The 168 Nineteen-Letter Verses

Abstract

QMR-001 independently reproduces the previously reported observation that 168 Quranic verses contain exactly 19 letters when the Coran Codex RAW exact-character counting policy is applied to DATA-001.

The count of 168 is reproducible under the stated corpus and counting rules.

However, the 168 observations are not 168 independent textual events. Repeated verses and repeated phrases contribute materially to the total. After identical verse texts are deduplicated, the 168 qualifying verse occurrences correspond to 125 unique verse texts, leaving a duplicate excess of 43 occurrences.

One repeated verse text alone,

فبأي آلاء ربكما تكذبان

contains exactly 19 letters and occurs 31 times in Surah 55.

The result is therefore reproduced as a corpus measurement, but its statistical interpretation requires explicit treatment of repetition and dependence.

This study follows METHOD-001.

Research Question

The confirmatory question is:

Under the predefined RAW exact-character counting rules of DATA-001, are there exactly 168 Quranic verse occurrences containing 19 letters?

A secondary robustness question is:

Does the count remain the same under the predefined BODY analytical view?

Because the target value, corpus and primary counting rule were fixed before this reproduction, the core reproduction is classified as Confirmatory.

The additional dependence and duplicate analyses are methodological audits of the reproduced result.

Dataset & Counting Rules

• Dataset: DATA-001
• Methodology: METHOD-001
• Primary view: RAW
• Control view: BODY

The source corpus is:

Tanzil Quran Text — Simple Clean — Version 1.1

Under the RAW policy:

• all 6,236 verse identifiers are retained;
• words are not used for the primary measurement;
• letters are counted as exact non-space Arabic letter characters;
• Hamza and Alif forms are not merged;
• ة is not converted;
• ى is not converted;
• no spelling is altered;
• no Basmala is removed;
• no post-hoc normalization is introduced to obtain a target count.

The unit of observation is a verse occurrence identified by its surah and verse identifier.

Method

For every one of the 6,236 verse occurrences in DATA-001:

1. Read the RAW verse text.
2. Remove spaces for the purpose of the letter-length measurement.
3. Count the remaining Arabic letter characters exactly as encoded in the validated corpus.
4. Select verse occurrences whose exact length equals 19.
5. Count the selected verse occurrences.
6. Separately deduplicate identical selected verse texts to measure textual dependence.
7. Repeat the length calculation under BODY as a predefined robustness comparison.

No spelling rule was changed after observing the result.

No letter class was merged in order to reach 19.

Results

Primary RAW result

The RAW analysis returns:

The previously reported count of 168 is therefore reproduced under the stated RAW rules.

Repetition within the 168 occurrences

The most important dependence is the repeated verse:

فبأي آلاء ربكما تكذبان

This text contains exactly 19 letters under the RAW policy and occurs 31 times in Surah 55.

Thirty of those 31 occurrences are duplicate excess beyond the first occurrence.

Therefore, this single repeated text accounts for:

30 of the 43 duplicate-excess occurrences, approximately 69.8% of the duplicate excess within the 168-result set.

This does not invalidate the corpus count of 168.

It does invalidate treating all 168 observations as though they represented 168 independent textual events.

Concentration by surah

The result is also concentrated in a small number of surahs.

Surahs 55, 37 and 26 together contribute 77 of the 168 occurrences, approximately 45.8% of the complete RAW result.

This concentration is relevant when considering independence assumptions.

Length Distribution Context

The number 19 is not evaluated only as an isolated count.

Among RAW verse occurrences, selected nearby high-frequency exact lengths include:

After identical verse texts are deduplicated:

The ranking therefore changes materially after repeated texts are removed.

This illustrates why occurrence counts and independent textual observations must not be conflated.

BODY Robustness Check

BODY applies the predefined Basmala transformation documented in DATA-001.

Under BODY, the number of verse occurrences containing exactly 19 letters becomes:

171

rather than 168.

The three additional qualifying verses are:

• 48:1
• 107:1
• 110:1

Each becomes a 19-letter verse after removal of the predefined opening Basmala prefix.

Therefore:

The value 168 is consequently a reproducible property of the RAW view, not an invariant count across every documented analytical view.

Statistical Testing

This study distinguishes between two different propositions.

Proposition 1 — Corpus measurement

RAW contains exactly 168 verse occurrences of length 19.

This is a deterministic corpus measurement and is reproduced.

Proposition 2 — Statistical exceptionalness

The occurrence of 168 such verses is exceptionally improbable under an appropriate null model.

That proposition requires substantially more assumptions.

A valid significance analysis must define:

• the null model;
• the relevant unit of randomization;
• the dependence produced by repeated texts;
• the dependence produced by repeated phrases;
• the search space that led to examining length 19;
• any other tested verse lengths;
• any other numerical targets or formulas examined;
• any post-hoc selection.

The presence of 125 unique texts among 168 occurrences, including a single 19-letter verse repeated 31 times, means that a model treating all 168 qualifying occurrences as independent observations is inappropriate.

Naive independence-based significance is not valid.

QMR-001 therefore does not publish an extreme p-value based on the false assumption that every qualifying verse occurrence is statistically independent.

Robustness

The core claim survives exact reproduction under RAW:

168 RAW verse occurrences contain exactly 19 letters.

However, the broader interpretation is sensitive to analytical choices.

RAW versus BODY

The count changes from:

168 → 171

under the predefined BODY transformation.

Duplicate treatment

The RAW set changes from:

168 occurrences → 125 unique texts

when identical verse text is counted once.

Dependence

Repeated verses and clustered contributions from particular surahs reduce the plausibility of simple independence assumptions.

The observation is reproducible, while stronger claims about statistical exceptionalness require a model that explicitly handles these dependencies.

Prior-Art Investigation

The core observation is classified as:

Previously Known

Coran Codex does not claim discovery of the number 168.

The purpose of QMR-001 is independent reproduction and methodological audit.

The prior-art classification applies to the central claim that 168 verses have an exact 19-letter length under the relevant counting convention.

Coran Codex separately documents methodological observations such as:

• the distinction between 168 occurrences and 125 unique texts;
• the size of the duplicate excess;
• the effect of the 31 repeated instances of فبأي آلاء ربكما تكذبان;
• the RAW/BODY divergence.

A prior-art label is not a claim that every analytical detail in this audit is itself previously known.

Interpretation

Fact

Under DATA-001 RAW exact-character counting, 168 verse occurrences contain exactly 19 letters.

Statistical interpretation

The 168 occurrences include substantial textual repetition and dependence.

Only 125 unique verse texts are represented, and one qualifying verse text occurs 31 times.

Any probability model that assumes all 168 observations are independent is therefore methodologically inadequate.

Claim

The previously reported RAW count of 168 is independently reproduced.

Not claimed

No supernatural or divine conclusion follows from this result.

QMR-001 does not establish intentional numerical design, divine authorship, or a uniquely improbable structure.

Those propositions would require additional independently justified evidence beyond the reproduced corpus measurement.

Limitations

This audit has several explicit limitations.

First, reproducing a deterministic count does not automatically validate a probability claim attached to that count.

Second, the result depends on the documented corpus representation and counting policy.

Third, RAW and BODY do not produce the same count.

Fourth, repeated texts create dependence that must be accounted for before interpreting occurrence counts statistically.

Fifth, prior-art investigation can establish known antecedents when they are found, but cannot prove exhaustively that no earlier work exists.

Finally, this study does not attempt to test every possible theory involving the number 19.

Its scope is the specific 168-verse claim.

Verdict

Status: Reproduced

The central RAW measurement is reproduced:

168 verse occurrences contain exactly 19 letters.

The observation is real under the stated DATA-001 RAW rules.

At the same time:

• the 168 occurrences correspond to only 125 unique verse texts;
• there is a duplicate excess of 43;
• one 19-letter verse occurs 31 times;
• BODY produces 171, not 168;
• naive independence assumptions are inappropriate.

The scientifically supported conclusion is therefore narrower than the numerical observation itself.

The result is a reproduced corpus property with important dependence and robustness limitations.

Reproduction

A reproduction of QMR-001 must state at minimum:

1. Dataset: DATA-001
2. Analytical view: RAW
3. Counting unit: exact non-space Arabic letter characters
4. Verse universe: all 6,236 verse identifiers
5. Target length: 19
6. No silent normalization
7. Treatment of repeated verse texts
8. Whether results refer to verse occurrences or unique verse texts

The expected baseline outputs are:

Any reproduction producing different values should first verify corpus identity, Unicode handling, space removal and Basmala policy before drawing a substantive conclusion.

Revision History

The structured revision history for this record is displayed with the article metadata.

Future material corrections to the corpus measurement, statistical interpretation, prior-art classification or scientific status will remain documented rather than silently replacing the earlier record.