quoted by Shakespeare in As You Like It. — Not a statistic. content-spec §2 (/the-idea/where-the-idea-came-from/) explicitly permits naming work and author while forbidding invented years and translations; the draft supplies neither a year nor a translator. Attribution verified against primary texts: the line is in the First Sestiad of Hero and Leander, and Phebe quotes it in As You Like It 3.5 ('Dead shepherd, now I find thy saw of might, / Who ever loved that loved not at first sight?').
PASSED (recorded for precedent, no action required): 'a first impression carries no information about how two people handle money, illness or disagreement' (directAnswer) and 'The experience is not an illusion, and it is not a prediction.' — Reads like a §0 rule 4 verdict but is not one: the content-spec addendum for this very page mandates the angle — 'миттєве враження нічого не прогнозує про сумісність — але й не робить почуття несправжнім'. The page also declines to adjudicate destiny explicitly ('not something this page will confirm or deny on anyone's behalf').
PASSED (recorded for precedent, no action required): safety paragraph, gendered-advice FAQ, prediction ban. — Safety paragraph present and correctly shaped (fear/control/coercion → professional or a helpline in your country, no phone numbers, no diagnosis) even though §2 does not mandate one on this page. No astrology, numerology, predictions, twin-flame stages or reunion coaching. FAQ 6 refuses gendered rules instead of supplying 'how to make him fall in love', which is the §0-correct handling of a how-to-change-another-person query. Internal links: 4 in-site, zero cross-network. FAQ count 6, directAnswer 54 words — both inside §0 GEO limits.
Attachment forms quickly under stress and disclosure. Unpredictable rewards hold attention more tightly than steady ones. — Two psychological findings stated as established fact with no study, year or URL — 'studies show' without a study. Identical shape to the already-rejected 'High-arousal situations produce fast attachment' (/meanings/soul-connection/) and the soul-tie reinforcement paragraph. content-spec §0.5.
A conflict that keeps recurring usually recurs because nothing has changed between the two people, not because a cosmic curriculum is being delivered. — (a) Unsourced causal generalisation ('usually recurs because') about relationships in general. (b) Verdict ruling out a metaphysical cause — content-spec §0 HARD rule 4 ('Твердження про існування або неіснування долі. Ми описуємо позиції, не виносимо вирок'); same precedent as 'not fate distributing rare matches' (platonic-soulmate) and 'not a sign the bond is supernatural' (soul-tie). 'Cosmic curriculum' also clips the §0 tone rule — a believing reader must not feel mocked.
Strip the interpretation away and the reported experience is fairly uniform. / The first half is reported by a lot of people in similar words. — Frequency/uniformity claims about a body of first-person reports, with no survey or study behind them. Precedent: 'The reports are remarkably consistent' (soul-connection), 'The descriptions repeat themselves with striking consistency' (twin-flame-vs-soulmate).
FAQ 'How do karmic relationships usually end?': The most common description is an ending in stages rather than at once. — Ranking claim about how accounts are distributed — poll-shaped without a poll, in an FAQ slot AI Overviews lift verbatim. Precedent: 'People frequently place this at the top of their list'.
FAQ 'Can karmic relationships last?': ... the term is often applied retrospectively, after a relationship has already ended, which is why so many descriptions of it are short. — The causal tail asserts a measured property of a body of descriptions ('so many ... are short') and its cause; nothing supports it. The usage observation itself is fine.
karmic is the only one of the three that is normally used in the past tense — Absolute empirical claim about usage ('the only one', 'normally') with no corpus or source. Low severity, but it is a measurable statement dressed as a definition.

The rules

Named source in the same sentence as the number, with the year.
Primary source only — we link the origin, not an article about it.
Dated verification: every statistic carries its verification date on the statistics page.
Unverifiable → unpublished. No exceptions.