76
of 167 current books are Urdu or Sindhi medium — theme-grade only
Overview / What is missing
Coverage and its limits
What is missing
Every figure on this page is computed from the catalogue at build time, so it cannot drift from what the study actually holds. A reference work that only lists what it has is telling you half of what you need to judge it.
There are four different kinds of absence here and they carry very different weight. A book that could not be found means a claim about that board's teaching may simply be uninformed. A book that was found but cannot be read at quotation grade means the study can describe its structure but must never quote it — and must say not established rather than absent about anything it fails to find there. A book that was read but not scored on an axis means a board mean rests on fewer books than the catalogue implies. Material deliberately excluded is not a gap at all, but it is listed so that nobody mistakes a boundary for an oversight.
1. Books that could not be found
Against the declared frame — 5 boards, classes 6–10, the subjects that carry social content — these cells are empty. Every one was searched for on the board's own site where one exists, and across the public mirrors credited in each record.
| Board | Missing | Count |
|---|---|---|
| Balochistan | Urdu 10 · Biology 10 · Physics 10 | 3 |
| Federal | English 6 · Urdu 6 · Islamiyat 6 · General Science 6 · English 7 · Urdu 7 · Islamiyat 7 · General Science 7 · English 8 · Urdu 8 · Islamiyat 8 · General Science 8 · Islamiyat 9 | 13 |
What this costs the study. Balochistan carries every one of these absences, and they are not evenly spread: its class-8 language and religious instruction and its class-9 English and Islamiyat are all missing. Any statement about how a Balochistan pupil's English or Islamiyat teaching progresses from class 6 to class 10 is therefore built on a broken sequence, and this site should not make one. Balochistan's board portal is the only working official portal of the four, which makes these absences more likely to be genuine non-publication than acquisition failure — but that is an inference, not a verified fact, and it has not been confirmed with the board.
A category that looks like a gap and is not
The boards do not agree on what the social-studies family is, so a naive frame reports twelve missing books that were never published. Sindh issues one integrated Social Studies volume per class; the other three split the same ground into separate History and Geography books.
| Board | Publishes at classes 6–8 |
|---|---|
| Punjab | Geography + History |
| Sindh | Social Studies |
| Khyber Pakhtunkhwa | Geography + History |
| Balochistan | Geography + History |
| Federal | — |
This is why the comparison pages compare the social-studies family rather than matching subject names, and why a cross-board table of "Social Studies" scores would be meaningless.
2. Books held but not readable at quotation grade
This is the study's largest limitation by a wide margin, and it is structural rather than fixable by more searching.
15
carry a publisher's own text layer, the only exact-quotation tier
46%
of the current corpus can never be quoted at sentence level
| Board | Urdu/Sindhi-medium | Of total held |
|---|---|---|
| Punjab | 23 | 45 |
| Sindh | 14 | 35 |
| Khyber Pakhtunkhwa | 18 | 36 |
| Balochistan | 16 | 36 |
| Federal | 5 | 15 |
The subjects worst affected are the ones that matter most to this study's questions: Biology · Chemistry · General Science · Geography · History · Islamiyat · Mutalea Quran · Pakistan Studies · Social Studies · Urdu. Urdu and Islamiyat are almost entirely unquotable across every board, and those are precisely the books that carry literary canon and religious instruction.
What follows, and it is a hard rule. Nothing on this site is ever quoted from Urdu or Sindhi material. Those extractions support statements about structure, chapter coverage and proper nouns — which survive degraded Nastaliq OCR — and nothing else. Where a probe of Urdu material finds nothing, this study says not established, never absent, because the method cannot tell those apart. Any finding that depends on an Urdu absence is the weakest kind of claim this project makes and is marked as such.
Most Pakistani pupils are taught in Urdu medium. The single most useful thing anyone could do to extend this study is a Nastaliq-tuned OCR pass over these 76 books. Until then, the honest description is that this study reads the English-medium curriculum well and the Urdu-medium curriculum only in outline. Tariq Rahman's study of world view in Urdu-medium schools, catalogued in the library, covers the ground this corpus cannot.
3. Books read but not scored on every axis
Coverage is uneven by design — each lens scored the books where its question is live, so a chemistry volume may carry three axis scores and a Pakistan Studies volume seven. The consequence is that two board means on this site do not rest on the same books, and a mean over a handful of volumes is not a board figure.
| Axis | Books scored | Of 167 | |
|---|---|---|---|
| A1 Religious saturation | 119 | 71% | |
| A2 Civic thickness | 49 | 29% | |
| A3 Inquiry vs received truth | 55 | 33% | |
| A4 National-narrative closure | 18 | 11% | |
| A5 Militarism | 71 | 43% | |
| A6 Othering | 56 | 34% | |
| A7 Gender conservatism | 113 | 68% |
A4, national-narrative closure, is the weakest row in the whole study and must not be quoted as a board figure. It can only be scored where a book carries the national narrative, which is a handful of volumes per board — for one board it rests on a single book. The axis is real; its denominators are too small to rank provinces on. Every mean on this site is printed with its n, and thin ones are marked in red.
4. Material held but deliberately not examined
These are boundaries, not oversights. They are listed so that the difference is visible.
| Historical editions, catalogued but never scored | 43 |
|---|---|
| — of those, outside classes 6–10 | 14 |
| — of those, with no recoverable printing year | 18 |
| — of those, with no source URL recorded | 25 |
- Historical editions are never scored. The rubric was built for the current curriculum generations; stretching it across a fifty-year gap would compare instruments rather than books. See the historical layer.
- The historical layer is Punjab only. No comparable run was found for Sindh, Khyber Pakhtunkhwa or Balochistan, so nothing here supports a cross-province claim about the past.
- 25 historical records carry no source URL. The files are held with full bibliographic records, but the retrieval URLs were not captured at acquisition. That is a real provenance failure and it is not recoverable after the fact.
- Madrassas, private schools and the Aga Khan board are entirely out of scope. This study covers four provincial government boards. A large share of Pakistani pupils are taught from none of these books.
- Classroom practice, actual examination papers and what pupils take away are all unobserved. The corpus shows what is printed, sequenced and examinable. It does not show what is taught aloud or what anyone concludes.
5. Provenance that is thinner than it should be
- No formal web-archive captures. Source links are recorded with a local retrieval date, not a Wayback or Perma.cc snapshot. Boards and mirrors reorganise; some of these links will rot, and there is no archived copy behind them.
- Checksums exist, but they prove less than they look like they prove. Every record carries the SHA-256 of the file the findings were computed from, so anyone can verify they hold identical bytes to ours. What it cannot do is prove that file matches what the source URL serves today — boards reprint, and a changed checksum tells you something changed without telling you what.
- Mirror-sourced files carry no imprint guarantee. Neither Punjab nor KP serves textbook PDFs from an official domain, so most of those books come from public mirrors. Edition years are as stated by the mirror or by the book's own imprint page, and several volumes have no recoverable imprint at all.
6. The books themselves — held, and now published
This project holds every book it catalogues — 167 PDFs, about 6.7 GB — and, as of 23 August 2026, serves all of them for download from each book's own record.
The reason is preservation. All four boards distribute these textbooks free and publish them online, and free access to school textbooks is federal policy. But source links rot and editions are withdrawn: several in this catalogue already have, Khyber Pakhtunkhwa serves no textbook PDFs from an official domain at all, and Balochistan's are pilot editions printed for a single academic year. A catalogue whose sources have disappeared is a catalogue that can no longer be checked — and being checkable is the only thing this project claims.
What this is not. It is not a licence. No board has granted written permission, and the boards' printed rights reservations stand — an explicit reservation appears in 21 Punjab books, 19 Sindh, 15 Balochistan and 8 Khyber Pakhtunkhwa. Copyright remains with each board. Files are served unmodified, credited to their board, without charge and with no advertising against them, and every record carries the SHA-256 of the file served so a reader can prove the copy is byte-identical to the one the findings were computed from.
Takedown. Any board, author or rights holder may have any file removed, for any reason, without giving one. It will be removed the same day, without argument, and the removal will be recorded publicly — what went, who asked, and when. A silent deletion would leave every citation to it broken and tell the reader nothing.
The scholarly library is not covered by this and stays catalogue-only. The argument above is specifically about school textbooks distributed free by their publishers as a matter of policy; it does not extend to commercially and institutionally published scholarship. The SDPI reports state in print that no part may be reproduced without written permission. One exception is hosted in full — Nosheen Ali's 2008 article, which carries a Creative Commons licence — and it is marked as such in the library.
If you can fill one of these gaps
Missing books are the most useful thing anyone can contribute, especially the eight Balochistan volumes above and any Sindh, KP or Balochistan edition printed before 2006. A Nastaliq-tuned OCR pass over the 76 Urdu-medium books would change more of this study than any other single intervention. Both routes are the same as the correction route in the standards.