Paper

Abstract

Paper 1 — The Record

This paper measures what changed in the establishment of eligibility to benefit from waqf (a legal-institutional arrangement in which property is dedicated to a designated purpose under a founder’s stipulations and managed by an administrator) in Egypt, distinguishing it from what remained constant. Its central argument is that the historical transformation occurred in the “reference point for framing”, rather than the “evidentiary instrument”: the answer’s reference point shifted from the founder’s stipulation and jurisprudence with named sources to the provisions of codified law. The reversal was sharp in the year Law 48 of 1946 on Waqf Provisions was enacted, and again during the decade of liquidation after 1952.

The document embodying the right—the waqf deed, its official copy and its sharia attestations—did not retreat before the law; its invocation instead peaked during that same decade of liquidation. Measurement draws on an original corpus with a frozen release, comprising (396) three hundred and ninety-six waqf entitlement texts covering (394) three hundred and ninety-four distinct instances with stable identifiers. These were manually coded following a complete automated enumeration of the published collection of Dār al-Iftāʾ fatwās. Three counters were applied to the corpus: (mode of inference, invocation of documents controlled for text length, and who the questioners were).

The comprehensive competing explanation was tested along four axes. It explained none of the gap in the central trend; the gap instead widened after normalisation. Findings after 1959 describe a thin sample. Whether the beneficiary eligibility register moved into a file controlled by the apparatus remains a question requiring administrative material that has not been fully assembled.

The replacement test was conducted using the German purpose foundation under the German Civil Code (BGB). Neither claim collapsed; the comparator’s provisions instead reproduced the rule underpinning them. Common ground with the Romano-Germanic family is established, while a system from a third family differs over the rank of the founder’s stipulation. What is measured is therefore waqf’s position among three families, rather than its uniqueness in relation to them.

Keywords: waqf; entitlement; Dār al-Iftāʾ; evidentiary authority; registration; establishment of eligibility; founder’s stipulation; Egypt.

The Paper’s Statement

First, type of contribution: a completed test of limited scope

This paper tests precisely two claims: (the reversal of the reference point for framing within the answer of the muftī (jurist who issues fatwās), and the document’s continuing centrality to the claim after that reversal). The test covered the entire corpus using a counter with a disclosed protocol. It was repeated with a second, independent dating key, then assessed against a comprehensive competing explanation tested along four axes. Whether the eligibility register moved into a file controlled by the apparatus has not been tested, because the requisite material has not been fully assembled; section 7 sets out this limitation. Accordingly, statements here about the documentary system and the reversal of the reference point are supported by an executed count, whereas statements about the administrative system describe the limits of the evidence rather than a measured result.

Second, each claim’s evidentiary tier is declared in advance

This paper follows the shared inference protocol of the “Architecture of Entitlement to Charitable Benefit” programme. The protocol distinguishes three evidentiary tiers and requires the wording of each claim to match its tier1:

1- The first tier (what the fatwās explicitly establish in their wording): the document’s place in the texts of claims, the muftī’s reference point in each individual answer, the beneficiary’s choice at the 1916 juncture, susceptibility to attachment and transfer, and the five instances of registration.

2- The second tier (what the corpus makes likely after controls and falsification of the competing explanation): the trend in the shift of the reference point for framing, the trend in the invocation of documents, and who the questioners were.

3- The third tier (what remains a hypothesis requiring administrative archives or judicial evidence): the movement of the beneficiary eligibility register into a file controlled by the apparatus, where the burden of correction falls, and every statement concerning the period after 1959 (language of certainty is not used at the second tier, nor language of findings at the third).

Third, the selection rule and the direction of its bias

The corpus is a complete enumeration, not a sample. Yet a complete enumeration of a published collection involves selection at a higher level, with two biases whose direction is known. First, the record preserves only those whose voices reached Dār al-Iftāʾ. It therefore omits blocked routes and silent groups, and gives greater prominence to claimants possessing documents than to those without them. Second, an editor selected the fatwās for publication: an editorial layer stands between the muftī’s answer and the compiler’s title and principles. Both biases operate in the same direction: they enhance the image of the old documentary system and magnify its difference from what followed. The corpus therefore shows that the instrument was available to those who pursued the route, rather than how widely it was available among entitlement-holders (mustaḥiqq). Any reading that treats this paper’s percentages as population rates exceeds what the source can support.

Fourth, the conditions for falsification are settled

The first falsifier: that the rise in inference based on law results from the automated dating key or from a single muftī. Testing used two independent keys, a breakdown by muftī, and a control for the same muftī serving across both decades, without overturning the finding. Section 8 sets out the details.

The second falsifier: that invocation of documents declines in proportion to the rise in inference based on law—the replacement hypothesis. This was tested by lexical counting across all the corpus texts using two measures: a binary measure at record level and a measure controlled for text length. Both rejected it. (The control for the composition of issues could not be implemented using the issue-type field because that field was associated with the temporal batch. It was subsequently implemented through three routes independent of that field; section 10 presents their results.)

Fifth, the strongest competing explanation

The strongest objection is that the three counters measure editorial conventions in the published collection rather than actual fatwā practice. Its strong form was addressed through four procedures set out in section 10. The most rigorous separated the count of the opening formula from the count of document mentions and measured the two trends independently. The trends differed in both their peaks and their direction. The competing explanation was therefore rejected in its strong form, while its weak form remained as specified in 10.2: rejected as an explanation of direction, still viable as an explanation of magnitude.

Sixth, what the measurement instrument has not achieved

The manual coding has not undergone an inter-coder reliability test, nor recoding blind to date and muftī. Instead, a convergence check was conducted against an independent automated instrument for the central variable alone; section 8 presents its results. Coding reliability between two readers therefore remains pending, and everything based on the manual coding must be read subject to this limitation.

Seventh, declaration of dependence

This paper and “The Entry Standard: Measured Content, a Hypothesised Source” are two analyses dependent on a single evidentiary base, rather than two independent studies whose findings agree. They attribute the same transformation to the same texts in the Waqf Entitlement Corpus, release WEC-432/1.0. The second has no quantitative coding of the institutional-source axis, so its quantitative support for the claim that the source shifted is the counter for the mode of inference implemented here. It follows that the test of the shared source axis occurs only in the first paper of this series, while the second is confined to the structure and content of the standard and the mechanisms of its abolition, translation and recombination. Agreement between the two papers consequently does not constitute independent corroboration: the evidence, the reader and the measurement on the shared axis are all the same. Independent corroboration requires a source outside this corpus: the incoming register of Dār al-Iftāʾ al-Miṣriyya (the Egyptian fatwā authority), the Ministry of Awqāf’s administrative records or comparison with the printed editions2.

Eighth, comparison against the legal comparator’s provisions

The paper’s quotations were checked against the law in force. The rank of the provision was corrected, the evidence furnished by numbering was confined to its proper scope, and § (82a) was included for comparison rather than inference. The documented record appears at the end of Appendix (8).

Footnotes

  1. The Common Appendix to the “Architecture of Entitlement to Charitable Benefit” programme: the inference protocol, variable matrix and declaration of dependence. It supplies the three tiers, the verbatim declaration of dependence and this paper’s row in the matrix. The appendix is reproduced in full in all four papers, after the list of sources and before each paper’s own methodological appendix. ↩

  2. See note 1. ↩

Methodological Disclosure

This paper was prepared with the assistance of “al-Warrāq” in the automated enumeration of the collection and in retrieving and comparing texts. Al-Warrāq was also used to implement the counters in sections 4 and 8 and the test of the comprehensive competing explanation. The author alone is responsible for the design, coding rules and judgements on each text. All texts were coded through manual reading. Two outstanding tasks are declared: a test of coding reliability between two readers and comparison of a sample against the printed editions. Every citation identifies a verifiable location.

1 · Introduction: From al-Mujāwirīn Cemetery to the University of Malaya

In Rajab (1313 AH / 1895), the Grand Muftī of Egypt, Shaykh Ḥassūna al-Nawāwī, received a communication from Cairo Governorate asking what established, under sharia, that cemetery land was dedicated as waqf for burying the dead. The muftī answered that Cairo’s cemetery was waqf under sharia and could not be used for purposes other than burial, grounding his answer in juristic texts and the account of al-Layth ibn Saʿd3.

Eighty-seven Hijri years later (Shaʿbān 1400 AH / June 1980), Muftī Shaykh Jād al-Ḥaqq ʿAlī Jād al-Ḥaqq received a letter from a researcher at the University of Malaya asking about the basis for abolishing family waqf. The muftī answered by setting out the schools’ positions on the binding force of waqf, then stated that the ruler’s actions are governed by benefit and that the Law Abolishing Waqf for Non-Charitable Purposes fell within that principle4.

Every element changes between these two scenes. In the first, the questioner is a state apparatus seeking confirmation concerning an existing waqf; in the last, a foreign university enquires into the basis for legislative abolition. The first reference point consists of juristic texts and an account of the conquest; the last consists of the ruler’s authority and a numbered statute. Between the two scenes lies the corpus’s entire span, from (1894) to (1979). The final scene must nevertheless be read as an individual instance, rather than an endpoint in a measured series. The methodological reason is the corpus’s weak representation of the 1960s and 1970s, discussed in section 10.

This paper asks what happened between these scenes to three powers, rather than one. The first is the power to define entitlement status: who enters the sphere of benefit, and by what standard. The second is the power to preserve and amend its record: which document embodies the right, who creates and preserves it, and who has the power to alter it. The third concerns the burden of discovering and correcting error: on whom it falls when the gatekeeper errs or distribution is unjust. This threefold distinction governs the analysis throughout the paper and provides the basis for its concluding table.

The paper does not assume in advance that the answer is a transition from “right” to “charity”, or from “orality” to “documentation”; waqf had been a rigorous system of documentation long before this entire period. It measures what its material permits it to measure and explicitly declares what that material cannot establish.

Its material is an original corpus: the record of a single framing authority spanning the entire transformation. Among the studies covered by our survey, we have found none that uses this type of source in this way. Before drawing inferences from it, we must explain what it is.

Footnotes

  1. Fatāwā Dār al-Iftāʾ al-Miṣriyya [Fatwās of the Egyptian Dār al-Iftāʾ] (compiled by al-Maktaba al-Shāmila from Dār al-Iftāʾ’s collection), “Waqf al-jabbānāt fī Miṣr” [The Waqf Status of Cemeteries in Egypt], fatwā of Ḥassūna al-Nawāwī, Rajab 1313 AH, location 2475. The instance carries its permanent identifier in the paper’s corpus (appendix). ↩

  2. The same source, “Ilghāʾ al-waqf al-ahlī wa-sanaduhu” [The Abolition of Family Waqf and Its Basis], fatwā of Jād al-Ḥaqq ʿAlī Jād al-Ḥaqq, Shaʿbān 1400 AH / 29 June 1980, location 3266. The questioner was a researcher at the University of Malaya; this is one of the instances involving an external questioner counted in section 9. The interval between the headings cited in footnotes 1 and 2 is eighty-seven Hijri years, corresponding to approximately eighty-five Gregorian years. The Hijri interval is given because that calendar is common to both headings. ↩

2 · Concept and Method: A Corpus of Claims and Framing

The unit of analysis and the institutional distinction between actors: the central unit of analysis in this study is the “instance of entitlement determination”, rather than the isolated fatwā or the unactivated legal provision. Such an instance is a situation in which a person’s or body’s inclusion in, or exclusion from, the sphere of waqf benefit is determined through a standard, an evidentiary instrument, a gatekeeper, a decision and a route of review.

The texts are coded into fields that strictly distinguish four authorities that must not be conflated: the gatekeeper (the nāẓir (trustee-administrator) or implementing official), the substantiation and registration authority (the Land Registry or court registers), the legal framing authority (the muftī and Dār al-Iftāʾ), and the adjudication and enforcement authority (the administrative courts or national courts). The muftī, whose record constitutes our corpus, is primarily a framing authority, rather than a gatekeeper or an enforcement judge. This distinction sets strict limits on what may be inferred from his fatwās.

Documentation of the corpus and filtering through the five layers: the corpus is cited through a single frozen release, designated WEC-432/1.0, frozen on 28 July 2026 with (396) three hundred and ninety-six accepted texts and recorded hashes. Every aggregate figure in this paper is calculated solely from that release, and every citation uses a stable identifier in the form F432-NNNNN, together with the text’s location, muftī and date. Neither of the two preceding intermediate layers is cited. The record of the corpus selection funnel that produced it, including its exclusion table and declared unknowns, is cited as a separate document5.

The corpus was constructed through five successive layers, each with its own count and unit: a crawl of four thousand nine hundred and thirty pages; segmentation into ten thousand and thirty-five fatwās; purely lexical screening yielding five hundred and forty candidates; a tightened inclusion rule retaining four hundred and five records; and manual coding retaining three hundred and ninety-six texts. Appendix 1 details the layers and the rules for entry into each67.

LayerSurvey
First layerAutomated survey and crawl of (4,930) pages
Second layerSegmentation of the material into (10,035) indexed fatwās
Third layerPurely lexical screening yielding (540) candidate fatwās
Fourth layerApplication of a tightened inclusion rule retaining (405) records
Fifth layerRigorous manual coding retaining (396) accepted texts (covering 394 distinct instances)

Table summarising the construction and layers of the corpus

Two limitations of this construction must be recorded rather than concealed. The first concerns (completeness of documentation): three aspects are incompletely documented. The transition from the third to the fourth layer cannot be rerun automatically. The judgements on the sixty records in the precision-test sample have not been preserved. Measurement of misses among excluded records has not been established: the known lower bound is approximately one and a half per cent (1.5%), and the likely lower bound approximately three (3%); this is a lower bound, not an estimate. The second concerns (units of description): the three hundred and ninety-six are units of description, rather than distinct instances; the number known with certainty is four hundred minus six. Appendix 1 details both limitations, their figures and their locations, while section 12 summarises them8.

The texts are assessed using the “five-tier source scale”, which prevents a text from being assigned an evidentiary status beyond its nature. Its tiers range from a general juristic rule to a nearly complete instance of determination; Appendix 1 defines the limits of each. All texts were manually coded on this scale after testing showed that automated coding could not handle this particular field.

The results determine the nature of the source: three hundred and forty-six instances are claims or documentary enquiries, amounting to eighty-seven per cent (87%). Of these, one hundred and fifty-one explicitly mention a document in their text. Thirty instances concern encounters with procedures, fourteen are procedural instances, five are nearly complete, and one is a general rule9.

These findings can be summarised in the following lists:

InstancePercentageDescription
(346) instances(87.4%)Formal claims or documentary enquiries (including 151 fatwās explicitly mentioning a document present in the text)
(30) instances(7.6%)Direct encounters with procedures
(14) instances(3.5%)Institutional requests for fatwās concerning procedures
(5) instances(1.3%)Nearly complete instances of adjudication
One instance×A general juristic rule

The result: the corpus provides a strong measure of four things: (1) how the claim was formulated, (2) which document was presented, (3) who requested framing, and (4) which rule the muftī relied upon. By itself, it does not measure success rates in practice, the regularity of financial disbursement or the effectiveness of judicial appeals. This defines the source rather than apologising for it: it is a corpus of claims and framing, and inferences are drawn from it in that capacity.

Entitlement-holders are divided into five categories, and this field governs the limits of generalisation. Named or descendant entitlement-holders constitute the corpus’s principal constituency, accounting for one hundred and ninety-four instances: forty-nine per cent (49%) of the entire corpus and sixty-six (66%) of the instances whose classification was resolved. Next are descriptively defined and identifiable beneficiaries in fifty-five instances, including students, Qurʾān readers and imams. Institutional bodies account for twenty-seven, local categories for thirteen, and the open public for seven; an individual in this last category owns nothing before distribution. One hundred instances remain unclassified; Appendix 1 sets out the rule for aggregating tags10. No judgement about the last category may be inferred from the legal position of the first. This is the declared limit of the categories.

Entitlement-holder categoryInstances observed in the corpusPercentage of the entire corpusPercentage of instances whose classification was resolved
Named entitlement-holder or descendant entitlement-holder194 instances49.0%65.5%
Descriptively defined and identifiable beneficiary55 instances13.9%18.6%
Institutional body27 instances6.8%9.1%
Local category13 instances3.3%4.4%
Open charitable public7 instances1.8%2.4%
Classification unresolved100 instances25.3%×

Table of the fivefold classification of entitlement-holders

The matrix distinguishing this paper from its three companion papers appears in the Common Appendix; this paper’s row is reproduced in Appendix 5. Its dependent variable is the mode of inference in the muftī’s answer by decade. Its two falsifiers have been tested and rejected. Its specific evidence consists of manual coding of three hundred and ninety-six texts, the counter for invocation of documents controlled for length, the questioner field, a case record for a single waqf, and instances of absence and registration11.

Footnotes

  1. The corpus selection funnel record and frozen release WEC-432/1.0. It records: the counts for the five layers (4,930 pages, then 10,035 fatwās, then 540 lexical candidates, then 405 after tightening, then 396 after manual coding); the record that the crawl ended by exhausting the collection rather than being cut short; the inability to rerun the tightened rule automatically because its script was lost; the limitation of the precision test, since 44 of the sixty records remained and 16 were excluded, while the judgements on them were not preserved; the identifiers of the nine late exclusions and their recorded reasons, six of whose titles contain the word “mawqif” [position]; rejection of the initial miss-rate result because the identifiers of its fifteen-record sample were not preserved, while reading a new sample at the time of freezing yielded two confirmed misses—F432-05731 on substitution of waqf assets in 1894 and F432-05987 on the sale of entitlement in 1903—two likely misses, F432-05939 and F432-05537, and one borderline instance, F432-01673, making the known lower bound 2 out of 135 and the likely lower bound 4 out of 135; and the two documented duplicate pairs with their identifiers, F432-06347 → F432-05671 and F432-06495 → F432-05673, with no identified third pair. Nothing from the review of misses was incorporated into the release: freezing preserves the accepted corpus unchanged. ↩

  2. The enumeration protocol and its outputs: the enumeration frame; the full inclusion and exclusion rule; deduplication; classifications; temporal distribution; and the manual precision test on sixty records and its effect on tightening the rule. These are documented in the enumeration report and corpus files (appendix). All must be read subject to the declared unknowns in the corpus selection funnel record, particularly the inability to rerun the tightened rule automatically, the loss of the judgements on the precision-test sample, and rejection of the initial miss-rate result. ↩

  3. See note 5. ↩

  4. See note 5. ↩

  5. Results of manual coding calculated from frozen release WEC-432/1.0. The third layer comprised four hundred and five records. Nine unrelated to waqf entitlement were excluded through an explicit tag in the tier column, leaving three hundred and ninety-six accepted texts, of which three hundred and ninety carry a year under the first dating key. The distribution across tiers is: 346 claims or documentary enquiries, including 151 with a document explicitly mentioned in the text; 30 encounters with procedures; 14 procedural instances; 5 nearly complete instances; and one general rule. The distribution of entitlement-holders, after aggregating detailed tags into their parent categories, is: 194 named or descendant entitlement-holders; 55 descriptively defined and identifiable beneficiaries (48 with basic tags and 7 with detailed tags); 27 institutional bodies (23 with basic tags and 4 with detailed tags); 13 local categories (11 with basic tags, one tagged by community affiliation, and one combined with the named entitlement-holder category); 7 instances involving the open public; and 100 unresolved classifications. The release preserves the raw tags; aggregation is this paper’s decision, declared where it is applied. ↩

  6. See note 9. ↩

  7. See note 1. ↩

3 · The Juristic Apparatus of Entitlement from Within: Text, Testimony and the Written Record

Before discussing any subsequent institutional transformation, the juristic starting point must be reconstructed from within, rather than from the accumulated judgements and legislation that followed it. Waqf entitlement was a system of legal positions with a full complement of constituent elements and mechanisms:

First: the sharia standard and the principles of deduction from the founder’s text

The governing standard is the founder’s text and its conditions. The governing maxim in Ḥanafī jurisprudence is that “the founder’s stipulation is like the Lawgiver’s text”. Its carefully qualified formulation is Ibn Nujaym’s statement in “al-Ashbāh wa-l-naẓāʾir [Analogues and Parallels]”: “The founder’s stipulation must be followed, for they say: the founder’s stipulation is like the Lawgiver’s text, that is, in the obligation to act upon it, and in implied meaning and signification”12.

The statement then became part of the base text under discussion: al-Ḥaṣkafī incorporated it into “al-Durr al-mukhtār [The Choice Pearl]”, and Ibn ʿĀbidīn commented on it in “Radd al-muḥtār [The Reply of the Perplexed]”13. The analogy concerns the method of deduction from wording and signification through the principles of jurisprudence, rather than the text’s legislative rank.

These are not two mutually supporting texts, but a single original text on which a later scholar commented. Ibn ʿĀbidīn added qualifications in two respects. He preferred “understanding and signification” to “implied meaning and signification”, “because implied meaning is not recognised by us in texts”. He also observed of the clause “the obligation to act upon it” that “this contradicts what we have just transmitted”, before reconciling the two positions14. Reading the deed is therefore a full exercise in the principles of jurisprudence, distinguishing whether the founder’s wording is unequivocal, manifest, equivocal or indeterminate. Ibn ʿĀbidīn transmitted this classification from Qāsim ibn Quṭlūbughā; he did not originate it15. The corpus shows this in practice: when a nāẓir confined waqf revenue to the upper tier of descendants, believing that it absolutely excluded the lower tier, the muftī determined that the general wording of the stipulation required equal treatment of the two tiers16.

Second: time and limitation—a procedural bar to hearing an action, rather than extinguishment of the right

The effect of time on a financial right is governed by rules on the “hearing of an action”, rather than the “extinguishment of the substantive right”. Shaykh Muḥammad ʿAbduh determined that an action concerning waqf entitlement could not be heard after fifteen years had elapsed without an excuse, and that silence throughout this period by the agent of an absent entitlement-holder likewise barred the action from being heard17. In an action brought by two nāẓirs against the Ministry of Finance, Shaykh Muḥammad Bakhīt quoted Article (587) of Qānūn al-ʿadl wa-l-inṣāf [Code of Justice and Equity], drawn from Tanqīḥ al-Ḥāmidiyya [Revision of al-Ḥāmidiyya]: “A right is not extinguished by the passage of time”. He then explained that an action concerning ownership could not be heard after fifteen years, or one concerning inheritance or waqf after thirty-three years, subject to the prescribed conditions18. In the juristic apparatus, limitation is thus a procedural bar to hearing an action, while the right remains established and binding in conscience. By comparison, registration in the land registration system is constitutive rather than declaratory.

Third: extinguishment, transfer and the rules governing litigation

Extinguishment and transfer are governed by the text, rather than unrestricted executive discretion. A founder stipulated the exclusion of anyone who acknowledged that a person outside the waqf’s beneficiaries was entitled to any of its revenue. The muftī distinguished the acknowledgement covered by the stipulation from assignment to another entitlement-holder for a specified period, ruling that acknowledgement and assignment were different and that no exclusion followed19. The nāẓir’s accountability takes the form of an action for an account and litigation: an entitlement-holder may sue him to establish substantive entitlement and to recover revenue arrears. Holding him to account for breach of trust and seeking his removal, however, require the judge’s authorisation and appointment of the entitlement-holder as a temporary mutawallī (waqf administrator). These procedural forms were codified in Ḥanafī jurisprudence20.

3.1 The Document’s Place in the Juristic Evidentiary Apparatus

A point commonly misrepresented in accounts of the earlier juristic system requires clarification here. It establishes a governing constraint on this paper’s entire thesis:

Jurisprudence recognises testimony as one means of proving waqf status; secondary testimony, women’s testimony alongside men’s, and testimony based on widespread repute and communal knowledge (tasāmuʿ) are admissible for this purpose. There follows, however, a decisive qualification concerning the basis of proof, expressed in the original passage examined as follows: “A documentary instrument, however, cannot serve as evidence, because one handwriting resembles another.” Ibn ʿĀbidīn transmits this passage from al-Khāniyya and al-Isʿāf rather than originating it himself21. Under the school’s basic evidentiary rule, therefore, a written document alone has no independent evidentiary authority: testimony is the primary means of proof, and writing serves merely as an evidentiary aid to recalling it.

The school nevertheless provides an exceptional ruling for cases in which oral proof is no longer available, allowing procedural recourse to the judicial register. Its source is al-Fatāwā al-Khayriyya, quoted verbatim by Ibn ʿĀbidīn: “If the waqf has a deed in the judges’ archive, customarily known among us as the register, and it is in their custody, its contents shall be followed by way of istiḥsān (juristic preference) if those concerned dispute it.” If the waqf has no deed registered in the archive, consideration is given to “what was established about its circumstances in earlier times, namely how its administrators used to act”; thereafter, “whoever establishes a right by proof shall have judgement given in their favour for that right”. The hierarchy thus has three levels: the register, the established practice of the administrators, and proof22.

Taken together, these two juristic passages entail three implications that delimit this paper. First, the judicial register serves as an evidentiary repository of rights within the juristic tradition itself; invoking it is no modern innovation. This must be qualified by the historical layer to which its source belongs: Radd al-muḥtār attributes the rule to al-Fatāwā al-Khayriyya, making it a later fatwā within the school, rather than a text from its earliest layer; it is not attributable to al-Khaṣṣāf. The identity and dates of the author of al-Khayriyya have not been examined in this paper. Second, the register has the standing of istiḥsān when the primary means of proof is unavailable, rather than that of the primary means itself. Third, and most importantly, the system this paper describes at its starting point is not jurisprudence in the abstract. It is the practice of Egypt’s sharia courts, which elevated the registered document above the position it occupies in juristic texts. The comparison is therefore between two successive administrative practices, not between pure jurisprudence and modern administration. This deliberately narrows the thesis and answers the objection that it compares an ideal with a reality.

Contemporary evidence from outside the corpus supports this third implication. In 1902, Shaykh Muḥammad ʿAbduh wrote that the proposition that the founder’s stipulation is like the Lawgiver’s text “has governed the practice of the courts from long ago to the present day: they adhere to words written in ‘waqf deeds’ which the founder may not even have understood, but which the scribe wrote”. Adherence to the wording of the document was thus described in 1902 as the practice of the courts, not merely a requirement of juristic texts. The author of this observation was himself one of the leading muftīs whose fatwās appear in this corpus23.

A source limitation qualifies this entire section. The two passages quoted here were taken from a modern comprehensive work of jurisprudence that cites Radd al-muḥtār and al-Khaṣṣāf. The cited passage in Radd al-muḥtār has been examined, allowing the original wording to be supplied more fully and its attribution to be corrected; the results are recorded in the citation footnote. Examination invalidated the attribution of the register rule to al-Khaṣṣāf: neither the passage nor its chapter was found in his Aḥkām al-awqāf, whereas Radd al-muḥtār explicitly attributes it to al-Fatāwā al-Khayriyya24. Further caution is required because the same source dates Egyptian Waqf Law 48 to 1952, while the instances in the corpus refer to it as a law of 1946, a date also favoured by an independent historical source25. This source is therefore used solely for its transmission of juristic texts, and no legislative facts are based on it.

Footnotes

  1. Ibn Nujaym, al-Ashbāh wa-l-naẓāʾir [Analogues and Parallels], the twenty-seventh rule in the Book of Waqf. Its base text was examined in the edition with commentary: Aḥmad ibn Muḥammad Makkī al-Ḥamawī, Ghamz ʿuyūn al-baṣāʾir fī sharḥ al-Ashbāh wa-l-naẓāʾir [Sharpening the Eyes of Insight: Commentary on Analogues and Parallels], Beirut: Dār al-Kutub al-ʿIlmiyya, 1st ed., 1405 AH / 1985, four volumes, page 228 in the continuous pagination of the combined version. The base text reads verbatim: “27 - The founder’s stipulation must be followed, for they say: the founder’s stipulation is like the Lawgiver’s text, that is, in the obligation to act upon it, 28 - and in implied meaning and signification, 29 - as we explained in Sharḥ al-Kanz [Commentary on the Treasure], except in certain cases”. The numbered text is Ibn Nujaym’s; what follows the divider “[Ghamz ʿuyūn al-baṣāʾir]” is al-Ḥamawī’s commentary, and the two must not be conflated. Before this round, the passage had been taken through ʿAbd al-Majīd ibn Muḥammad al-Subayyil, Qāʿidat sharṭ al-wāqif ka-naṣṣ al-shāriʿ [The Maxim That the Founder’s Stipulation Is Like the Lawgiver’s Text], the first section explaining the maxim in the Ḥanafī school. Its wording was completed, then its original examined here, removing dependence on the intermediary. The same passage in al-Subayyil states that Ibn Nujaym observed in al-Baḥr al-rāʾiq [The Clear Sea] that the proposition was not unrestricted: some founders’ stipulations are valid and must be acted upon, while others are invalid. No edition of al-Baḥr was examined, so this subsidiary point remains cited through al-Subayyil. The original of the Ḥanafī statement that what counts is what actually occurred, rather than what was written, was examined here in Radd al-muḥtār [The Reply of the Perplexed], at the location specified in footnote 5. It is attributed there to al-Fatāwā al-Khayriyya [The Khayriyya Fatwās], in these words: “In al-Khayriyya, they explicitly state that, concerning stipulations, what counts is what actually occurred, rather than what was written in the waqf document. If evidence is produced for something not found in the waqf deed, it must undoubtedly be acted upon, because the written document is mere writing and carries no weight, being outside the recognised forms of sharia evidence”. This is a source independent of al-Zuḥaylī for the proposition stated in 3.1 that writing alone is not self-standing evidence. The statement belongs to al-Khayriyya, rather than Ibn Nujaym: this corrects the attribution without changing the ruling. ↩

  2. Ibn ʿĀbidīn, Ḥāshiyat Radd al-muḥtār ʿalā al-Durr al-mukhtār [The Reply of the Perplexed: A Gloss on The Choice Pearl] (al-Ḥalabī edition). The two examined locations are: “[Section: Renting a House Containing Trees]”, followed by “Section on Their Saying That the Founder’s Stipulation Is Like the Lawgiver’s Text” and the material following it up to “Section on Cash Stipends in Awqāf”. Examination corrected the textual layer without changing the ruling. An earlier version of this text stated that Ibn ʿĀbidīn had “formulated” the maxim by saying “that is, in implied meaning and signification, and the obligation to act upon it”, citing Abū ʿUmar Dubyān ibn Muḥammad al-Dubyān, al-Muʿāmalāt al-māliyya aṣālatan wa-muʿāṣaratan [Financial Transactions: Classical Foundations and Contemporary Practice], page 508, which states verbatim: “Ibn ʿĀbidīn said in his gloss: their saying that the founder’s stipulation is like the Lawgiver’s text means: in implied meaning and signification, and the obligation to act upon it”. Examination of Radd al-muḥtār showed that this wording belongs to the base text being explained, rather than to the glossator. Ibn ʿĀbidīn introduces it using the commentary formula “(His words: their saying that the founder’s stipulation is like the Lawgiver’s text)”, then “(His words: that is, in implied meaning and signification, etc.)”, then “(His words: and the obligation to act upon it)”. These are lemmata from the base text. The error lay in the intermediary source, passed into this paper and has now been corrected. Ibn ʿĀbidīn objects to and qualifies the statement rather than merely transmitting and endorsing it. He writes: “Such is the wording in al-Ashbāh, whereas what appears in al-Baḥr, transmitted from the learned Qāsim, is ‘in understanding and signification’. This is appropriate, because implied meaning is not recognised by us in texts”. On the second clause he writes: “(His words: and the obligation to act upon it) This contradicts what we have just transmitted”, before reconciling the positions with “It appears to me that there is no incompatibility”. His statement “Such is the wording in al-Ashbāh” traces the wording back to Ibn Nujaym, ruling out its treatment as a second independent text. The classification into unequivocal, manifest, equivocal and indeterminate wording is also attested verbatim at the same location, introduced by: “The learned Qāsim said: I say, if the meaning is as stated, then any wording of the founder that is unequivocal, admitting neither restriction nor interpretation, must be acted upon…”. This is therefore Qāsim ibn Quṭlūbughā’s statement, transmitted rather than originated by Ibn ʿĀbidīn, as direct examination confirms. The edition number, year and place of publication are not recorded in the examined version, so the location is identified by the section title. ↩

  3. See note 13. ↩

  4. See note 13. ↩

  5. Fatāwā Dār al-Iftāʾ al-Miṣriyya [Fatwās of the Egyptian Dār al-Iftāʾ], “Ḥaqq al-mustaḥiqq fī al-rujūʿ bi-mā lam yastaḥiqqahu” [The Entitlement-Holder’s Right to Recover What He Was Not Entitled To], fatwā of ʿAbd al-Majīd Salīm, location 2817. The questioner was a woman, Khadūja Muḥammad. The fatwā’s heading reads “Ṣafar 1350 AH - 12 June 1932”. The correspondence between the dates is inconsistent, since Ṣafar 1350 AH corresponds to June 1931. Both dates are therefore retained as given, with the discrepancy flagged; neither is corrected by reference to the other without examination of the printed edition. The muftī’s wording is: “An entitlement-holder whom the nāẓir has not paid his due in past years, believing him not to be entitled, has the choice of recourse against the nāẓir or against the entitlement-holders of the upper tier. If he proceeds against the nāẓir, the nāẓir may in turn recover from them what was recovered from him”. This rests on a passage in Tanqīḥ al-Ḥāmidiyya [Revision of al-Ḥāmidiyya] transmitted from al-Bazzāziyya [Al-Bazzāzī’s Fatwā Collection]. ↩

  6. The same source, “ʿAdam samāʿ daʿwā al-istiḥqāq li-muḍiyy al-mudda” [Barring an Entitlement Action from Being Heard Owing to the Passage of Time], fatwā of Muḥammad ʿAbduh, Jumādā al-Ākhira 1320 AH, location 3063. ↩

  7. The same source, “al-Taqādum lā yusqiṭ al-ḥaqq mahmā ṭāla al-amad” [The Passage of Time Does Not Extinguish a Right, However Long It May Be], fatwā of Muḥammad Bakhīt, Ramaḍān 1338 AH, location 2656. It concerns action 82 of 1916, brought against the Ministry of Finance before the Miṣr Sharia Court, on the basis of a waqf deed dated 20 Rajab 1174 AH. All quotations are from the body of the answer, in which the muftī quotes Article 587 of Qānūn al-ʿadl wa-l-inṣāf [Code of Justice and Equity], drawn from Tanqīḥ al-Ḥāmidiyya [Revision of al-Ḥāmidiyya], verbatim. He then states that consideration of the bar to hearing the action depends on Article 376 of Courts Regulation 31 of 1910. The quotation’s exact wording was checked against the fatwā text in the combined version and matched. The text reads: “He stated in Article (587) of Qānūn al-ʿadl wa-l-inṣāf, drawing on Tanqīḥ al-Ḥāmidiyya (a right is not extinguished by the passage of time…”. Attribution to al-Tanqīḥ is therefore explicit in the text. It must be noted that the fatwā’s title was editorially supplied later and states the limitation rule without qualification, whereas the answer’s text qualifies it by reference to possession and ownership. The citation here is therefore to the wording of the text, rather than the title or the editorially formulated principle. This qualification applies wherever the limitation rule is reused. The heading contains no Gregorian date, so only the Hijri date is given. ↩

  8. The same source, “al-Tanāzul ghayr al-iqrār” [Assignment Is Not Acknowledgement], fatwā of ʿAbd al-Majīd Salīm, Ramaḍān 1347 AH / 14 February 1929, location 2736. ↩

  9. Mudawwanat aḥkām al-waqf al-fiqhiyya [Compendium of Juristic Rulings on Waqf] (1st ed., 1439 AH / 2017), pages 240 and 215 in the continuous pagination of the combined version, on the forms of litigation by an entitlement-holder against the mutawallī in Ḥanafī jurisprudence. Both locations were examined, requiring two qualifications to the citation. First, page 240 sets out the first form: that “the entitlement-holder litigates against the mutawallī in an action brought to establish his entitlement to the administratorship, or that he is among the waqf’s entitlement-holders, or that the mutawallī has not paid him all the revenue to which he is entitled”. The same context expressly affirms the entitlement-holder’s right to litigate “although he is not a party to the action, and the party is the mutawallī”. Second, page 215 makes the branch concerning an account for breach of trust and a request for removal conditional on the judge’s authorisation: “The judge authorises one of the entitlement-holders to litigate against the mutawallī, if the latter was appointed by the founder, by appointing that entitlement-holder as a temporary mutawallī and entrusting him with the task of bringing proceedings against the original mutawallī, such as holding him to account for breach of trust or seeking his removal for incapacity or neglect”. Removal is thus conditional on the judge’s authorisation and appointment—a qualification that must be recorded rather than omitted. The volume is not identified in the examined version, so only the continuous pagination is given. The citation is through this modern compendium; the original Ḥanafī texts were not examined for this paper. ↩

  10. Wahba al-Zuḥaylī, al-Fiqh al-Islāmī wa-adillatuhu [Islamic Jurisprudence and Its Evidence], p. 7667, chapter five. Both passages are quoted verbatim from this location, which refers to Radd al-muḥtār, 3/408 and 441–444. Examination of the cited passage in Radd al-muḥtār requires three differences between the intermediary source and its original to be recorded. First, the original reads: “A documentary instrument, however, cannot serve as evidence, because one handwriting resembles another.” Ibn ʿĀbidīn attributes this to al-Khāniyya [The Khāniyya Fatwās] and al-Isʿāf [Assistance], rather than originating it. Second, the original attributes the register rule to al-Fatāwā al-Khayriyya [The Khayriyya Fatwās], stating: “And in al-Khayriyya: if the waqf has a deed in the judges’ archive, customarily known among us as the register, and it is in their custody, its contents shall be followed by way of istiḥsān if those concerned dispute it.” Attribution of the rule to al-Khaṣṣāf is therefore al-Zuḥaylī’s, and finds no support in the passage he cites. Examination of al-Khaṣṣāf’s original invalidated this attribution, resolving what had previously remained an outstanding item in the evidentiary record: Aḥkām al-awqāf [Rulings Governing Awqāf], by Abū Bakr Aḥmad ibn ʿAmr al-Shaybānī, known as al-Khaṣṣāf, was examined by reading its entire contents list—one hundred and thirty-nine sections, from its introduction to “Chapter on a Man Who Dedicates Land as Waqf for His Clients”—without finding any chapter or discussion on proving waqf status through the register or on awqāf of long standing. Four root-based searches confined to the book were then conducted: “records of judges’ archives”, “its witnesses died long ago”, “judge’s archive waqf deed”, and “istiḥsān those concerned disputed it”. None yielded material on this subject. This conclusion is supported by the fact that al-Khaṣṣāf’s entire book takes the form of a dialogue (“I said… he said…”), whereas the passage attributed to him takes the form of a declaratory fatwā. This paper therefore does not attribute the ruling to al-Khaṣṣāf, but to al-Fatāwā al-Khayriyya on Ibn ʿĀbidīn’s authority. The argument in the main text that the register serves as an evidentiary repository of rights within the juristic tradition itself remains unchanged, but its dating shifts from al-Khaṣṣāf’s historical layer to that of a later fatwā. Third, the original includes an intermediate level between the register and proof that al-Zuḥaylī omits: “Otherwise, consideration shall be given to what was established about its circumstances in earlier times, namely how its administrators used to act.” This is followed by juristic analogy: “Whoever establishes a right by proof shall have judgement given in their favour for that right.” The original hierarchy thus has three levels: the register, the established practice of the administrators, and proof. Ibn ʿĀbidīn adds in the same passage that reliance on the register is an application of istiḥsān where the documentary instrument agrees with it, “because, if it has a deed agreeing with what is in the judges’ register, this strengthens it, especially if the deed bears the handwriting of former judges”. This examination entails no revision to any implication in 3.1 or to any claim; it corrects the attribution and its location and supplies the original wording more fully. The qualification concerning al-Zuḥaylī’s legislative information remains: in the same passage he dates Egyptian Waqf Law 48 to 1952, stating, “Then Article One of Egyptian Waqf Law (48 of 1952) provided”. He is therefore not relied upon for legislative facts. ↩

  11. See note 21. ↩

  12. Muḥammad ʿAbduh, “Shurūṭ al-wāqifīn” [Founders’ Stipulations], al-Manār [The Beacon], issue 6, 22 June 1902 CE, pp. 201–217; the quotation is from pp. 210–211, and reads in full: “The statement of some jurists that ‘the founder’s stipulation is like the Lawgiver’s text’ has become a familiar saying, widely known among people. It has governed the practice of the courts from long ago to the present day: they adhere to words written in ‘waqf deeds’ which the founder may not even have understood, but which the scribe wrote among his conventional formulations. By adhering to these words, they sometimes disregard what the Lawgiver and the founder intended by the waqf.” Attribution to Muḥammad ʿAbduh rests on the article-author field in the periodicals archive, not the journal-proprietor field; its proprietor was Muḥammad Rashīd Riḍā. An internal reference in “Fatāwā al-Manār” [Fatwās of al-Manār], issue 20, 23 December 1904 CE, pp. 778–782, supports the identification of the passage: “Most of our scholars have said that the founder’s stipulation is like the Lawgiver’s text and cannot be changed, but some have refuted this statement with strong evidence… For the detailed discussion, see (p. 210) of the fifth volume of al-Manār.” The article-author field for this second item is blank in the archive, so it is not attributed to a named author. Muḥammad ʿAbduh is one of the muftīs represented in this corpus, and this paper cites three instances attributable to him; the evidence thus belongs to the same framing layer. It describes court practice and supports an established implication, rather than supplying a count included in this paper’s measures. ↩

  13. See note 21. ↩

  14. Ibrāhīm al-Bayyūmī Ghānim, al-Awqāf wa-l-siyāsa fī Miṣr [Awqāf and Politics in Egypt], p. 444: “We noted earlier that the state intervened for the first time to lay down rules for the waqf system itself when it issued Waqf Law 48 of 1946”; and p. 460: “On 14 September 1952 it issued Decree-Law 180 of 1952, which provided for the retention only of purely charitable waqf”; see also p. 434 on the subsequent fate of family awqāf. The date 14 September 1952 matches the date referred to in the corpus instances as the commencement of the law’s application. ↩

4 · A Document with Evidentiary Authority: Measuring the Presence of the Waqf Deed in the Production of Claims

4.1 The Formula and the Procedural Setting

One of the clearest findings from reading the corpus is a formula repeated in the fatwā texts so frequently that it almost constitutes a procedural ritual carrying evidentiary authority: “We have examined the question and an official copy of the waqf deed.” The muftī answers on the basis of the attached sharia document, rather than the questioner’s oral account or assertion. He frequently makes his answer conditional when this document is absent, or calibrates it to the degree of authentication when the copy is unofficial. In this sense, issuing fatwās on entitlement is a continuous exercise in documentary examination: the deed is read, the stipulation extracted, and its provisions applied to the case. The evidentiary instrument is thus not the system’s silent background but the setting in which it operates.

The corpus contains every link in the documentary chain: (1) the deed issued by the sharia court, with its date and archival reference number; (2) subsequent attestations of amendments, additions and removals of beneficiaries; and (3) the official copies used by entitlement-holders, together with the sharia court register, which one questioner describes as protected from any taint of forgery26.

Its evidentiary authority can be invoked against the government administration itself. The action brought by two nāẓirs against the Ministry of Finance rested on a sharia waqf deed dated (20 Rajab 1174 AH / 1761), challenging the government’s possession of the property. The muftī presented it as grounds for litigation a century and a half after its issue27. The outcome of the action before the Miṣr Sharia Court falls outside this paper’s evidence: the source belongs to the framing layer, whereas the judgement belongs to the adjudication layer.

The system has acknowledged internal limits. A waqf initiated through a private instrument is subject to judicial uncertainty (section 7 will show that attestation is a condition for hearing an action, not for validity). Loss of the waqf deed permits proof through established dealings and public repute. An attested instrument that has not undergone land registration after the registration laws occupies an intermediate position28.

4.2 The Count Conducted: Did the Document Recede as Law Advanced?

The preceding account must be tested against the entire corpus. The greatest threat to this paper’s claim is the apparently self-evident replacement hypothesis: if codified law became the reference point of the answer, the document would be expected to recede from the process of making claims, since the general statutory provision would render the particular stipulation unnecessary.

The count was conducted across all texts in the corpus using disclosed lexical matching patterns, over six successive periods, with three separate counters: (1) mention of the document in any of its formulations; (2) mention of the expression “waqf deed” alone; and (3) occurrence of the opening formula “We have examined”.

Matching was confined to the record’s text after removal of its heading and the publishing website’s footer; the protocol and its patterns appear in Appendix 129. A second measure, adjusted for text length, supplements the binary counter: the frequency of the formulations per thousand words. This is necessary because the probability of an occurrence within a record increases automatically with its length. Length is calculated from the cleaned text, rather than the length field recorded in the files, because that field compresses the relative differences between periods to approximately one-third of their size30. The following table presents the results for both measures, calculated on the frozen release:

PeriodnInvocation of documentsPer thousand words“Waqf deed”“We have examined” formulaMean text lengthReliance on law
1894–191310625%1.4314%3%4371%
1914–193312366%3.0350%74%6062%
1934–19455260%2.9938%96%5652%
1946–19534072%4.0958%75%68460%
1954–19594879%3.9960%60%74790%
1960–19802152%3.8643%29%71552%

Table 2: Invocation of documents and reliance on law across the six periods. Percentages are calculated against the number of dated instances in each period; length is measured in words in the cleaned text. The final row describes a thin sample rather than an observed trend, as explained in section 10.

The direction of the results is consistent: both measures invalidate the replacement hypothesis:

  • The binary indicator reaches its maximum of seventy-nine per cent (79%) in the same period (1954–1959 CE) in which reliance on law reaches ninety per cent (90%).

  • The length-adjusted indicator peaks (4.09 per thousand words) in the immediately preceding period (1946–1953 CE), and is almost matched in the following period (3.99).

  • Under both measures, therefore, invocation of documents peaks within the decade of legislative liquidation. Mention of “waqf deed” also reaches its maximum in 1954–1959.

The two curves rose together rather than exchanging positions. This suggests a change in the reference point through which the same evidentiary instrument was read, rather than the replacement of one evidentiary instrument by another: the claimant retains the deed, while the law used to assess it and derive its legal consequences changes.

Five qualifications to this count must be recorded rather than passed over:

  • First, the count is lexical and automated: it measures the presence of an expression, not its weight in the reasoning. Coding the document’s weight remains pending.

  • Second, the increase between the first and second periods admits an editorial explanation. The rise in “We have examined” from three per cent (3%) to seventy-four (74%) coincides with the rise in mentions of the document.

  • Third, a finding that limits that explanation is that the two curves subsequently diverge in both peak and direction. This is addressed in 10.2.

  • Fourth, the difference between the two measures in the timing of the peak remains and must not be overlooked. The binary indicator places it in 1954–1959, while the length-adjusted indicator places it in 1946–1953. The evidence supports a peak within the decade of liquidation, without identifying a single period within that decade.

  • Fifth, the final period cannot sustain a reading of trends at all because its two decades are poorly represented. Its row should therefore be read as a description of a sample; the figures for its representation appear in 10.131.

4.3 Case Record: The al-Minshāwī Waqf Before Dār al-Iftāʾ

Evidentiary authority is tested in practice, not in abstract description. The waqf of Aḥmad al-Minshāwī Pasha was therefore traced through five successive episodes between (1907 and 1921), whose questioners were the Ministry of Public Instruction, an individual nāẓir, and the headship of an institute.

Across these episodes, the deed remains the sole, indispensable reference point: its copy is invoked in every episode, and the answer rests on the precise wording of its stipulation. What changes is the type of questioner and the type of issue submitted. At the centre of the first system is thus a deed held in a court’s record office, rather than a government register. The episodes, their questioners, issues and texts are detailed in Appendix 2 and Appendix 9.

4.4 When the Document Is Absent: Thirteen Negative Instances

The more exacting test of the document’s position and evidentiary authority arises when it is absent rather than directly present: changes in the muftī’s conduct and reference point in its absence indicate its weight. Instances in which the muftī expressly states that he has not examined the waqf deed or sharia deed were counted using the formula “We have not examined”. There were thirteen negative instances. Their distribution on either side of the 1946 dividing point was calculated using the year in the fatwā heading rather than the automatically extracted year: five before it and eight after it32.

Reading these instances reveals a sharp division in the reference point of the answer, rather than in whether the muftī declines to issue a fatwā. None of the five instances preceding the law relies on codified law: three rely on jurisprudence with named sources, and two on the stipulation without naming sources. All eight subsequent instances rely on codified law, eight out of eight. Before the law, the document’s absence leaves the muftī to turn to quotations from jurists; afterwards, its absence leaves him to turn to a numbered article.

In 1938, ʿAbd al-Majīd Salīm states that he has not examined the waqf deed “because it was not sent to us”. He then answers the question of whether agricultural machinery forms part of the waqf property on the basis of what is apparent and citations from juristic texts33. In 1952, Ḥasanayn Muḥammad Makhlūf states: “We have examined the question but have not examined the waqf deed.” He then proceeds to answer that no separate portion is to be apportioned for the sum, “pursuant to Article 41 of Waqf Law 48 of 1946”34.

The count does not, however, identify making the juristic answer conditional on the accuracy of the question as a characteristic of the first period. The classification rule counts an answer as conditional if it opens with a conditional expression making the ruling dependent on the truth of the question, including “if it is” and “if it is correct”.

On this basis, the conditional instances number one out of five (1 out of 5) before 1946 and two out of eight (2 out of 8) afterwards. Conditional answering thus persists on both sides of the dividing line (the best-known later example is an instance from 1951 in which ʿAllām Naṣṣār qualifies his answer with “if what is stated in the question is correct”, then bases the ruling on Law 48 of 194635. This is an instance that departs from the pattern and must be recorded rather than passed over).

This evidence supports a change in the reference point of the answer when the document is absent, rather than a change in the muftī’s ability to answer. It supports the first claim through evidence distinct from the overall count, and supports nothing in the second.

Two limitations qualify this evidence. The first concerns sample size: thirteen instances out of three hundred and ninety-six is a small number, insufficient on its own to sustain a claim about a period. The second concerns the division: instances are divided at 1946 using the fatwā heading. Some records in the corpus have an unreliable automated dating key, including two of these thirteen. Their locations are identified in section 8.

Footnotes

  1. Fatāwā Dār al-Iftāʾ al-Miṣriyya [Fatwās of the Egyptian Dār al-Iftāʾ], “al-Iqrār bi-l-istiḥqāq fī al-waqf” [Acknowledgement of Waqf Entitlement], fatwā of ʿAbd al-Majīd Salīm, Dhū al-Qaʿda 1358 AH / 3 January 1940, location 3033. It contains the questioner’s description of the sharia court register as protected from any taint of forgery. This description comes from a single questioner’s statement, not the muftī’s answer, and must not be generalised to questioners as a whole. ↩

  2. The same source, “al-Taqādum lā yusqiṭ al-ḥaqq mahmā ṭāla al-amad” [The Passage of Time Does Not Extinguish a Right, However Long It May Be], fatwā of Muḥammad Bakhīt, Ramaḍān 1338 AH, location 2656. It concerns action 82 of 1916, brought against the Ministry of Finance before the Miṣr Sharia Court, on the basis of a waqf deed dated 20 Rajab 1174 AH. All quotations are from the body of the answer, in which the muftī quotes Article 587 of Qānūn al-ʿadl wa-l-inṣāf [Code of Justice and Equity], drawn from Tanqīḥ al-Ḥāmidiyya [Revision of al-Ḥāmidiyya], verbatim. He then states that consideration of the bar to hearing the action depends on Article 376 of Courts Regulation 31 of 1910. The quotation’s exact wording was checked against the fatwā text in the combined version and matched. The text reads: “He stated in Article (587) of Qānūn al-ʿadl wa-l-inṣāf, drawing on Tanqīḥ al-Ḥāmidiyya (a right is not extinguished by the passage of time…”. Attribution to al-Tanqīḥ is therefore explicit in the text. It must be noted that the fatwā’s title was editorially supplied later and states the limitation rule without qualification, whereas the answer’s text qualifies it by reference to possession and ownership. The citation here is therefore to the wording of the text, rather than the title or the editorially formulated principle. This qualification applies wherever the limitation rule is reused. The heading contains no Gregorian date, so only the Hijri date is given. ↩

  3. The same source, “Waqf khayrī bi-ishhād lam yushhar” [A Charitable Waqf Constituted by an Attested Instrument That Has Not Undergone Land Registration], fatwā of Aḥmad Harīdī, Jumādā al-Ākhira 1381 AH / 28 November 1961, location 2863. ↩

  4. The lexical counting protocol appears in section 4, with its patterns and scope detailed in the appendix. The entire counter was rerun on the frozen release WEC-432/1.0 when references to that release were standardised. Percentages are calculated against the dated instances in each period, numbering 106, then 123, 52, 40, 48 and 21. The effect of cleaning the matching scope was measured and disclosed: removing the heading and excluding the name of the Hijri month changed the document-indicator classification of twenty-one dated records out of three hundred and ninety. Most were in the first period, whose headings contain only the Hijri date. The indicator percentages consequently fell from 36%, 69%, 65%, 72%, 81% and 62% to those shown in the table. The “waqf deed” and “We have examined” counters were unaffected by cleaning, since neither formulation occurs in the headings. It must be noted that rerunning the count on the frozen release changed the final-period row: previously calculated as 57%, 3.58 and 48%, with a mean length of 797 words, it became the row shown in the table. The difference arises from a single record whose cleaned text differed between the earlier calculation and the release; it changed neither measure’s peak period. Recalculation reproduced the values of all five other rows unchanged. ↩

  5. The collective test of the competing explanation on the Waqf Entitlement Corpus, with its code retained alongside it. The test read text cleaned of extraneous material from the publishing website without modifying the source. The following figures are drawn from it. Length: median cleaned length, 2,668 characters; extraneous website material, 4,893 characters, with a deviation of 469; median length by decade, 2,480, 2,463, 3,629 and 2,435; question length did not increase (1,066 → 1,080), while answer length almost doubled (893 → 1,689); the gap within length quartiles was 83.3, 65.3, 66.7 and 74.5 points; normalisation, 80.5% → 80.9%; logistic regression on 390 instances, with a log-length coefficient of 0.692 (standard error 0.308) and a period coefficient of 4.434 (standard error 0.414), and a period odds ratio of 91.8 → 84.3; the sensitivity test requires length to increase 77.8-fold, against an observed increase of 1.47; citation density per ten thousand characters: statutory citations, 0.52 → 10.69, and named juristic citations, 3.15 → 1.13. Composition of issues: questions of devolution and transfer, 41.1% → 60.0% → 78.7%; the core family of issues accounted for between 70.2% and 91.7% of each decade, and within it the figures were 0.0% (n=41) → 17.9% (n=28) → 82.5% (n=63); within named and descendant entitlement-holders, 0.0% → 27.3% → 82.8%; after excluding every text whose question mentions a law or liquidation, 4.5% (n=88) → 78.5% (n=65), a gap of 73.9 points; combined four-way normalisation, 90.7%, with a gap of 83.1 points and a 95% confidence interval based on two thousand resamples [71.0, 92.4], subject to the qualification that 38 weighting units out of 92 have no comparator. Questioner identity: individuals’ share, 17.9% → 41.7% → 63.6% → 88.2%, and the combined share of courts and ministries, 35.7% → 5.9%; within individuals alone, 0.0% → 20.0% → 79.6%; cross-tabulation between 1930–1949 and 1950–1959: individuals, 12% (n=25) → 80% (n=49); courts, 8% (n=26) → 75% (n=12); unspecified, 6% (n=31) → 83% (n=12). The decisive test and its figures appear in Table 4. Publication density: correlation between decade size and the percentage relying on law, −0.26, and between waqf’s share of the decade’s fatwās and the percentage relying on law, −0.49; annual series: 1944, zero out of seven; 1946, two out of two; and 1955, 90.9% out of 22; excluding 1955 entirely leaves the percentage for the 1950s at 76.4%. Controlling for the muftī: Ḥasanayn Makhlūf, 41.7% (n=12) in the 1940s and 76.2% (n=21) in the 1950s; in the 1950s, Ḥasan Maʾmūn, 89% (n=35), Mughīth, 89% (n=9), and Naṣṣār, 64% (n=11). Representation of decades relative to their share of the published collection: 1890 = 3.96, 1900 = 1.48, 1910 = 1.33, 1920 = 0.95, 1930 = 1.19, 1940 = 0.65, 1950 = 0.91, 1960 = 0.37, and 1970 = 0.19. ↩

  6. See note 30. ↩

  7. Enumeration of instances of the document’s absence in the corpus using the formula “We have not examined”: thirteen instances, divided by the year in the heading into five before 1946 and eight afterwards. The first five are: “Waqf istiḥqāq wa-bināʾ al-wāqif ʿalayhi” [A Waqf with Designated Beneficiaries and the Founder’s Construction on Its Property], ʿAbd al-Raḥmān Qarāʿa, Dhū al-Qaʿda 1329 AH / 27 July 1921, location 3047—the correspondence between the dates in its heading is inconsistent, since Dhū al-Qaʿda 1329 AH corresponds to October 1911 CE; both dates are therefore retained as given, and neither is corrected by reference to the other without examination of the printed edition; its row in the distribution using the second dating key must be read subject to this qualification—and it is the only instance on this side with a conditional answer, using the formula “if the waqf deed transcribed in the question is valid”; “Waqf bi-sharṭ muḥāsabat al-nāẓir” [A Waqf Subject to the Stipulation That the Nāẓir Render an Account], ʿAbd al-Majīd Salīm, Rajab 1351 AH / 31 October 1932, location 2893; “Isqāṭ al-istiḥqāq wa-l-naẓar wa-jaʿluhu li-l-ghayr” [Waiver of Entitlement and Administration and Their Assignment to Another], ʿAbd al-Majīd Salīm, 3 November 1932 CE, location 3027; “Ḥukm al-ālāt al-zirāʿiyya allatī ustuḥdithat baʿd al-waqf” [Ruling on Agricultural Machinery Added after the Creation of the Waqf], ʿAbd al-Majīd Salīm, Ṣafar 1357 AH / 23 April 1938, location 2826, which states, “We have not examined the waqf deed because it was not sent to us”; and “Ḥukm wafāt aḥad al-nāẓirayn” [Ruling on the Death of One of the Two Nāẓirs], ʿAbd al-Majīd Salīm, 2 Muḥarram 1362 AH / 30 November 1943, location 2969. The correspondence between the dates in this heading is also inconsistent, since 2 Muḥarram 1362 AH corresponds to 9 January 1943 CE; both dates precede 1946, so they do not change the division of instances. None relies on codified law. All eight subsequent instances rely on it, including “Waqf istiḥqāqī” [A Waqf with Designated Beneficiaries], ʿAllām Naṣṣār, Ramaḍān 1370 AH / 26 June 1951, location 3038, which states, “We have examined the question but have not examined the waqf deed - if what is stated in the question is correct…” and then bases the ruling on Waqf Law 48 of 1946; this is an instance with a conditional answer after 1946. Another is “al-Waqf ʿalā qirāʾat al-Qurʾān al-karīm” [Waqf for the Recitation of the Noble Qurʾān], Ḥasan Maʾmūn, Dhū al-Ḥijja 1376 AH / 21 July 1957, location 2857, the second conditional answer on this side. 🔴 This passage previously stated that the two dating keys differed by more than two years in two of the thirteen records; the measured number is four. The count was rerun on the frozen release by comparing the Hijri field, converted using the adopted formula, with the extracted Gregorian field. Four locations differed: location 3038 (1370 AH → 1951 CE, with 1914 in the Gregorian field, a difference of thirty-seven); location 3027 (1315 AH → 1897 CE, with 1932 in the field, a difference of thirty-five); location 2855 (1376 AH → 1957 CE, with 1942 in the field, a difference of fifteen); and location 3047 (1329 AH → 1911 CE, with 1921 in the field, a difference of ten). In all four, the discrepancy lies in the automatically extracted Gregorian field, not the Hijri date recorded in the heading. ⚠️ This does not alter the division into five instances before 1946 and eight afterwards, since that division uses the Hijri year in the heading. Locations 3027 and 3047 precede 1946 under both dating keys, while locations 3038 and 2855 follow it under the Hijri key, which is the one adopted. ↩

  8. See note 32. ↩

  9. The same source, “Waqf ahlī wa-khayrī” [A Family and Charitable Waqf], fatwā of Ḥasanayn Muḥammad Makhlūf, Muḥarram 1372 AH / 19 October 1952, location 2840; the quotation comes from the answer itself. See also “Waqf istiḥqāqī bi-sharṭ” [A Waqf with Designated Beneficiaries Subject to a Stipulation], fatwā of Aḥmad Ibrāhīm Mughīth, Jumādā al-Thāniya 1374 AH / 25 January 1955, location 2908, in which the muftī bases his answer on “Article 22 of Waqf Law 47 of 1946”, as stated in the answer itself, whereas the other instances in the corpus refer to Law 48 of 1946. Consultation of al-Jarīda al-Rasmiyya [Official Gazette] was not possible to resolve this discrepancy; both numbers are therefore retained as given, and neither is corrected by reference to the other. ↩

  10. See note 32. ↩

5 · Entitlement as Economic Value Open to Transactions

Was entitlement a charitable hope, or a position with value that could circulate in the market? The corpus answers through a three-part conceptual apparatus that defines the “economic potential of entitlement”: susceptibility to valuation (determining the monetary and material value of a portion), liability to attachment and use as security (the permissibility of judicial attachment of entitlement revenue for creditors), and transferability (the legal limits on transferring or assigning the right). These dimensions are not equivalent. This is the author’s term in this paper, confined to the definition given here.

The corpus documents four things: (1) creditors’ attachment of entitlement, alongside a prohibition on waiving it in favour of someone without entitlement, in a fatwā of Shaykh Muḥammad ʿAbduh36; (2) priority in attachment, with maintenance taking precedence over debts, in a fatwā of Bakrī al-Ṣadafī37 and a comparable case concerning attachment for a divorced woman38; (3) limited transferability, since assignment for a specified period is permissible39, a purported sale is tested against jurisprudence40, and inheritance is precluded altogether41; and (4) founders’ stipulations excluding beneficiaries who incur debts, in four instances between 1921 and 195542. The instances and their texts are detailed in Appendix 4, and the full text of this section appears in Appendix 9.

This section is subject to a limitation on enumeration: an instance involving the sale of entitlement in 1903 was not included in the frozen release and is classified as a confirmed miss. What is presented here is therefore a lower bound on the collection’s evidence of capacity for monetisation, not an exhaustive enumeration of it43.

The conclusion is proportionate to the evidence: waqf entitlement is valued, attached and pledged through its receipt, and the entitled position passes according to the founder’s stipulation. It is an economic value fully open to transactions, even though the underlying legal position cannot lawfully be sold.

Footnotes

  1. The same source, “Ḥukm isqāṭ al-rayʿ” [Ruling on the Waiver of Revenue], fatwā of Muḥammad ʿAbduh, Ramaḍān 1321 AH, location 2994; the account is supplied by the superintendent of the al-Ḥilmiyya Awqāf. The three quotations come from the answer itself, not from the compilation’s principles. ↩

  2. The same source, “Nafaqat al-mustaḥiqq wa-zawjatihi wa-ʿiyālihi muqaddama ʿalā sāʾir al-duyūn” [Maintenance for the Entitlement-Holder, His Wife and His Dependants Takes Precedence over All Other Debts], fatwā of Bakrī al-Ṣadafī, Ṣafar 1329 AH, location 2976; the questioner is Ṣāliḥ Effendi, agent for the affairs of al-Sitt Fāṭima Hānim. The quotation comes from the answer, in which the muftī bases his ruling on passages from Tanqīḥ al-Ḥāmidiyya and Fatāwā al-Khayriyya [The Khayriyya Fatwās]. The sentence “Maintenance for the entitlement-holder, his wife and his dependants takes precedence over all other debts” is the compilation’s title wording, not the muftī’s wording. The heading gives no Gregorian date; Ṣafar 1329 AH corresponds to February 1911. ↩

  3. The same source, “Istimrār ḥukm nafaqat al-muṭallaqa fī al-ʿidda” [Continuation of a Divorced Woman’s Maintenance Order during the Waiting Period], fatwā of Muḥammad Bakhīt, Rabīʿ al-Thānī 1336 AH / 22 January 1918, location 583. ↩

  4. The same source, “al-Tanāzul ghayr al-iqrār” [Assignment Is Not Acknowledgement], fatwā of ʿAbd al-Majīd Salīm, Ramaḍān 1347 AH / 14 February 1929, location 2736. ↩

  5. The same source, “Al-tanāzul ʿan al-istiḥqāq fī al-waqf wa-bayʿuhu” [Assignment and Sale of Waqf Entitlement], fatwā of ʿAbd al-Majīd Salīm, Dhū al-Qaʿda 1356 AH / 22 January 1938, location 3032. ↩

  6. The same source, “Al-istiḥqāq fī al-waqf laysa tarika” [Waqf Entitlement Is Not an Estate], fatwā of ʿAbd al-Majīd Salīm, Shaʿbān 1353 AH / 13 November 1934, location 3030. ↩

  7. The corpus contains four instances of stipulations excluding beneficiaries for incurring debts: “Waqf istiḥqāqī wa-khayrī” [A Waqf with Designated Beneficiaries and Charitable Purposes], ʿAbd al-Raḥmān Qarāʿa, 18 September 1921, location 3016 (al-Minshāwī’s stipulation concerning interest-bearing debts); “Waqf istiḥqāqī wa-khayrī” [A Waqf with Designated Beneficiaries and Charitable Purposes], ʿAbd al-Raḥmān Qarāʿa, Rabīʿ al-Awwal 1344 AH / 15 November 1925, location 3020, which includes a stipulation excluding “any of the aforementioned waqf recipients who incurs a debt and fails to repay his creditors, resulting in the creditor or his representative intending to attach the waqf properties”; “Ishtirāṭ al-wāqif ʿadam sadād al-dayn min ghallat al-waqf” [The Founder’s Stipulation That Debt Not Be Repaid from Waqf Revenue], ʿAbd al-Majīd Salīm, Jumādā al-Ākhira 1349 AH / 13 November 1930, location 2892, which stipulates “that a debt incurred by any entitlement-holder shall not be discharged from waqf revenue”; and “Waqf istiḥqāqī wa-ḥirmān bi-sharṭ” [A Waqf with Designated Beneficiaries and Exclusion under a Stipulation], Ḥasan Maʾmūn, Muḥarram 1375 AH / 5 September 1955, location 2912, which stipulates that exclusion must precede the attachment order by ten days. The quotations come from the questions in the first three instances and from the answer in the fourth. ↩

  8. See note 5. ↩

6 · The Resident Student at al-Azhar: A Longitudinal Trajectory and the Turning Point of 1916

The “resident student at al-Azhar” is among the most extensively documented descriptively defined beneficiaries in this paper’s sources, although this has not been measured by counting. His historical trajectory makes it possible to trace a descriptively defined beneficiary, rather than a descendant entitlement-holder, across successive documentary systems through four historical snapshots:

The first snapshot concerns the entry standard (entry in the register): ʿAbd al-Raḥmān Katkhudā’s regular provision allocates two loaves each day to a specified number of teachers and students entered in the “register”, in order of registration. Those absent or deceased are replaced from among those waiting, and actual engagement in learning is required44. Here, the register records a right; it is not a subsistence roll. Governance is organised through communities around registration: each community has a register held by its naqīb (community head) and a shaykh who exercises authority within it, as ʿAlī Mubārak records45. The full texts from the waqf deed and the Khiṭaṭ appear in Appendix 6.

The second snapshot concerns receipt and value (the secondary market): regular provision consists of loaves in kind, varying between the riwāqs (residential halls), and resident students sell their provisions or exchange some of them for food to accompany their bread at the gates of al-Azhar Mosque46. This created a secondary market in the flow of benefits in kind, not in the original registration. The waqf and al-Azhar trajectories intersect within the corpus itself: founders dedicate awqāf to students, and fatwās define their qualifying descriptions through prevailing custom47.

The third snapshot concerns the turning point of 1916 (seeking authorisation under the deed and relocating the choice): on 31 January 1916, the Secretariat of the Supreme Council of al-Azhar wrote to the muftī, Shaykh Muḥammad Bakhīt, in communication no. 1255. The Council had considered a memorandum from the Dasūq Institute “concerning the proposed conversion of the students’ in-kind allowances into monthly cash payments, after obtaining a sharia fatwā permitting this”. It decided to refer the matter to the muftī “to examine the founders’ stipulations and issue a sharia opinion on the permissibility of that conversion”. Copies of the stipulations in the waqf deeds of Aḥmad Bey al-Sharīf and al-Minshāwī Pasha accompanied the letter48.

The instance encapsulates the entire programme on a single page: a modern apparatus, with a supreme council, a secretariat and incoming correspondence numbers, seeks to convert a benefit from provision in kind to cash. Yet it cannot proceed with implementation until the muftī has examined the founders’ stipulations—the new authority seeks authorisation under the old deed.

The express wording of the answer then goes further than the request. Shaykh Bakhīt did not merely grant permission; he relocated the choice, relying on Article 115 of Qānūn al-ʿadl wa-l-inṣāf [Code of Justice and Equity], drawn from “Radd al-muḥtār”. It reads: “If specified bread and meat are stipulated for the entitlement-holders each day or month, and the administrator wishes to pay them their value in cash, he may not do so. The choice belongs solely to the entitlement-holders: they may demand the specific entitlement designated for them, or demand its value in cash.” The muftī then held that the choice belonged to the entitlement-holders: “If they wish, they may take the bread itself, as stipulated by the founder; if they wish, they may take its value in cash”49. This was not the first recognition of that choice: Muḥammad ʿAbduh had recognised it earlier, in 1314 AH50.

The descriptively defined beneficiary thus possesses a legal choice against an administrative apparatus seeking to change the form of the benefit. That choice derives from the founder’s stipulation as read through jurisprudence. Here, the first system appears with particular clarity: changing the form of the benefit depends on the beneficiary’s consent. Whether, under the succeeding system, such a change became a decision issued by a body and challenged by the affected person remains an open question. It belongs within the limited administrative evidentiary gateway set out in section 7, and this paper’s material does not supply the answer.

The corpus later includes a fatwā on a stipulated fixed amount. Its editorially formulated principle in the compilation, rather than the text of its answer, states that the amount constitutes a fixed stipend, unaffected by changes in prices or in the value of the Egyptian pound. This is the other side of the question of monetary erosion51.

The fourth snapshot falls outside this paper’s detailed scope, and only its boundary is drawn: reforms at al-Azhar converted “regular provision in kind” into a “cash allowance”, and the “resident student at al-Azhar” into a “school student” within a public educational institution. The details are left to the material devoted to that subject52.

Footnotes

  1. ʿAlī Mubārak, al-Khiṭaṭ al-Tawfīqiyya al-jadīda [The New Tawfīqī Topography] (2nd ed., 1425 AH / 2004), 50/4. In two lines, the text brings together the geographical standard, registration, the reserve list and the replacement rule. ↩

  2. The same source, 49/4. The text concerns resident students generally, organised into communities, rather than particular riwāqs, and should be read within that limit. ↩

  3. Aḥmad Amīn, Qāmūs al-ʿādāt wa-l-taqālīd wa-l-taʿābīr al-Miṣriyya [Dictionary of Egyptian Customs, Traditions and Expressions] (Hindawi edition based on the 1953 edition), letter alif, entry on al-Azhar. The sale should be assessed through section 5: it is customary monetisation of what is received, not transfer of a legal position. The questions for examining it are: did the naqīb recognise it? Was the name in the register changed? ↩

  4. Fatāwā Dār al-Iftāʾ al-Miṣriyya [Fatwās of the Egyptian Dār al-Iftāʾ], “Waqf ʿalā ṭalabat al-ʿilm bi-l-Jāmiʿ al-Aḥmadī” [A Waqf for Students at al-Aḥmadī Mosque], fatwā of Muḥammad ʿAbduh, Dhū al-Ḥijja 1314 AH, location 2982; “Waqf khayrī” [A Charitable Waqf], fatwā of ʿAbd al-Raḥmān Qarāʿa, Rajab 1344 AH / 27 January 1926, location 2807, on the entitlement of students in the al-ʿAbbāsī riwāq; and “Waqf istiḥqāq” [A Waqf with Designated Beneficiaries], fatwā of Bakrī al-Ṣadafī, Dhū al-Ḥijja 1332 AH, location 3000, on a waqf for Jabartī resident students, in which the status of “bearers of the Qurʾān” is defined through the founder’s custom. ↩

  5. The same source, “Istibdāl al-jirāya bi-l-nuqūd” [Conversion of Regular Provision into Cash], fatwā of Muḥammad Bakhīt, Rabīʿ al-Awwal 1334 AH / 1 February 1916, location 3003; and the letter from the Secretariat of the Supreme Council of al-Azhar dated 31 January 1916, no. 1255, accompanied by a copy of correspondence from the Office of the Shaykh of al-Azhar, no. 39 of 10 July 1915, concerning the stipulations in the waqf deeds of Aḥmad Bey al-Sharīf and al-Minshāwī Pasha. The three quotations come from the question and the answer. The verbatim quotation of Article 115 was checked against the fatwā text in the collected edition and matched it. The text states that the article is “from Qānūn al-ʿadl wa-l-inṣāf, drawn from Radd al-mukhtār”; the attribution to Radd al-muḥtār is thus stated in the fatwā itself, not added to it. The spelling in the text is “Radd al-mukhtār”. ↩

  6. The same source, “Istibdāl al-jirāya bi-l-nuqūd” [Conversion of Regular Provision into Cash], fatwā of Muḥammad Bakhīt, Rabīʿ al-Awwal 1334 AH / 1 February 1916, location 3003; and the letter from the Secretariat of the Supreme Council of al-Azhar dated 31 January 1916, no. 1255, accompanied by a copy of correspondence from the Office of the Shaykh of al-Azhar, no. 39 of 10 July 1915, concerning the stipulations in the waqf deeds of Aḥmad Bey al-Sharīf and al-Minshāwī Pasha. The three quotations come from the question and the answer. The verbatim quotation of Article 115 was checked against the fatwā text in the collected edition and matched it. The text states that the article is “from Qānūn al-ʿadl wa-l-inṣāf, drawn from Radd al-mukhtār”; the attribution to Radd al-muḥtār is thus stated in the fatwā itself, not added to it. The spelling in the text is “Radd al-mukhtār”. ↩

  7. The same source, location 2982 (previously cited in footnote 32), recognising students’ choice to receive their designated entitlement in cash or as regular provision. The wording comes from the compilation’s principles; the ruling itself is stated in the answer. ↩

  8. The same source, “Waqf istiḥqāqī wa-l-taghyīr fīhi” [A Waqf with Designated Beneficiaries and Changes to It], fatwā of Ḥasan Maʾmūn, Rajab 1376 AH / 4 February 1957, location 2928. The sixth principle reads: “The stipulated fixed amount for the founder’s son constitutes a fixed stipend, entirely unaffected by changes in prices, by increases or decreases in the value of the Egyptian pound, or by differences between periods.” This wording comes from the compilation’s principles, not from the answer, and must be read subject to that qualification. On the same issue, see “Waqf khayrī wa-istiḥqāqī maʿa tafsīr sharṭ al-wāqif” [A Charitable Waqf with Designated Beneficiaries, with an Interpretation of the Founder’s Stipulation], fatwā of Ḥasanayn Muḥammad Makhlūf, Shawwāl 1367 AH / 18 August 1948, location 2904, which considers the currency current in each period. ↩

  9. The two cautions stated in the design ruling are: regular provisions in different riwāqs must not be presumed equivalent in their stipulations or substantiation; nor must the transformation be presumed linear, since regular provision, salary, allowance and free service may overlap for the same person. ↩

7 · From the Document to Central Registration: Tracing Movement Towards Administration, Not Claiming a Transformation

Methodological disclosure of the limits of the design:

The limit of this section is explicitly stated in the design ruling: what is traced here is the movement of the record of deeds and rights towards modern administration. It does not claim that an administrative official’s file came to record individual beneficiary eligibility, or that it imposed on the beneficiary the burden of correction and litigation. That proposition remains contingent on an administrative evidentiary gateway whose three required materials have not been assembled: an instrument identifying a beneficiary, an instance of amendment or refusal of registration, and a route for correction.

The first turning point: the relationship between form and existence (1935)

The Detailed Survey and Registration Agency asked who was entitled to payment for land incorporated into a public utility project, where its owner had dedicated it as waqf to a mosque through a private instrument whose date had been established before the Manūf Native Court in 1918, without formal sharia attestation. In 1935, Shaykh ʿAbd al-Majīd Salīm held that the waqf was valid, and that the requirement of attestation under Article 137 of the Regulations for the Organisation of the Sharia Courts was “a condition for the hearing of a waqf action in the event of denial, not a condition for the validity of the waqf itself”53. Documentary formality here is a procedural condition for judicial protection, not a substantive condition of existence. A land registration system reverses this distinction when registration becomes constitutive rather than declaratory (this paper has examined no provision establishing that reversal, nor does the corpus contain an instance that substantiates it; it should therefore be read as pointing towards a development beyond what has been measured, not as a supported claim).

The contrasting Ottoman comparator (1922):

Alongside this turning point, a comparator from outside Egypt is recorded as a description of the rule’s structure, not as evidence included in the count. In 1922, in a series of numbered propositions published in al-ʿIrfān presenting waqf rulings under Ottoman law, Ḥusnī ʿAbd al-Hādī stated: “If the provisions of the waqf deed conflict with the imperial land-register entries, established dealings are examined, and whichever accords with longstanding dealings is applied.” For him, longstanding, settled dealings thus adjudicate between the sharia deed and the Ottoman state register54.

The four instances in which the modern registration apparatus needed the fatwā authority (1953–1962):

The corpus captures four instances in which the modern registration apparatus needed the muftī to determine what to register. All postdate the liquidation law:

The first instance (1953): four entitlement-holders sought to register their ownership shares in waqf properties after the promulgation of Decree-Law 180 of 1952. Their application was made “in accordance with the procedures and rules governing inheritance rights laid down in Law 164 of 1946 Regulating Land Registration”. Completing the procedures required each entitlement-holder’s portion to be determined, so the matter was referred to the muftī with a request to determine the portions according to the text of the waqf deed55. Modern land registration could not proceed until the deed had been read through jurisprudence.

The second instance (1955): the questioner explicitly states that “the Land Registration Agency in al-Minyā requested that a fatwā be obtained on the wife’s portion, so that the dissolution of the waqf could be registered on that basis”56. The administrative apparatus thus requests a fatwā as an instrument for implementing the registration procedure.

The third instance (1955): the Land Registration Office in Damanhūr wrote to ask about the legal characterisation of sums stipulated by a founder in a waqf. An “application for registration under no. 46 of 1955” had been submitted to the office to register it. The question was whether these were stipends in the sense to which the law attached its provisions, or non-recurring gifts. Shaykh Ḥasan Maʾmūn set out the conditions for a stipend. He held that the requirement of recurrence must be explicit and conclusive, and that an ambiguous provision could not be relied upon57. The four instances, with their correspondence and reference numbers, are detailed in Appendix 3.

The fourth instance (1962): an acknowledgement drawn up at the Land Registration and Notarisation Office in Cairo in 1952 was presented to the muftī, who was asked to explain its effect on entitlement58.

This is what the material establishes: records of waqf deeds and endowed property increasingly entered the sphere of modern administration through applications for registration and entries in the land register, while legal characterisation at critical moments remained in the hands of the old framing authority. During the decade of the liquidation of family waqf, the modern registration apparatus derived the content of what it registered from a reading of the deed, not from a general provision.

As for the “eligibility register of the individual beneficiary”—the form identifying that person, the decision removing them, and the route through which their entry could be corrected—the paper has not located its direct instruments. The subtitle was therefore formulated to match the evidence: “to the administrative register”, not “to the official’s file”. The limited administrative evidentiary gateway remains the declared programme of work for completing the second side of the comparison. This is a claim in the third evidentiary tier: it cannot be upgraded through an accumulation of indications within the corpus, but requires opening a source outside it59.

Footnotes

  1. Fatāwā Dār al-Iftāʾ al-Miṣriyya [Fatwās of the Egyptian Dār al-Iftāʾ], “Al-waqf bi-waraqa ʿurfiyya” [Waqf by Private Instrument], fatwā of ʿAbd al-Majīd Salīm, Ṣafar 1354 AH / 28 June 1935, location 2822; the questioner is the Detailed Survey and Registration Agency. The quotation comes from the answer. The year 1918 in the automated record must be identified as the date on which the private instrument’s date was established before the Manūf Native Court, as stated in the question, not the date of the fatwā. An earlier version of this text gave it as the date of the request for a fatwā; this was corrected. ↩

  2. Ḥusnī ʿAbd al-Hādī (Nablus), “Ḥuqūq al-awqāf” [The Rights of Awqāf], al-ʿIrfān [Knowledge], a series whose first instalment appeared in issue 7, 1 April 1922, pp. 394–396, and whose second appeared in issue 9, 1 June 1922, pp. 527–530. The quotation comes from p. 530. Its context concerns an imperial land-register entry lacking the restriction “generation after generation”, and the statement that “the imperial land-register entries are not applied; rather, the waqf deed is applied as before”. Conversely, if established dealings accord with the imperial land-register entry and conflict with a waqf deed that has not been followed, the entry governs. On the same page, Article 172 states: “In matters concerning waqf stipulations, regard is had to the founder, not to the scribe’s drafting.” It is compared with Article 1737 of the Majalla and Article 3 of the Law on Disposition: “Evidence contradicting the official entry—that is, the imperial land-register entry—shall not be heard.” The archive’s author field attributes the passage on p. 530 to ʿAbd al-Ḥamīd al-Rāfiʿī, but this is a linking error: the page bears the heading “Ḥuqūq al-awqāf”, and its waqf discussion is signed “Nablus / Ḥusnī ʿAbd al-Hādī”. What is attributed to al-Rāfiʿī consists of verses following it on the same page, signed “Tripoli in the Levant” (Ṭarābulus Shām). Attribution therefore follows the signature in the text, not the archive field. The setting is Ottoman and Levantine; it supplies no basis for inference about Egypt. This text enters none of the paper’s counts and does not raise section 7 beyond its declared limit. ↩

  3. The same source, “Waqf istiḥqāqī wa-kayfiyyat aylūlatihi ṭibqan li-qānūn 180 sanat 1952” [A Waqf with Designated Beneficiaries and Its Devolution under Law 180 of 1952], fatwā of Ḥasanayn Muḥammad Makhlūf, 8 March 1953, location 2748; the quotation comes from the question. Its heading gives “Jumādā al-Thāniya 1272 AH”, an evident error in the source that is recorded and not corrected without examining the printed edition. The Gregorian date is used here. ↩

  4. The same source, “Waqf istiḥqāqī” [A Waqf with Designated Beneficiaries], fatwā of Aḥmad Ibrāhīm Mughīth, Rajab 1374 AH / 19 March 1955, location 2842; the quotation comes from the question. ↩

  5. The same source, “Waqf istiḥqāqī shuriṭa bihi dafʿ mabāligh muʿayyana li-muʿayyan” [A Waqf with Designated Beneficiaries Stipulating Payment of Specified Sums to a Named Person], fatwā of Ḥasan Maʾmūn, Rabīʿ al-Thānī 1375 AH / 3 December 1955, location 2914; and letter no. 1824 from the Land Registration Office in Damanhūr, dated 15 October 1955, received under cover of a letter from the Damanhūr Sharia Court of First Instance dated 18 October 1955. The waqf concerned in the application was registered before the Sharia Court of al-Buḥayra Province on 24 March 1902. ↩

  6. The same source, “Ḥaqq al-rujūʿ ʿan al-waqf” [The Right to Revoke a Waqf], fatwā of Aḥmad Harīdī, Ramaḍān 1381 AH / 28 February 1962, location 2933. ↩

  7. See note 1. ↩

8 · The Shift in the Source of Evidentiary Authority Within the Fatwā: The Central Test

Although the sharia deed remained at the heart of the claim and did not recede, the reference point through which it was read was transformed. This was an institutional reversal measurable within the corpus itself, and it is the subject of the test conducted in this paper.

8.1 The Statistical Counting Procedure and Its Result

Each text was manually coded by its mode of inference into seven categories, set out in Appendix 1. The categories “modern law only” and “modern law interpreted through jurisprudence” were then combined into an indicator of reasoning based on codified law. The following are the results by decade, with their absolute counts

DecadeTotal instances (n)Reasoning based on codified lawJurisprudence with named sources (count and percentage)
1890511 (2%)8 (16%)
1900440 (0%)10 (23%)
1910571 (2%)17 (30%)
1920480 (0%)18 (38%)
1930561 (2%)26 (46%)
1940366 (17%)14 (39%)
19507762 (81%)4 (5%)
1960178 (47%)5 (29%)
1970430

Table 3: Mode of inference by decade for the three hundred and ninety instances dated using the first dating key (the six instances excluded from it are included in the count using the second key; the rows for the 1960s and 1970s describe thin samples and should not be read as indicating either decline or stability).

The two curves cross: reasoning based on law is virtually absent until the 1930s, then reaches seventeen per cent in the 1940s (17%) and eighty-one (81%) in the 1950s. Most instances in the latter decade rely on law alone, without an accompanying juristic derivation: fifty-one of sixty-two instances. By contrast, jurisprudence with named sources rises to its peak in the 1930s, at twenty-six of fifty-six, then collapses to five per cent (5%) in the 1950s. The peak of juristic documentation thus immediately precedes its replacement.

The gap tested in this paper has two defined endpoints. The first is the period 1930–1949, with ninety-two instances and a proportion of reasoning based on law of seven point six per cent (7.6%). The second is the 1950s, with seventy-seven instances and a proportion of eighty point five per cent (80.5%). The gap is therefore seventy-two point nine percentage points (72.9). All the following tests concern this gap60.

8.2 The First Falsifier: Is This an Effect of the Automated Dating Key?

Because the first year key was extracted automatically, it was liable to capture the year of an old deed mentioned in the question rather than the year in which the fatwā was issued. If the reversal in the inference curve resulted from this error, the claim would collapse entirely.

The entire calculation was therefore repeated using an independent second key: the date recorded in the fatwā’s own heading. The rule for converting its Hijri dates is set out in Appendix 1 (the date could be extracted for all three hundred and ninety-six instances). The two keys differed by more than two years in ten records. The most extreme cases were a fatwā assigned to 1895 by the first key whose heading gives 1953, and another assigned to 1914 whose heading gives 195161.

With the second key, the result held: reasoning based on codified law remained at zero in the first four decades, then rose to two per cent (2%) in the 1930s, fourteen (14%) in the 1940s, eighty-one (81%) in the 1950s, and forty-seven (47%) in the 1960s. Jurisprudence with named sources rose from fourteen per cent (14%) to forty-five (45%) in the 1930s, then collapsed to five (5%). The first falsifier was tested and did not overturn the claim; it made the legislative reversal sharper.

8.3 The Second Falsifier: Generational Shift or an Institutional Response?

The generational explanation attributes the rise in reasoning based on law to successive generations of muftīs: one trained only in jurisprudence, another trained in law. If this were correct, the reversal would be gradual and spread across the period, rather than tied to the year of a particular enactment.

The check was conducted at the level of the individual muftī. Shaykh ʿAbd al-Majīd Salīm is the most prolific muftī in the corpus, with ninety-four instances extending from 1912 to 1944. In forty-eight of these, or fifty-one per cent (51%), he reasoned through jurisprudence with named sources. He relied on codified law in only one instance. He thus represents the old juristic pattern in its purest form until the end of his tenure.

The decisive control here is a muftī whose tenure spans both decades. Shaykh Ḥasanayn Muḥammad Makhlūf himself crosses the 1946 divide: reasoning based on law accounts for forty-one point seven per cent (41.7%) of his twelve instances in the 1940s, rising to seventy-six per cent (76.2%) of his twenty-one instances in the 1950s. The same man changed his reference point without a change in the holder of the office. No muftī in the 1950s departs from this pattern: Shaykh Ḥasan Maʾmūn stands at eighty-nine per cent (89%) of thirty-five instances, Shaykh Aḥmad Ibrāhīm Mughīth at eighty-nine (89%) of nine, and Shaykh ʿAllām Naṣṣār at sixty-four (64%) of eleven. The leap is therefore neither the effect of an individual muftī nor the replacement of one person by another, but a comprehensive institutional response 62.

Correspondingly, reasoning based on law begins in the very year of its promulgation: two instances in 1946, both by Makhlūf and both based on law; then one in 1947 and two in 1948, all by him; then six in 1951, all by ʿAllām Naṣṣār; and twenty of twenty-two instances in 195563.

The shape of the annual series is itself evidence: if the curve reflected slow editorial drift, it would be smooth and gradual. What is observed is a sharp boundary between 1944 and 1946, with seven instances in the former year and none reasoning from law. The higher level then persists, without returning to its previous level64.

The apparent pattern is therefore a change in the working reference point at the moment of promulgation, rather than a slow generational shift. This suggests that the institutional explanation is more likely than the generational one. That judgement is explicitly qualified: few muftīs have instances spanning 1946, and the pre-1946 counts in their records are small. Ḥasan Maʾmūn has one instance before 1946, and thirty-six of forty-two after it. The individual-level check is therefore weaker than the year-level check. The preference for the institutional explanation rests on the coincidence of the leap with the enactment, supported by the Makhlūf control.

The year of the enactment is itself independently verified outside the corpus. Ibrāhīm al-Bayyūmī Ghānim states that the state intervened to establish rules for the waqf system itself “for the first time when it issued Waqf Law 48 of 1946”. He dates Decree-Law 180 of 1952 to 14 September 195265. The two turning points coinciding with the reversal are thus supported by a source independent of the measured texts. The wording of the individual articles, however, is drawn from their applications within the corpus, not from al-Jarīda al-Rasmiyya [Official Gazette].

A contemporary press account, published three years before the law, also describes the reform climate in which it emerged. Writing in an editorial in al-Risāla in 1943 on the appointment of a new minister, Aḥmad Ḥasan al-Zayyāt described the Ministry of Awqāf as containing “waqf deeds with their peculiar stipulations, transactions with their dubious forms, and systems that complicate the simple, impede the swift and make the easy difficult”, as well as “cases that delay the vindication of rights, complaints awaiting justice, and rights that deny their entitlement-holders”66. Its conceptual relevance is that the working reference point described on the eve of the law was the waqf deed’s stipulation, while the declared rationale for reform centred on entitlement-holders and their access to their rights. This supplies an indication from outside the corpus supporting the “institutional” rather than the “generational” explanation. Its limit must remain clear: it is a reformist press text from the public-opinion layer, not an instance of framing or material for counting, and it enters none of this paper’s measures.

8.4 The Decisive Test: Where Does the Law Number Appear—in the Question or the Answer?

After all the preceding tests, one question still distinguishes between two opposing explanations: did law enter the fatwā because questioners began asking questions in statutory language, or because the muftī began answering in it? The distinction concerns the layer of analysis, not merely degree. The first explanation implies a change in what reached Dār al-Iftāʾ; the second implies a transformation in the reference point for framing itself.

The separation was carried out across all the texts in the corpus. Each text was divided at the boundary between question and answer, and each part was searched separately for an explicit law number, using a strict pattern that did not capture a bare reference to an article. These are the results

DecadeNumber of instances (n)In the questionIn the answerIn the answer but not the question
1930562%0%0%
1940353%6%3%
19507513%80%67%
19601712%41%29%

Table 4: The location of an explicit law number within the structure of the text (the final column gives the proportion of texts in which the number appears in the answer but not in the question).

The implication of this table is that legal language enters the fatwā through the muftī, not through the questioner: in two-thirds of the fatwās from the 1950s, the muftī introduces law where the questioner has not. Its occurrence in the question remains at thirteen per cent (13%), less than a quarter of its occurrence in the answer. What is measured is therefore not the composition of incoming questions to Dār al-Iftāʾ, but the reference point for framing used in the answer. This is the most precise response to the contention that the count measures a change in the questions received.

8.5 Instrument Reliability: What Was and Was Not Done

The central variable in this section was coded manually by a single coder who knew the hypothesis, in batches ordered by period. This arrangement has been established to affect an adjacent field: the field for type of issue shares no label across the five batches. Was the central variable affected in the same way?

The procedure implemented to answer this question was a convergence check between two independent instruments. An automated instrument was constructed to classify an instance as relying on law, using formulations disclosed in Appendix 1. Its classifications were then compared with the manual coding across all three hundred and ninety-six instances. Agreement reached ninety-eight and a half per cent (98.5%), with a kappa coefficient of ninety-five hundredths. There were only six disagreements67. This entails that classification in this field depends on the reader’s subjective judgement only within narrow limits, and is reproducible using an instrument unaware of the hypothesis.

This is not termed a reliability test: reliability is measured between two human readers working on the same sample, blind to date and muftī, which was not done. What was implemented was a convergence check between an automated instrument and the reader, supporting the field for mode of inference alone. It provides no support for the fields for evidentiary tier, entitlement-holder category or type of issue; findings based on these fields must be read subject to this limitation. A reliability test between two readers takes priority over the other items in the testing programme. Moreover, the same reader worked on both sides of the declared dependence between this paper and the second paper of this series. No calculated agreement rate therefore compares this paper’s manual coding with its companion’s automated coding68.

8.6 What Remains Unresolved in the Testing Programme

The pivotal question remains unresolved pending completion of the administrative and judicial evidence: where the manifest meaning of a waqf stipulation explicitly conflicted with a new statutory provision, which settled the dispute, and how was that preference justified in juristic terms? The coded sample identified instances in which law settled questions arising from liquidation, including the devolution of the portion of a woman who died before becoming entitled, and the limitation of freedpersons’ entitlement69. Their exhaustive enumeration and an analysis of their reasoning remain items in the testing programme.

Footnotes

  1. See note 30. ↩

  2. The counting protocol for section 8. Coding was manual, using seven categories. The counts of instances dated by the first key across the nine decades are 51, 44, 57, 48, 56, 36, 77, 17 and 4. With the second key, the fatwā heading, the year could be extracted for all three hundred and ninety-six instances, distributed across the decades as 37, 58, 54, 50, 58, 36, 81, 17, 3 and 2, the last being the 1980s with two instances. The proportions reasoning from law under this key are zero, zero, zero, zero, 2%, 14%, 81% and 47%, followed by one of three in the 1970s and two of two in the 1980s. The proportions using jurisprudence with named sources are 14%, 22%, 31%, 36%, 45%, 42%, 5% and 29%, followed by one of three and zero of two. The two keys differ by more than two years in ten records. Instances relying on law are distributed across the first years after its promulgation as follows: two in 1946, one in 1947, two in 1948, six in 1951, six in 1952, seven in 1953, five in 1954, and twenty in 1955. ↩

  3. See note 30. ↩

  4. See note 61. ↩

  5. See note 30. ↩

  6. See note 25. ↩

  7. Aḥmad Ḥasan al-Zayyāt, “Wizārat al-Awqāf al-jadīda” [The New Ministry of Awqāf], al-Risāla [The Message], issue 526, Monday, 1 Shaʿbān 1362 AH / 2 August 1943, pp. 601–602. Both quotations come from the editorial. The first passage in full reads: “Shrines are not a word foreign to the language of awqāf: there are waqf deeds with their peculiar stipulations, transactions with their dubious forms, and systems that complicate the simple, impede the swift and make the easy difficult! There are also cases that delay the vindication of rights, complaints awaiting justice, and rights that deny their entitlement-holders.” The occasion was ʿAbd al-Ḥamīd ʿAbd al-Ḥaqq’s appointment to the Ministry of Awqāf. Attribution follows the article-author field in the periodicals archive, which in this journal identifies its proprietor and editor-in-chief. The same article estimates the total waqf holdings as “exceeding two hundred thousand feddans and thirty-six thousand properties”, with revenue amounting to “fifty million pounds dedicated as waqf”. These are press figures, not statistical figures: they enter none of this paper’s measures and are recorded here to convey the tone, not a quantity. The text belongs to the reformist public-opinion layer and cannot serve as evidence for either the framing layer or the operational layer. ↩

  8. A convergence check between the two instruments on the indicator of reasoning based on law; its automated formulation appears in the appendix. It was implemented across the entire corpus of three hundred and ninety-six texts, yielding agreement of 98.5% and a kappa coefficient of 0.95. Disagreements comprised four instances tagged as relying on law by the automated instrument but not by the manual coding, and two instances in the opposite direction. This is a convergence check between two instruments, not a reliability test between two readers. ↩

  9. See note 1. ↩

  10. Fatāwā Dār al-Iftāʾ al-Miṣriyya [Fatwās of the Egyptian Dār al-Iftāʾ], “Waqf istiḥqāqī ʿalā al-wāqifa wa-ʿalā ʿutaqāʾihā” [A Waqf with Designated Beneficiaries for the Female Founder and Her Freedpersons], fatwā of Aḥmad Ibrāhīm Mughīth, Ramaḍān 1374 AH / 18 May 1955, location 2846. ↩

9 · Who Addressed the Source Itself: From the State Apparatus to Individuals to the Region

The source used to measure the historical period itself changes during that period. This is an established historical fact, not a defect in the material. Its change was measured using two counters applied to the narrator (questioner) field, both with the dated instances in each period as their denominator:

The first counter: institutional questioners

Instances in which the questioner was an official body—from the Ministry of Justice, the Dīwān (bureau) of General Awqāf, judges and governorates—numbered thirty out of one hundred and six between (1894 and 1913), constituting twenty-eight per cent (28%) of all the material. This proportion then fell to zero out of twenty-one instances after 1960. (The decline was not monotonic: the fourth period exceeded the third in both count and proportion; the series and its counts appear in Appendix 7.)70

The second counter: questioners from outside Egypt

Among the twenty-one dated instances after 1960, eight questions came from outside Egypt. Their sources were Homs, Khartoum, Jerusalem, Kuwait, Bahrain, Taif and Malaysia, together with a request for a fatwā from an Israelite community in Egypt71. (These eight instances are not read as a percentage, because their denominator is a sample that cannot support proportional inference: the 1960s’ share of the corpus amounts to only thirty-seven hundredths (0.37) of that decade’s share of the published collection, and the 1970s’ share to only nineteen hundredths (0.19)72. These instances are therefore recorded individually and used to describe the widening range of questioners; no conclusive magnitude or trend is inferred from them.)

The measured trajectory thus has three phases, measured in the first two and described in the third. The first phase: a legal framing authority for the official state apparatus. The second phase: a legal channel for individuals, including entitlement-holders, nāẓirs, lawyers and women. The third phase: a regional authority which, at the end of the corpus, explains the legal basis for abolishing what it had previously framed.

Reading this trajectory requires allowance for five competing explanations that the corpus alone cannot resolve: changes in the policy for preserving and publishing fatwās; changes in the referring authority within the state; the effect of abolishing family waqf in drying up local material; the expansion of Dār al-Iftāʾ’s regional standing; and selection by the collection’s editors. This count alone gives no grounds for preferring any one of these explanations. What the evidence establishes is that those addressing the source changed, and that all subsequent inference must be read with this awareness. A change in those addressing it does not warrant an inference of change in the institution itself: that would move from a description of the record to a description of the institution without an evidentiary bridge.

Footnotes

  1. A count of who the questioners were, using the narrator field in the manual coding files. Both counters have the same denominator: dated instances. Instances with institutional questioners across the six periods numbered 30 of 106, then 17 of 123, then 4 of 52, then 5 of 40, then 2 of 48, then zero of 21. Questioners from outside Egypt in the period 1960–1980 accounted for eight of 21: Homs 1961, Khartoum 1962, Jerusalem 1965, Kuwait 1966, Bahrain 1967, Taif 1971, an Israelite community in Egypt 1973, and Malaysia (with a heading dated 1980). The coding batch for this period contains 27 instances, of which 6 have no year under the first dating key. These eight instances are not expressed as a proportion of their denominator, since the post-1959 interval cannot support proportional inference owing to its weak representation. They are therefore recorded individually with their locations, as description rather than magnitude. ↩

  2. See note 70. ↩

  3. See note 30. ↩

10 · Competing Explanations: What Has Been Ruled Out and What Remains

Four competing explanations for the paper’s central claim are presented here. The first is comprehensive and encompasses the remaining three, so it comes first:

10.1 The First, Comprehensive Explanation: A Biased Corpus or a Changing Structure?

The strongest objection is that this paper’s trends measure bias in the corpus rather than a transformation of the actual structure. The published fatwā collection is no neutral mirror: its texts grew longer, the composition of its issues and the identity of its questioners changed, and publication density varied from decade to decade. Perhaps the count measures the combined effect of these four factors.

All three factual premises of this objection are established: (1) the texts grew longer: median cleaned text length reached two thousand four hundred and eighty characters in the 1930s and three thousand six hundred and twenty-nine in the 1950s. (2) The composition of issues changed: the proportion of questions concerning devolution and transfer rose from forty-one per cent (41%) to seventy-nine (79%). (3) The identity of the questioner changed: the share of individuals rose from eighteen per cent (18%) to sixty-four (64%) and then to eighty-eight (88%)73.

The conclusion, however, does not necessarily follow from these premises. Testing was carried out along all four axes, against the decisive gap of seventy-two point nine points (72.9). The result is that they explain none of it:

The first axis, length: the combined instances from the two periods were divided into quartiles by text length, and reasoning was compared within each quartile. The gap persisted in every quartile, ranging from sixty-five to eighty-three points. Normalising the distribution of the 1950s to that of the preceding period raises the percentage relying on law from eighty point five to eighty point nine: length thus explains less than zero of the gap.

The statistical sensitivity test indicates that producing this jump through length alone would require median length to increase approximately seventy-eight times; the observed increase was one and a half times. An indicator wholly unaffected by length settles the issue: citation density per ten thousand characters. The density of statutory citations increased by a factor of twenty and a half between the 1930s and the 1950s, while the density of citations to named juristic sources fell by sixty-four per cent (64%) over the same period. A single measure of length cannot produce two opposing movements in the same texts74.

The second axis, the composition of issues: this was the control announced in the design but impossible to implement using the field for type of issue. That field was labelled by temporal batch, with zero shared labels across the five batches, making it impossible to separate type from period using that instrument. The control was implemented through three routes independent of that field, and the competing explanation was ruled out in each:

  • First: holding constant the core family of issues—devolution, interpretation of the stipulation and the office of nāẓir—which accounted for between seventy and ninety-one per cent (70% and 91%) of each decade. Within this family alone, the percentage relying on law rose from zero in the 1930s to eighteen in the 1940s and eighty-two and a half in the 1950s.

  • Second: holding entitlement-holder category constant. Among individually designated and descendant entitlement-holders alone, the proportion rose from zero (0%) to twenty-seven (27%) and then approximately eighty-three (83%).

  • Third, the most demanding test: excluding every text whose question mentions a law or liquidation. The gap persisted between eighty-eight instances with a rate of four and a half per cent (4.5%) and sixty-five with a rate of seventy-eight and a half (78.5%): a gap of seventy-three point nine points, larger than the raw gap75.

The premise is therefore correct and the conclusion invalid: the composition of issues changed, but it explains none of the gap; holding it constant actually widens the gap. (This is qualified by a limitation of the measurement instrument: combined four-way normalisation produces a gap fourteen per cent (14%) larger than the raw gap. Yet thirty-eight weighting units out of ninety-two in the earlier period have no comparator in the 1950s. The two-way and three-way normalisations are therefore the ones relied upon; the four-way result is reported with its qualification.)76

The third axis, questioner identity: the objection would imply that reliance on law increased because questioners had become bodies that spoke its language. The observation is precisely the opposite: bodies using legal language receded while reliance on law rose, with the combined share of courts and ministries falling from approximately thirty-six per cent (36%) to six (6%).

Among individual questioners alone, the percentage relying on law rose from zero (0%) in the 1930s to twenty (20%) in the 1940s and approximately eighty (80%) in the 1950s. Cross-tabulating who asked with what was cited leaves no category unaffected: the rate for individuals rose from twelve to eighty (12% - 80%), for courts from eight to seventy-five (8% - 75%), and for unspecified questioners from six to eighty-three (6% - 83%)77. Added to this is the decisive test presented in (8.4): law enters through the muftī rather than the questioner in two-thirds of the fatwās of the 1950s. The third axis is therefore not merely ruled out; it operates in the direction opposite to the objection.

The fourth axis, publication density and the representation of periods: if the trend were merely an effect of publication density, the two would be positively correlated. What is observed is a negative correlation: minus twenty-six hundredths (-0.26) between the number of texts in a decade and the proportion relying on law, and minus forty-nine hundredths (-0.49) between the waqf share of that decade’s fatwās and the proportion relying on law. The weights of the decades do not affect the proportion relying on law at all, since it is calculated separately within each decade. The result’s robustness to the sample was measured: excluding the whole of 1955—the most densely represented year in the corpus, with twenty-two instances—leaves the rate for the 1950s at seventy-six point four per cent (76.4%), rather than eighty point five (80.5%)78.

Here the competing explanation has one genuine area of applicability: the “post-1959 tail”. The representation of decades varies sharply relative to their share of the published collection: over-representation of the 1890s by approximately four times, and under-representation of the 1960s at thirty-seven hundredths (0.37) and the 1970s at nineteen hundredths (0.19). Given the small sample—seventeen texts followed by four—this tail cannot support an interpretation of direction. The values of forty-seven per cent (47%) and seventy-five (75%) in Table 3 describe a thin, selected sample, rather than recording either decline or stability. Every claim concerning this interval belongs to the third evidentiary tier7980.

The judgement on the comprehensive competing explanation is therefore: partially ruled out. It is ruled out for this paper’s principal trend—the shift in the reference point for framing—since the four axes together explain approximately zero of its gap, and the gap actually widens after normalisation. It remains applicable only to the post-1959 tail.

10.2 The Second, Third and Fourth Explanations: Their Results

After the comprehensive explanation was ruled out, three subsidiary competing explanations remained. Their results, with the full account and figures, are detailed in Appendix 9:

(1) The editorial layer: this was addressed through four methodological procedures. The most rigorous separated the count of the opening formula (“We have examined”) from the count of document mentions. The two trends differed in both peak and direction, ruling out the strong version of the explanation while leaving the weak version standing: rejected as an explanation of direction, still applicable to magnitude. (2) The composition of issues in the document trend: an initial check using a disclosed lexical flag found no difference between instances flagged for liquidation and other instances within the same period. Its limitation is that the flag is coarse and cannot substitute for coding. (3) Generational shift was ruled out by controlling for the same muftī serving across both decades. It retains a limited role in the speed with which the new form became established, which is an interaction rather than an alternative explanation.

A fifth qualification is required here, concerning the thesis itself rather than its explanation: the paper describes a shift in the reference point for framing within the muftī’s answer, an explicitly declared single layer of analysis. It does not necessarily follow that the judiciary applied the new reference point in the same way, or that the executive administration made disbursements accordingly. Inference from the “framing layer” to the “operational layer” is impermissible without an evidentiary bridge.

Footnotes

  1. See note 30. ↩

  2. See note 30. ↩

  3. See note 30. ↩

  4. See note 30. ↩

  5. See note 30. ↩

  6. See note 30. ↩

  7. See note 1. ↩

  8. See note 30. ↩

11 · Constrained Comparison

The following table brings together the three institutional powers across the three systems. Its final column records only what the material, within its limits, supports:

Institutional powerJudicial document system (classical)Transitional central registration (measured)Subsequent administrative file (apparatus)
Definition of legal positionThe founder’s text; the muftī frames it and the judge adjudicates [documented through instances]The text remains; codified law competes as a reference point [documented through an executed count]Statute and regulations; details unknown
Preservation and amendment of the recordA sharia deed in the court record office; amendment through attestation before the judge [documented]An application for land registration; a property register entry; a fatwā is sought on legal characterisation [documented through four instances]Unknown: the instrument for identification and updating has not been documented
Burden of detecting errorThe entitlement-holder brings an action against the nāẓir and calls him to account; limitation bars the hearing of an action without extinguishing the right [prescribed in a fatwā]Mixed: actions persist while administrative decisions expand [partially documented]Unknown: the route for correction has not been documented

Table 5: The three powers across successive systems. (The grade in square brackets refers to the evidence in the cell, not the strength of the right it describes; the final column expressly acknowledges its limits.)

What, then, does the executed count suggest? Two findings at the second evidentiary tier. First, the sharia deed and its official copy remained central to the production of claims through the end of the measured interval; indeed, invocation of documents peaked in the decade of liquidation itself on both measures. Second, the reference point for framing shifted from the stipulation and jurisprudence with named sources to codified law: a sharp shift in the year the 1946 law was enacted, and then in the decade of the 1952 liquidation. It holds under two independent dating keys and after testing the comprehensive competing explanation along all four axes. (The change in who addressed the source is measured in section 9, but five competing explanations remain unresolved there, so it is not counted among the tested findings.)

What remains an empirical question? The fate of the second and third powers after the administrative apparatus was fully established: who had the power to control the individual beneficiary’s eligibility file? Who updated and withdrew it? Who bore the burden of correcting it? By its explicit commitment, this paper does not answer that question. Nor does it infer from the silence of its material either weakness or strength in the beneficiary’s subsequent evidentiary authority. It passes the question to the appropriate avenue of investigation, with the instruments needed to answer it specified.

11.1 The Replacement Test: The Third Comparator

The result first: this paper’s claim does not collapse when its direct comparator replaces waqf, and Islamic waqf is not distinct in this conceptual respect from its comparator within the family of “purpose institutions” in comparative law.

The comparator put in place of waqf here is the private purpose foundation under the German Civil Code (BGB) (Stiftung bürgerlichen Rechts). Its methodological neutrality has two grounds:

First: it is not a party to this paper’s comparison.

Second: it is a standard institution within its system, not a marginal exception: the Civil Code itself regulates it in §§ 80–88, and foundations are subject to mandatory administrative supervision.

The test consists of putting the comparator in place of waqf in the two tested claims, then examining whether they remain valid81:

The first claim—the shift in the reference point for framing—is reproduced by the comparator. The 2021 Act on the Unification of Foundation Law [Gesetz zur Vereinheitlichung des Stiftungsrechts und zur Änderung des Infektionsschutzgesetzes], effective from 1 July 2023, changed both numbering and substance, so that a reference to § 85 before the reform denotes something different from a reference after it. The reference point of the answer thus shifts to the provision identified by its number, in the manner measured by this paper. The limitation is explicit: the inference concerns the change in numbering, not the provision’s substance. The current § 85 concerns the conditions for amending the charter, not the beneficiary’s legal position. The latter is addressed in the current § 86 no. 4, which supplies evidence for the second claim.

The second claim—that the document remains central to the claim—is reproduced by the comparator in wording more explicit than that in our material. § 86 no. 4 requires, when one foundation is absorbed into another, protection of the rights of those on whom the transferring foundation’s charter conferred claims to its benefits, expressly grounding entitlement in the document. § 83(2) requires both the foundation’s governing bodies and the supervisory authority to respect the founder’s will. In 2016, the Federal Court of Justice held that claimability turns on what the founder established in the founding instrument or charter: a right exists where the document sets objective criteria through which standing is acquired without discretionary selection, and does not exist where selection is left to one of the foundation’s governing bodies82. This is precisely what our corpus shows in contrasting the individually designated entitlement-holder with the open public.

The rule underlying this paper—that “the founder’s stipulation is like the Lawgiver’s text”—therefore carries no exceptional distinctiveness for waqf. The German comparator reproduces it in its provisions and binds the supervisory authority itself to it.

The position of this judgement on the map must be specified, rather than left as an unqualified negation. Reproduction is established within the “Romano-Germanic family” by two cases, not one: Germany’s § 83(2) binds both governing bodies and the supervisory authority; and the case law of the French Council of State prevents the administration from interfering with the donor’s stipulations, allowing a stipulation to be changed only through a civil judge’s ruling preceding approval of the amendment.

A system from a “third family” differs from both. Section 28(1) of the Tamil Nadu Hindu Religious and Charitable Endowments Act 1959 binds only the trustee and places the terms of the trust alongside the usage of the institution and the directions of the competent authority; it contains no provision requiring the supervisory apparatus to respect those terms. It follows that what is measured on this axis is waqf’s position among three families, not its uniqueness in relation to them, and that the distinction is an association with legal family, not a causal explanation. Waqf sits at the top of the hierarchy of the founder’s stipulation alongside the German and French institutions, not outside it.

The second piece of evidence is reinforced by a corresponding procedure not previously recorded: § 86e(2) requires the authority to hear those holding claims created by the charter at least one month before a decision on absorption or merger, and to alert them to its consequences. The document thus creates not only the right but also procedural standing before the authority. This strengthens the claim that the document remains central to the claim, without altering its limit.

The replacement test has two limitations. First, it concerns the structure of the rule rather than a measured series: no comparable count corresponds to this paper’s corpus. Second, one difference emerged outside what was tested; it is recorded without being used as a basis for inference. In our material, the entitlement-holder litigates against the nāẓir and calls him to account, whereas the Federal Court of Justice held that the Civil Code’s provisions grant the beneficiary no supervisory powers over the foundation. This difference concerns the third power in Table 5, the power whose final column acknowledges that it is unknown.

This difference must remain within its proper limits: the negation concerns powers over the foundation, not standing before the authority. Beneficiaries’ standing in matters of supervision varies across systems of purpose institutions; it is not absent. (In the German system, it is absent in relation to the foundation but exists as a right to be heard before the authority under § 86e(2); in Indian law, it exists as a right for any interested person to object to the decision and challenge it judicially, under sections 34(1) and 34(4) of the Tamil Nadu Act.) Beneficiary silence is therefore not a necessary feature of the purpose institution, but a design choice with an operational alternative in a system currently in force. No judgement on distinctiveness is based on this here, since it concerns the third power, which falls outside what this paper tested.

Footnotes

  1. The material for the replacement test consists of the provisions governing the private purpose foundation in the German Civil Code (BGB). Their German wording is reproduced in full in Appendix 8 of this paper, including: § 80(1), on the foundation as a legal person without members; § 80(2), requiring administrative recognition for its creation; § 81(1), requiring the founder to establish a charter specifying the purpose; § 83(2), requiring both the foundation’s governing bodies and supervisory authorities to respect the founder’s will; § 83c(1), requiring the principal of the capital to be preserved undiminished; and § 86 no. 4, protecting, upon absorption, the rights of “persons on whom the charter of the transferring foundation has conferred enforceable claims to benefits from the foundation”. The version reproduced is that in force since 1 July 2023 pursuant to the Act on the Unification of Foundation Law of 16 July 2021 (BGBl. I S. 2947). The reform changed both numbering and substance, so every pre-2023 reference to § 85 denotes a different provision. The judicial rule comes from the Federal Court of Justice’s judgement of 15 December 2016, case I ZR 63/15, headnotes (a) and (b) and paragraphs 22 and 28, including: “What is decisive … is what the founder established in the founding instrument or charter”, and “The Civil Code’s provisions on foundations do not grant beneficiaries … any legal position in the sense of powers resembling those of members or supervisory powers”. A limitation stated at the beginning of Appendix 8 must be noted: access to the official text at gesetze-im-internet.de and to the official gazette was unavailable. The wording of the provisions was therefore obtained from a consolidated mirror of federal legislation (buzer.de), which records each provision’s authority for entry into force from the official gazette. The evidentiary status of these texts is that of a reliable mirror supported by references to the official gazette, not the official gazette itself. The government bill (Drucksache 19/28173) is not used as a source for the wording of any provision, since the enacted law differs from the bill in places; it is consulted only to illuminate the rationale. The Federal Court of Justice’s judgement, by contrast, was examined in its official text and is a primary source for the judicial rule alone. No quantitative count comparable to this paper’s count was undertaken for the comparator: the test concerns the structure of the rule, not a measured series. ↩

  2. See note 81. ↩

12 · Declared Limitations

1. Type of contribution: a completed test of exactly two claims: (the reversal in the reference point for framing and the continued centrality of the document to claims). Everything else in this paper is either a description supported by documented instances or a declared limitation.

2. Analytical layer: the paper is confined to the legal framing layer; rates of access, success and actual disbursement remain dependent on archival material beyond the sources examined.

3. Release cited: references are exclusively to the approved frozen release WEC-432/1.0, containing three hundred and ninety-six (396) accepted texts, and every aggregate figure is calculated from that release alone. The difference between the two counts is stated explicitly: (405), four hundred and five, is the candidate layer before coding; the approved count after coding is three hundred and ninety-six.

4. Selection rule: the corpus exhaustively enumerates a published collection, and its bias operates in one known direction: enhancing the apparent extent of the documentary system and exaggerating the difference. The data therefore show that the instrument was available to those who pursued this route, not that it was automatically widespread among entitlement-holders. Every difference presented in this paper’s comparisons is an upper bound, not an estimate.

5. The corpus selection funnel and its limitations: the transition from the third to the fourth layer follows a rule known in textual form but not reproducible through automated execution. The judgements for the sixty-record precision-test sample have not been preserved; nor have the identifiers for the first sample used to measure the miss rate, whose zero result has been rejected. The known minimum miss rate is approximately one and a half per cent (1.5%), and the likely figure is approximately three (3%). This is a known lower bound, not an estimate.

6. Distinctness of instances: the identifiers establish with certainty (394) distinct instances (four hundred less six); the other two duplicate pairs have not been identified. The approved count is three hundred and ninety-six units of description.

7. Category: the corpus’s principal constituency is “named entitlement-holders and descendant entitlement-holders”, who account for forty-nine per cent of it (49%). Their position cannot support a judgement about the open public, represented by seven instances.

8. Counting and dating procedures: counting in section 4 is automated and lexical; in section 8 it is manual and conceptual. Figures in both sections are calculated using the year extracted from the text and recalculated using the fatwā-heading dating key. Six instances without a year under the first key are excluded from its chronological count, and the two keys differ by more than two years in ten records. Every length figure is calculated from the cleaned text, since the recorded length field compresses differences between periods to approximately one third of their size.

9. Number of comparators in the replacement test: the test in 11.1 used one neutral comparator from one family (the German foundation). Its finding was then compared with a comparator from a third family (Hindu religious and charitable endowments) and with a control from the first comparator’s family, solely on the axis of the rank accorded to the stipulation. A finding concerning one text cannot therefore be extended to its family without a second source of evidence from within that family, nor can it be generalised to all systems of purpose institutions.

The third family was not put in place of waqf on the paper’s other axes of analysis; this is declared as a limitation and is not estimated. A further limitation concerns the evidentiary record: neither the French comparator’s text nor that of the third-family comparator is recorded in this paper’s bibliography, neither has a footnote at its point of use in 11.1, and neither original was examined here. The two sources of evidence should therefore be read as descriptions of the structure of the rule drawn from the examination of comparators, not as source attributions verified in this paper.

10. Measurement instrument: manual coding was undertaken by one person who knew the hypothesis, in batches ordered by period. No reliability test between two readers was conducted. What was conducted was a convergence check against an automated instrument for the mode-of-inference field alone, with agreement of ninety-eight and a half per cent (98.5%); it provides no support for the other subsidiary fields. The issue-type field is tagged by temporal batch and cannot serve as a control for the composition of issues; that control was implemented using instruments independent of it.

11. Editorial layer: a difference between the muftī’s answer and the compiler’s title and principles is established in at least one instance: the title of the fatwā on limitation states its rule without qualification, whereas the answer restricts it by reference to possession and ownership. Comprehensive comparison against the printed edition remains a future task.

12. The surviving competing explanation: the comprehensive competing explanation has been ruled out for the central trend along four completed axes, but remains established in the post-1959 tail alone. The editorial layer affects the magnitude of the two trends, though not their direction. The possible effect of the composition of issues on the document trend has been narrowed by a preliminary, coarse check that does not substitute for coding.

13. Temporal scope of inference: measured statistical inference, and what it makes likely, ends in 1959; what follows is a description of a thin sample: (the representation of the 1960s is thirty-seven hundredths (0.37) of their share in the collection, and that of the 1970s nineteen hundredths (0.19); their texts number seventeen (17) and then only four (4), respectively).

14. Declaration of dependence: this is stated explicitly in item seven of the paper’s statement and in the Common Appendix. It establishes that this paper and “The Entry Standard: Measured Content, a Hypothesised Source” are two analyses dependent on a single evidentiary base, and that their statistical agreement does not constitute independent corroboration83.

15. Sources not searched where relevant evidence would be expected: the paper explicitly identifies the sources that were not searched: (the Ministry of Awqāf’s internal regulations and reports, appointment files at the National Archives, issues of the Official Gazette, the original al-Fatāwā al-Khayriyya [The Khayriyya Fatwās], and al-Baḥr al-rāʾiq [The Clear Sea] on the question of valid stipulations). What was not found here was not sought in its expected location and cannot count as a negative result. (The original work by al-Khaṣṣāf, however, was searched and the matter resolved; the finding is recorded in the footnote tracing the quotation.)

Footnotes

  1. See note 1. ↩

13 · Conclusion

This paper’s central question is: what happened to the establishment of waqf beneficiary eligibility between the scenes of 1894 and 1980? The answer the evidence can sustain is that institutional change occurred in one of two layers:

1. At the layer of the reference point (the reversal in the reference point for framing): measurement establishes a sharp reversal. The muftī’s answer came to rest on a statutory provision cited by number, where it had previously rested on the founder’s stipulation and quotations from jurists. This was measured across the entire corpus and repeated using a second, independent dating key. Generational shift was ruled out as an explanation by controlling for a single muftī serving across both decades, such as Shaykh Ḥasanayn Makhlūf; that shift retains a limited role in the speed with which the new form became established. Framing was then tested against a comprehensive competing explanation along four axes, which explained none of the gap. The direction of the effect is established by the fact that legal language enters the fatwā through the muftī rather than the questioner in two thirds of the fatwās of the 1950s (67%).

2. At the layer of the instrument (the continued centrality of the document to claims): the sharia document did not retreat before codification. The waqf deed and its official copy remained in the claimant’s hands, and their invocation peaked in the very decade of liquidation under both measures, the binary indicator and the indicator adjusted for body-text length. The land registration apparatus continued to seek fatwās on the legal characterisation of what it registered. What changed, therefore, was not the instrument supporting the right, but the law through which that same instrument was interpreted.

A third comparator from the waqf’s family was then put in its place: the purpose foundation under the German Civil Code (BGB). Neither claim collapsed under replacement. The comparator’s provisions reproduced the rule supporting both claims, and its § 86e(2) reproduced procedural standing for the claimant against the authority.

The position of this finding on the comparative map is stated explicitly: the founder’s stipulation ranks highly in the Romano-Germanic family, supported by two sources of evidence, German statutory text and French case law. It ranks lower in a system from a third family, the Tamil Nadu Hindu Religious and Charitable Endowments Act, which binds the trustee alone. What is measured is the waqf’s position among three families, not its uniqueness relative to them. The difference on this axis is an association with legal family, not a causal explanation. Establishing commonality where uniqueness is presumed contributes to dialogue between East and West; it does not constitute a failure of that dialogue, since it challenges the binary division on which much comparative writing rests. The detail appears in 11.1.

Two limitations apply to both findings. First, the editorial layer affects the magnitude of the two trends, even though it has been ruled out as an explanation of their direction; its contribution cannot be estimated before comparison with the printed editions. Second, the measured period ends in 1959. What follows is a description of a sample that supports neither a claim of decline nor a claim of stability.

The implication for interpreting the change is that the familiar formulation—a transition of waqf from documentation to neglect—does not describe what happened in this material. More accurately, the document remained in place while the authority with the power to interpret it changed. This opens the next paper’s question rather than closing it. If the reference point for interpretation shifted, who came to have power over the entry itself, and who bears the burden of correcting it?

This paper opens two avenues of inference for subsequent work. The first is the “paper on entry standards”, for which the corpus has documented standards relating to institution, poverty, conduct and non-marriage, with supporting cases; its quantitative support on the source axis, however, comes from this paper’s counter, not an independent one. The second is the “paper on claimability”, for which the classical side of the comparison has been documented: an entitlement-holder arguing against the Ministry on the basis of his document; limitation barring the hearing of an action without extinguishing the right, qualified in the answer by possession and ownership; maintenance taking priority over creditors; and a regular provision whose holder can choose to receive it in kind or in cash.

Sources and References

First: Principal Source

— The Waqf Entitlement Corpus (1894–1979 CE / 1313–1400 AH), approved release WEC-432/1.0, frozen on 28 July 2026 with three hundred and ninety-six texts: Fatāwā Dār al-Iftāʾ al-Miṣriyya [Fatwās of the Egyptian Dār al-Iftāʾ], compiled by al-Maktaba al-Shāmila from the authority’s publication, book number 432 in al-Shāmila’s numbering, comprising ten volumes. Identifiers are fixed in the form F432-NNNNN, and their stability is anchored by the hash of the original pages file. Each citation gives the fatwā’s title, muftī, heading date and location in the compilation’s numbering. The bibliographical details of the reference printed edition—publisher, place, year and edition number—were not examined in this paper; establishing them remains a declared task.

— Sijill qumʿ al-mudawwana wa-l-iṣdār al-mujammad [Record of the Corpus Selection Funnel and Frozen Release]: an independent document describing the five layers, their counts and entry rules, the exclusion table with its identifiers, and what is declared unknown (2026 CE).

Second: Juristic Sources

— Abū ʿUmar Dubyān ibn Muḥammad al-Dubyān, al-Muʿāmalāt al-māliyya aṣālatan wa-muʿāṣaratan [Financial Transactions: Classical Foundations and Contemporary Practice], 2nd ed., 1432 AH, twenty volumes, pp. 507–508 (quoting Ibn ʿĀbidīn’s gloss on the meaning of the maxim concerning the founder’s stipulation, with references to the gloss at 4/366 and 4/434). Only this passage was examined for this paper; the place of publication and publisher are therefore not recorded, and the edition of Radd al-muḥtār [The Reply of the Perplexed] was not examined at these two precise locations.

— Ibn ʿĀbidīn, Ḥāshiyat Radd al-muḥtār ʿalā al-Durr al-mukhtār [The Reply of the Perplexed: A Gloss on The Choice Pearl] (al-Ḥalabī edition), “Maṭlab fī qawlihim sharṭ al-wāqif ka-naṣṣ al-shāriʿ” [Discussion of the Maxim That the Founder’s Stipulation Is Like the Lawgiver’s Text] and the related discussions through “Maṭlab al-jāmakiyya fī al-awqāf” [Discussion of Stipends in Awqāf] (containing excerpts from the base text and the commentary on them, the passage from al-Fatāwā al-Khayriyya stating that what matters is the actual situation rather than what is written, and the quotation from the scholar Qāsim on unequivocal, manifest, equivocal and indeterminate wording). The edition number, year and place of publication are not recorded in the copy examined; the passage is therefore identified by the discussion title.

— Ibn ʿĀbidīn, Ḥāshiyat Radd al-muḥtār ʿalā al-Durr al-mukhtār, the passage examined in the six-volume compilation, p. 413, corresponding to the passage al-Zuḥaylī cites as 3/441–444 (containing the rule concerning the judges’ registry, attributed to al-Fatāwā al-Khayriyya; the statement “the documentary instrument, however, cannot serve as evidence”, attributed to al-Khāniyya [The Khāniyya Fatwās] and al-Isʿāf [Assistance]; the rank accorded to the established practice of the administrators; and the requirement that the documentary instrument agree with the judicial register). The edition number, year and place of publication are not recorded in the copy examined; the passage is therefore identified by its continuous pagination.

— Ibn Nujaym, al-Ashbāh wa-l-naẓāʾir [Analogues and Parallels], rule 27 among the rules of the Book of Waqf; examined in the edition with commentary: Aḥmad ibn Muḥammad Makkī al-Ḥamawī, Ghamz ʿuyūn al-baṣāʾir fī sharḥ al-Ashbāh wa-l-naẓāʾir [Sharpening the Eyes of Insight: Commentary on Analogues and Parallels], Beirut: Dār al-Kutub al-ʿIlmiyya, 1st ed., 1405 AH / 1985, 4 vols., p. 228 in the compilation’s continuous pagination (where Ibn Nujaym’s text is numbered and al-Ḥamawī’s commentary follows beneath a divider; the two must not be conflated).

— Al-Khaṣṣāf, Abū Bakr Aḥmad ibn ʿAmr al-Shaybānī, Aḥkām al-awqāf [Rulings Governing Awqāf], the compilation comprising one hundred and thirty-nine sections. Its entire contents list was examined, and four matching searches restricted to the book were conducted. The registry rule attributed to it in modern compilations did not appear. It is recorded here as an examination yielding a negative result.

— ʿAbd al-Majīd ibn Muḥammad al-Subayyil, Qāʿidat sharṭ al-wāqif ka-naṣṣ al-shāriʿ: dirāsa fiqhiyya muqārana [The Maxim That the Founder’s Stipulation Is Like the Lawgiver’s Text: A Comparative Juristic Study], the first discussion, on the Ḥanafī school. Neither the edition nor the page number was examined; the passage is therefore identified solely by its discussion title. Examination of the original removed the need to rely on this intermediary for the quotation from Ibn Nujaym, but that reliance remains for the discussion of al-Baḥr al-rāʾiq.

— Wahba al-Zuḥaylī, al-Fiqh al-Islāmī wa-adillatuhu [Islamic Jurisprudence and Its Evidence], Damascus: Dār al-Fikr, revised fourth edition, ten volumes, p. 7667 in the compilation’s continuous pagination, chapter five, on substantiating waqf under sharia and statute (containing the statement “a written documentary instrument cannot serve as evidence” and the rule concerning entries in judges’ registers, attributed to al-Khaṣṣāf, with references to Radd al-muḥtār 3/408 and 441–444). The edition’s year and the volume number are not recorded in the copy examined. Its attribution of the rule to al-Khaṣṣāf is invalidated by examination of both originals: Radd al-muḥtār attributes it to al-Fatāwā al-Khayriyya, while al-Khaṣṣāf’s Aḥkām al-awqāf does not contain it. The detail appears in the footnote tracing the quotation; this source is therefore not relied upon to attribute juristic quotations to their authors. It should also be noted that this source dates Egyptian Waqf Law 48 to 1952, contrary to the references in the corpus’s instances and Ghānim’s account; it is therefore not relied upon for legislative facts.

— Mudawwanat aḥkām al-waqf al-fiqhiyya [Compendium of Juristic Rulings on Waqf], 3 vols., 1st ed., 1439 AH / 2017, pp. 240 and 215 in the compilation’s continuous pagination. The place of publication, publisher and volume number are not recorded in the copy examined.

Third: Contemporary Sources and Legislative Texts Cited

— ʿAlī Mubārak, al-Khiṭaṭ al-Tawfīqiyya al-jadīda li-Miṣr wa-l-Qāhira wa-mudunihā wa-bilādihā al-qadīma wa-l-shahīra [The New Tawfīqī Topography of Egypt, Cairo, and Their Ancient and Celebrated Towns and Places], 20 vols., Cairo: Egyptian National Library and Archives, revised 2nd ed., 1425 AH / 2004.

— Aḥmad Amīn, Qāmūs al-ʿādāt wa-l-taqālīd wa-l-taʿābīr al-Miṣriyya [Dictionary of Egyptian Customs, Traditions and Expressions], Cairo: 1953 CE, Hindawi edition.

— Muḥammad ʿAbduh, “Shurūṭ al-wāqifīn” [Founders’ Stipulations], al-Manār [The Beacon] (Cairo, proprietor Muḥammad Rashīd Riḍā), issue 6, 22 June 1902 CE, pp. 201–217. Attribution follows the article-author field in the digital periodicals archive; the passage cited is on pp. 210–211.

— Aḥmad Ḥasan al-Zayyāt, “Wizārat al-Awqāf al-jadīda” [The New Ministry of Awqāf], al-Risāla [The Message] (Cairo, proprietor and editor-in-chief Aḥmad Ḥasan al-Zayyāt), issue 526, 2 August 1943 CE, pp. 601–602. Attribution follows the article-author field in the digital periodicals archive, which here agrees with the journal’s proprietor.

— Ḥusnī ʿAbd al-Hādī, “Ḥuqūq al-awqāf” [The Rights of Awqāf], al-ʿIrfān [Knowledge] (Sidon, proprietor Aḥmad ʿĀrif al-Zayn), a series in issue 7, 1 April 1922 CE, pp. 394–396, and issue 9, 1 June 1922 CE, pp. 527–530. The passage cited is on p. 530; attribution follows the signature in the text, not the archive field, as explained in its footnote.

— Legislative texts cited in the corpus’s instances: Waqf Law 48 of 1946; Law No. 164 of 1946 on the Regulation of Real Estate Registration; Decree-Law 180 of 1952; the Regulations for the Organisation of the Sharia Courts and Related Procedures (Article 137); and Courts Regulation 31 of 1910 (Article 376). The substance of their provisions is drawn from their applications within the corpus. Issues of the Official Gazette could not be consulted, so nothing unexamined is recorded as established. The year of the 1946 law and the date of the 1952 decree have been verified through an independent source, Ghānim.

— The German Civil Code (BGB; Bürgerliches Gesetzbuch), in the version in force since 1 July 2023 under the Act on the Unification of Foundation Law and on the Amendment of the Infection Protection Act of 16 July 2021 (BGBl. I S. 2947). The provisions relied upon in this paper are §§ 80, 81, 83, 83c, 85, 86 and 86e. § 82a is reproduced in Appendix 8 for comparison, and no finding rests upon it. The wording of § 86e is not reproduced in Appendix 8 or footnote 51; what rests upon it in 11.1 must therefore be read subject to this limitation. Also cited is the Federal Court of Justice’s judgement of 15 December 2016, case I ZR 63/15. The German provisions are reproduced in full in Appendix 8, together with the authority for each provision’s entry into force. The official text was not examined in the Federal Law Gazette or on gesetze-im-internet.de.

— Qānūn al-ʿadl wa-l-inṣāf fī al-qaḍāʾ ʿalā mushkilāt al-awqāf [Code of Justice and Equity in Resolving Waqf Problems] (Articles 115 and 587), Tanqīḥ al-fatāwā al-Ḥāmidiyya [Revision of the Ḥāmidī Fatwās], and al-Fatāwā al-Khayriyya: this paper draws their texts from the bodies of the fatwās quoting them within the corpus; the originals were not examined.

Fourth: References and Programme Papers

— Ibrāhīm al-Bayyūmī Ghānim, al-Awqāf wa-l-siyāsa fī Miṣr [Awqāf and Politics in Egypt], Cairo: Dār al-Shurūq, 1998 CE, pp. 434, 444 and 460.

— Bandar, al-Mulḥaq al-mushtarak li-barnāmaj “Miʿmār istiḥqāq al-manfaʿa al-khayriyya”: brūtūkūl al-istidlāl wa-maṣfūfat al-mutaghayyirāt wa-iʿlān al-tabaʿiyya [The Common Appendix to the Architecture of Entitlement to Charitable Benefit Programme: Inference Protocol, Variable Matrix and Declaration of Dependence] (2026 CE). Reproduced in full in the four papers after the bibliography.

— Bandar, Ikhtibār al-tafsīr al-munāfis al-jamāʿī ʿalā mudawwanat fatāwā al-istiḥqāq al-waqfī: arbaʿat maḥāwir [The Collective Test of the Competing Explanation on the Waqf Entitlement Corpus: Four Axes] (2026 CE). With the preserved execution code.

— Bandar, papers of the “Architecture of Charitable Benefit Production” programme and its synthesis study (2026 CE). Drafts. These include the al-Minshāwī file from the perspective of creation.

Appendix 1 · Methodology: The Corpus and Its Protocol

The release cited is WEC-432/1.0, identified at every first citation. Its hash is recorded in its statement, and the stability of its identifiers is anchored by the hash of the original pages file. Deletion from the release leaves a gap in the sequence without shifting subsequent records; no record is therefore renumbered in a later release. Satisfying the two closure conditions—the stratified reliability sample and comparison against the printed edition—requires a subsequent release using the same identifiers.

The five layers of corpus construction, reproduced from section 2: the pages of the published collection were crawled, yielding four thousand nine hundred and thirty pages; crawling ended because the collection had been exhausted, not because it was interrupted. The pages were then segmented into individual fatwās, yielding ten thousand and thirty-five fatwās in their standard structure. These were screened using a documented morphological lexicon, through strictly lexical rather than semantic screening, yielding five hundred and forty waqf candidates. The inclusion rule was then tightened after a manual precision test on sixty records, leaving four hundred and five records in the fourth layer. All these were then manually coded in five temporal batches, and nine records were excluded because they did not concern waqf entitlement despite containing its vocabulary. These included records on zakāt, divorce and adoption; six had titles containing “mawqif” (“position”), not “waqf”. The approved count thus settled at three hundred and ninety-six texts.

The five-tier source scale and the boundaries between its tiers: the lowest tier is a general juristic rule that constructs the apparatus without documenting an instance. Next comes a claim or documentary enquiry in an individual case. Then comes a procedural instance of registration, licensing or appointment, in which the muftī is asked to characterise an existing administrative procedure. Next is a procedural encounter over entitlement. Finally comes a near-complete instance of determination whose text records a dispute, a decision and an outcome.

The rule for aggregating entitlement-holder-category tags: each detailed tag is grouped under its parent category. The descriptively defined category comprises forty-eight unqualified tags and seven detailed tags; the institution category comprises twenty-three unqualified tags and four detailed tags; and the local category comprises eleven unqualified tags and two tagged by community or in combination with the named category. The release preserves the raw tag; aggregation is a decision made by this paper and declared where it is applied.

The corpus files, which form the paper’s basis and can be deposited after final documentation, are: the raw texts; the first-level database containing ten thousand and thirty-five structured fatwās; the lexical candidate set of five hundred and forty records; the entitlement corpus of four hundred and five records after the rule was tightened; the five manual-coding files organised by period; the exclusions file; and the frozen-release file and its statement. Published texts must not cite the two intermediate layers, since they retain records subsequently excluded.

The inclusion rule, in its tightened form after the precision test: an explicit institutional waqf term—founder, waqf deed, waqf recipient, nāẓir or yield—occurring twice or more, or in both the title and introduction. Uses meaning to set something aside, such as “the unborn child’s share is set aside”, are excluded, as are titles containing “mawqif”. The initial rule and its results are preserved for comparison. It is declared that the script implementing this rule has not been preserved, that literal reapplication of its wording does not reproduce the division, and that it was accompanied by manual review whose decisions were not recorded individually.

The lexical-counting protocol in section 4: the matching expressions are “kitāb al-waqf” (“waqf deed”); “ḥujjat al-waqf”, “ḥujjat waqf” and “al-ḥujja” (“the waqf deed”, “a waqf deed” and “the deed”); “ṣūra rasmiyya” (“official copy”) in its three forms; and “ishhād” (“attestation”). The document indicator records the occurrence of any of these. The second counter uses “kitāb al-waqf” alone, and the third uses “iṭṭalaʿnā” (“we have examined”). Matching covers the body of the record after removal of the heading, defined as everything preceding the first occurrence of “principles”, “question” or “answer”, and after truncation of the publishing website’s footer at the first of its declared markers. Before matching, the Hijri month name is removed in its forms “ذو الحجة”, “ذي الحجة” and “ذى الحجة” (Dhū al-Ḥijja and its inflected and orthographic variants). Each period’s percentage is calculated over its dated instances. Alongside the binary indicator, the rate of occurrence of the expressions per thousand words of the same cleaned body text is calculated to control for text length. Length is calculated from this body text, not from the length field recorded in the files: that field measures the text together with a nearly constant website addition of approximately four thousand eight hundred and ninety-three characters, compressing relative differences between periods to approximately one third of their size.

The protocol for counting the mode of inference in section 8: manual coding uses seven categories: jurisprudence or a stipulation without a named source; jurisprudence with named sources; modern law alone; modern law interpreted through jurisprudence; citation of unspecified provisions; the Majalla or an Ottoman codification; and other. The law indicator combines “modern law alone” and “modern law interpreted through jurisprudence”. The first dating key is the year extracted automatically from the text; the second is the year recorded in the fatwā heading, extracted by matching within the record’s first two hundred characters: the Gregorian year where given, and otherwise a conversion from the Hijri year using the formula CE = AH + 622 − (AH ÷ 33).

The convergence-check instrument: an instance is automatically classified as relying on law if its answer contains any of the following: “law number”, “the law number”, “Waqf Law number”, “Article number”, “under Article”, “under the law”, “Decree-Law”, or either 1946 or 1952 coupled with a law number. This is compared with the manual coding across the entire corpus, and agreement and the kappa coefficient are calculated.

The protocol for the decisive test in 8.4: the cleaned text is divided at the first occurrence of “question” and then the subsequent occurrence of “answer”; the intervening text forms the question segment and the following text the answer segment. A text is excluded unless both segments meet the minimum length threshold. The tightened law-number expression is “the law number”, “law number”, “Decree-Law”, or “of the year” followed by a year in the twentieth century. It does not capture a bare reference to an article, to avoid capturing provisions of Qānūn al-ʿadl wa-l-inṣāf [Code of Justice and Equity] and other juristic codifications.

The protocol for testing the comprehensive competing explanation: it is applied to the text after removal of additions from the publishing website, not to the recorded length field. It has four axes: length, using four tests—stratified comparison by quartile, logistic regression, a sensitivity test, and citation density per ten thousand characters; the composition of issues, by holding the family of issues constant, holding entitlement-holder category constant, removing every text whose question mentions a law or liquidation, and applying combined four-way normalisation with a confidence interval obtained through resampling; questioner identity, using the five-category classification and a cross-tabulation; and publication density, using correlations, the annual series, removal of the year with the highest density, the distribution of muftīs, and the representation ratio of decades. The execution code is preserved with the report.

The liquidation-issues flag in section 10: a record is tagged as concerning liquidation if its body contains any of the following: “180 of 1952”, “abolition of waqf for non-charitable purposes”, “devolution”, “land registration”, “dissolution of the waqf”, or “termination of the waqf”. This is a coarse flag and does not substitute for coding issue type.

The testing programme, which has not been carried out and is not credited to this paper: a test of coding reliability between two readers on a random sample of at least sixty instances mixing periods, blind to date and muftī, with agreement reported field by field; recoding issue type using a single rule declared before reading, then measuring the document trend within each type; stratified comparison against the printed edition for twenty-four to thirty texts, to distinguish the questioner’s words, the muftī’s answer, and the editor’s title and principles; completion of the dates of the six instances; coding the document’s weight in reasoning rather than the presence of its vocabulary; exhaustive identification of the dates of deeds stipulating the exclusion of indebted beneficiaries, to test the claim that such stipulations became more numerous over time; exhaustive identification of conflicts between stipulation and law and analysis of their reasoning; complete reading of the one hundred and thirty-five excluded records to determine the miss rate; and identification of the third and fourth duplicate pairs.

The limited administrative evidentiary gateway, which is a condition for any upgrading of the claims in sections 7 and 11: an administrative instrument identifying the beneficiary; an instance of amendment, deletion or refusal of an entry; and a route to correction. Some of these should preferably postdate 1959. Their declared likely sources are the Ministry’s internal regulations, its reports, and appointment files at the National Archives.

Limitations of corpus construction, expanded from section 2; the documentation limitation declared there as a single item is detailed here in three respects:

Three limitations of this construction must be recorded rather than left implicit. First, the transition from the third to the fourth layer follows a rule known in textual form but not reproducible through automated execution: its implementing script was not preserved, literal reapplication of its wording at freezing did not reproduce the division, and it was accompanied by manual review whose decisions were not recorded individually. Second, the sixty-record precision test retains its identifiers but lacks its judgements: all that is known is that forty-four records remained and sixteen were excluded. Third, and most importantly, measurement of the miss rate among excluded records has not been established. The first review’s zero result is rejected because its identifiers are missing; it was also contradicted by a new sample read at freezing. That sample contained two confirmed misses, one concerning substitution of waqf assets in 1894 and the other the sale of entitlement in 1903, two likely misses and one borderline case. The known minimum miss rate is therefore two out of one hundred and thirty-five, approximately one and a half per cent (1.5%), with a likely figure of approximately three (3%). This is a lower bound, not an estimate, since not all one hundred and thirty-five records were read. None was added to the frozen release84.

The count of three hundred and ninety-six refers to units of description, not distinct instances: the corpus contains two documented duplicate pairs, each instance having been published under two titles. The number of distinct instances known with certainty is therefore four hundred less six. Programme papers recorded a reduction to approximately three hundred and ninety-two on the basis of four pairs, but the third and fourth pairs remain unidentified. The documented figure is therefore retained and the lower figure rejected85.

Footnotes

  1. See note 5. ↩

  2. See note 5. ↩

Appendix 2 · The al-Minshāwī Waqf Case Record: Five Episodes

Details of the summary in 4.3, with the source for each episode in its footnote here:

The first episode (1907): the Ministry of Public Instruction submitted a communication to the muftī concerning two hundred and fifty pounds allocated by the founder in his waqf deed, recorded at Ṭanṭā Sharia Court, to admit a number of poor pupils to Ṭanṭā School. It asked what the stipulation requiring consultation and acknowledgement meant, and whether the nāẓir could be required to pay the sum for the preceding period if the school had had no pupils of this kind86.

The second episode (1908): the questioner was one of those named by the founder in his stipulation. The issue was whether recipients of money and agricultural land fell within the charitable disbursement purposes to which the founder had stipulated priority87.

The third episode (1909): the issue concerned the hospital: should expenditure on it come from total revenue or from the allocated sum of two thousand pounds? The muftī’s answer adhered to the allocation’s literal terms, ruling that the founder’s wording explicitly required expenditure on those items of equipment to come from the two thousand pounds, not from total revenue88.

The fourth episode (1915): the Office of the Shaykh of al-Aḥmadī Mosque submitted a communication accompanied by a copy of al-Minshāwī’s waqf deed, asking which teachers qualified for entitlement by description. Muḥammad Bakhīt detailed the qualifying descriptions according to the founder’s stipulation89.

The fifth episode (1921): the issue was the interpretation of a stipulation added by the founder through a deed of amendment issued in 1314 AH, excluding indebted waqf beneficiaries90.

Footnotes

  1. Fatāwā Dār al-Iftāʾ al-Miṣriyya [Fatwās of the Egyptian Dār al-Iftāʾ], “Waqf”, fatwā of Bakrī al-Ṣadafī, Jumādā al-Ūlā 1325 AH, location 2782. The question was submitted in a communication from the Ministry of Public Instruction dated 7 July 1907, No. 2353, concerning 250 pounds in Aḥmad al-Minshāwī Pasha’s waqf deed, recorded at Ṭanṭā Sharia Court. The passage draws on the substance of the question. ↩

  2. The same source, “Waqf”, fatwā of Bakrī al-Ṣadafī, Rabīʿ al-Awwal 1326 AH, location 2784. The questioner was Basyūnī Bey al-Khaṭīb, one of those named by the founder in his stipulation. The waqf deed was issued by the Sharia Court of al-Gharbiyya Province on 10 Shawwāl 1322 AH. ↩

  3. The same source, “Takhṣīṣ mablagh min al-rayʿ li-muʿayyan yataqayyad bihi al-nāẓir” [Allocation of a Sum from Revenue to a Named Beneficiary Binds the Nāẓir], fatwā of Bakrī al-Ṣadafī, Rabīʿ al-Thānī 1327 AH, location 2947. The questioner was Shaykh ʿAbd al-Razzāq al-Qāḍī. The quotation is from the answer. The heading gives no Gregorian date; 1904 CE in the automated record is the date of the waqf deed mentioned in the question, 17 December 1904, not the date of the fatwā. ↩

  4. The same source, “al-Waqf ʿalā al-mudarrisīn bi-l-Jāmiʿ al-Aḥmadī” [Waqf for Teachers at al-Aḥmadī Mosque], fatwā of Muḥammad Bakhīt, Shaʿbān 1333 AH, location 2884. The question was submitted in a communication from the Office of the Shaykh of al-Aḥmadī Mosque, accompanied by a copy of the text of al-Minshāwī Pasha’s waqf deed. ↩

  5. The same source, “Waqf istiḥqāqī wa-khayrī” [A Waqf with Designated Beneficiaries and Charitable Purposes], fatwā of ʿAbd al-Raḥmān Qarāʿa, Muḥarram 1340 AH / 18 September 1921, location 3016. The questioner was Shaykh ʿAbd al-Rāziq al-Qāḍī. The waqf deed was drawn up by the Sharia Court of al-Gharbiyya Province in Ṭanṭā on 7 Muḥarram 1313 AH, register No. 180; the deed of amendment containing the stipulation was dated 26 Shawwāl 1314 AH, registered No. 55, from al-Sanṭa Sharia Court. ↩

Appendix 3 · Four Instances of the Registration Apparatus Seeking Fatwās

Reproduced from section 7, where their sources appear in the footnotes:

The first (1953): four entitlement-holders wished to register their ownership shares in waqf real property following Decree-Law 180 of 1952, “in accordance with the procedures and rules governing inheritance rights laid down in Law 164 of 1946 Regulating Land Registration”. Completing the procedures required determination of each entitlement-holder’s portion, so the matter was referred to the muftī with a request to determine the portions according to the waqf deed.

The second (1955): the questioner expressly states that “the Land Registration Agency in al-Minyā requested that a fatwā be obtained on the wife’s portion, so that the dissolution of the waqf could be registered on that basis”.

The third (1955): the Land Registration Office in Damanhūr wrote through Damanhūr Sharia Court of First Instance to enquire about the characterisation of sums stipulated by the founder of a waqf for which it had received an application for registration: were these stipends within the meaning to which the law attached its rules, or non-recurrent giving? Ḥasan Maʾmūn set out the conditions of a stipend in jurisprudence and law, ruling that recurrence must be stipulated explicitly, certainly and conclusively, and that an ambiguous text could not be relied upon.

The fourth (1962): an acknowledgement drawn up at the Land Registration and Notarisation Office in Cairo in 1952 was submitted to the muftī, who was asked to explain its effect on entitlement.

Appendix 4 · The Economic Potential of Entitlement: Case Texts

Reproduced from section 5, where their sources appear in the footnotes:

Priority in attachment: when creditors’ judgements competed with a maintenance order for an entitlement-holder’s wife and two daughters, Bakrī al-Ṣadafī based his answer on quotations from Tanqīḥ al-Ḥāmidiyya [Revision of al-Ḥāmidiyya] and al-Fatāwā al-Khayriyya, then ruled that “the holders of the established debts mentioned in this question have no exclusive claim to the said husband’s entitlement to the yield of that waqf; rather, they receive what remains after his maintenance and that of his wife and dependants”.

Forms of stipulations excluding indebted beneficiaries, represented by four instances between 1921 and 1955 and varying in severity: a stipulation that a debt incurred by any entitlement-holder must not be paid from the waqf’s yield; a stipulation excluding anyone whose debt resulted in attachment of waqf assets; and a stipulation making exclusion take effect ten days before the attachment judgement.

Appendix 5 · This Paper’s Row in the Programme’s Variable Matrix

Reproduced from section 2. It is included to prevent the four papers from being read as four labels for a single institutional change.

Dependent variableFalsifier and resultDistinct evidence not shared
Mode of inference in the muftī’s answer, seven categories, by decade(a) The effect of the dating key or of a single muftī: tested using two keys and controlling for the muftī, and ruled out. (b) The replacement hypothesis—the document recedes in proportion to the rise of law: tested with two measures and ruled outA field recording the mode of inference, manually coded for 396 texts · the counter for invocation of documents, controlled for length · the questioner field · the case record of the al-Minshāwī waqf · thirteen instances in which the document was absent · five instances of registration

Appendix 6 · The Resident Student at al-Azhar: The Waqf Deed and the Khiṭaṭ

Reproduced from section 6, where their sources appear in the footnotes:

  • The stipulation governing ʿAbd al-Raḥmān Katkhudā’s regular provision: two loaves daily for a specified number of teachers and students “from those entered in the register, in order of entry; if one is absent or dies, a replacement enters from those waiting at the door, in order of arrival, and he stipulated that only someone engaged in learning may receive it”, specifically from among Upper Egyptians.

  • ʿAlī Mubārak’s description of the governance of registration: among the resident students, “each community has a place of its own where its members live with their belongings and receive regular provisions and stipends; each community has a register held by its naqīb and a shaykh who adjudicates among them and defends them”.

  • The convergence of the two routes in the corpus: founders establish awqāf for students at al-Aḥmadī Mosque, for students of the al-ʿAbbāsī riwāq, and even for Jabartī resident students; the fatwā specifies their qualifying descriptions by reference to prevailing custom.

Appendix 7 · Who the Questioners Were: The First Counter’s Series

Reproduced from section 9, where its source appears in the footnote. Each period’s denominator is its dated instances.

Instances involving institutional questioners numbered thirty out of one hundred and six between 1894 and 1913, then seventeen out of one hundred and twenty-three, then four out of fifty-two, then five out of forty, then two out of forty-eight, and finally zero out of twenty-one after 1960.

Footnotes

Appendix 8 · Cited Provisions in Force Governing the Purpose Foundation

Limitation of the evidentiary record: the texts below come from a consolidated mirror of federal legislation (buzer.de), which identifies each provision’s authority in the Federal Law Gazette (BGBl. 2021 I S. 2947, in force 1 July 2023). The Gazette text itself was not examined.

The provisions are reproduced in German without translation because what requires verification is the wording of the text, not the intermediary’s rendering. The passages presented in the paper’s main text are translations to be compared against this wording.

The 2021 reform changed both numbering and substance. Any reference in literature published before 2023 to “§ 85 BGB” or “§ 87 BGB” therefore denotes a text different from the one reproduced here.

§ 82a is reproduced below for comparison alone; no finding in the main text rests upon it.

§ 80 BGB — Ausgestaltung und Entstehung der Stiftung

(1) 1 Die Stiftung ist eine mit einem Vermögen zur dauernden und nachhaltigen Erfüllung eines vom Stifter vorgegebenen Zwecks ausgestattete, mitgliederlose juristische Person. 2 Die Stiftung wird in der Regel auf unbestimmte Zeit errichtet, sie kann aber auch auf bestimmte Zeit errichtet werden, innerhalb derer ihr gesamtes Vermögen zur Erfüllung ihres Zwecks zu verbrauchen ist (Verbrauchsstiftung). (2) 1 Zur Entstehung der Stiftung sind das Stiftungsgeschäft und die Anerkennung der Stiftung durch die zuständige Behörde des Landes erforderlich, in dem die Stiftung ihren Sitz haben soll. 2 Wird die Stiftung erst nach dem Tode des Stifters anerkannt, so gilt sie für Zuwendungen des Stifters als schon vor dessen Tod entstanden.

§ 81 BGB — Stiftungsgeschäft

(1) Im Stiftungsgeschäft muss der Stifter 1. der Stiftung eine Satzung geben, die mindestens Bestimmungen enthalten muss über a) den Zweck der Stiftung, b) den Namen der Stiftung, c) den Sitz der Stiftung und d) die Bildung des Vorstands der Stiftung sowie 2. zur Erfüllung des von ihm vorgegebenen Stiftungszwecks ein Vermögen widmen (gewidmetes Vermögen), das der Stiftung zu deren eigener Verfügung zu überlassen ist. (2) Die Satzung einer Verbrauchsstiftung muss zusätzlich enthalten: 1. die Festlegung der Zeit, für die die Stiftung errichtet wird, und 2. Bestimmungen zur Verwendung des Stiftungsvermögens, die die nachhaltige Erfüllung des Stiftungszwecks und den vollständigen Verbrauch des Stiftungsvermögens innerhalb der Zeit, für welche die Stiftung errichtet wird, gesichert erscheinen lassen. (3) Das Stiftungsgeschäft bedarf der schriftlichen Form, wenn nicht in anderen Vorschriften ausdrücklich eine strengere Form als die schriftliche Form vorgeschrieben ist, oder es muss in einer Verfügung von Todes wegen enthalten sein. (4) 1 Wenn der Stifter verstorben ist und er im Stiftungsgeschäft zwar den Zweck der Stiftung festgelegt und ein Vermögen gewidmet hat, das Stiftungsgeschäft im Übrigen jedoch nicht den gesetzlichen Anforderungen des Absatzes 1 oder des Absatzes 2 genügt, hat die nach Landesrecht zuständige Behörde das Stiftungsgeschäft um die Satzung oder um fehlende Satzungsbestimmungen zu ergänzen. 2 Bei der Ergänzung des Stiftungsgeschäfts soll die Behörde den wirklichen, hilfsweise den mutmaßlichen Willen des Stifters beachten. 3 Wurde im Stiftungsgeschäft kein Sitz der Stiftung bestimmt, ist im Zweifel anzunehmen, dass der Sitz am letzten Wohnsitz des Stifters im Inland sein soll.

§ 82a BGB — Übertragung und Übergang des gewidmeten Vermögens

1 Ist die Stiftung anerkannt, so ist der Stifter verpflichtet, das gewidmete Vermögen auf die Stiftung zu übertragen. 2 Rechte, zu deren Übertragung eine Abtretung genügt, gehen mit der Anerkennung auf die Stiftung über, sofern sich nicht aus dem Stiftungsgeschäft ein anderer Wille des Stifters ergibt.

§ 83 BGB — Stiftungsverfassung und Stifterwille

(1) Die Verfassung der Stiftung wird, soweit sie nicht auf Bundes- oder Landesgesetz beruht, durch das Stiftungsgeschäft und insbesondere die Satzung bestimmt. (2) Die Stiftungsorgane haben bei ihrer Tätigkeit für die Stiftung und die zuständigen Behörden haben bei der Aufsicht über die Stiftung den bei der Errichtung der Stiftung zum Ausdruck gekommenen Willen, hilfsweise den mutmaßlichen Willen des Stifters zu beachten.

§ 83c BGB — Verwaltung des Grundstockvermögens

(1) 1 Das Grundstockvermögen ist ungeschmälert zu erhalten. 2 Der Stiftungszweck ist mit den Nutzungen des Grundstockvermögens zu erfüllen. 3 Zuwächse aus der Umschichtung des Grundstockvermögens können für die Erfüllung des Stiftungszwecks verwendet werden, soweit dies durch die Satzung nicht ausgeschlossen wurde und die Erhaltung des Grundstockvermögens gewährleistet ist. (2) 1 Durch die Satzung kann bestimmt werden, dass die Stiftung einen Teil des Grundstockvermögens verbrauchen darf. 2 In einer solchen Satzungsbestimmung muss die Stiftung verpflichtet werden, das Grundstockvermögen in absehbarer Zeit wieder um den verbrauchten Teil aufzustocken. (3) Durch Landesrecht kann vorgesehen werden, dass die nach Landesrecht zuständigen Behörden auf Antrag einer Stiftung für einen bestimmten Teil des Grundstockvermögens eine zeitlich begrenzte Ausnahme von Absatz 1 Satz 1 zulassen können, wenn dadurch die dauernde und nachhaltige Erfüllung des Stiftungszwecks nicht beeinträchtigt wird.

§ 85 BGB — Voraussetzungen für Satzungsänderungen

(1) 1 Durch Satzungsänderung kann der Stiftung ein anderer Zweck gegeben oder der Zweck der Stiftung kann erheblich beschränkt werden, wenn 1. der Stiftungszweck nicht mehr dauernd und nachhaltig erfüllt werden kann oder 2. der Stiftungszweck das Gemeinwohl gefährdet. 2 Die Voraussetzungen des Satzes 1 Nummer 1 liegen insbesondere vor, wenn eine Stiftung keine ausreichenden Mittel für die nachhaltige Erfüllung des Stiftungszwecks hat und solche Mittel in absehbarer Zeit auch nicht erwerben kann. 3 Der Stiftungszweck kann nach Satz 1 nur geändert werden, wenn gesichert erscheint, dass die Stiftung den beabsichtigten neuen oder beschränkten Stiftungszweck dauernd und nachhaltig erfüllen kann. 4 Liegen die Voraussetzungen nach Satz 1 Nummer 1 und Satz 3 vor, kann eine auf unbestimmte Zeit errichtete Stiftung auch abweichend von § 83c durch Satzungsänderung in eine Verbrauchsstiftung umgestaltet werden, indem die Satzung um Bestimmungen nach § 81 Absatz 2 ergänzt wird. (2) 1 Durch Satzungsänderung kann der Stiftungszweck in anderer Weise als nach Absatz 1 Satz 1 oder es können andere prägende Bestimmungen der Stiftungsverfassung geändert werden, wenn sich die Verhältnisse nach Errichtung der Stiftung wesentlich verändert haben und eine solche Änderung erforderlich ist, um die Stiftung an die veränderten Verhältnisse anzupassen. 2 Als prägend für eine Stiftung sind regelmäßig die Bestimmungen über den Namen, den Sitz, die Art und Weise der Zweckerfüllung und über die Verwaltung des Grundstockvermögens anzusehen. (3) Durch Satzungsänderung können Bestimmungen der Satzung, die nicht unter Absatz 1 oder Absatz 2 Satz 1 fallen, geändert werden, wenn dies der Erfüllung des Stiftungszwecks dient. (4) 1 Im Stiftungsgeschäft kann der Stifter Satzungsänderungen nach den Absätzen 1 bis 3 ausschließen oder beschränken. 2 Satzungsänderungen durch Organe der Stiftung kann der Stifter im Stiftungsgeschäft auch abweichend von den Absätzen 1 bis 3 zulassen. 3 Satzungsbestimmungen nach Satz 2 sind nur wirksam, wenn der Stifter Inhalt und Ausmaß der Änderungsermächtigung hinreichend bestimmt festlegt.

§ 86 BGB — Voraussetzungen für die Zulegung

Durch Übertragung ihres Stiftungsvermögens als Ganzes kann die übertragende Stiftung einer übernehmenden Stiftung zugelegt werden, wenn 1. sich die Verhältnisse nach Errichtung der übertragenden Stiftung wesentlich verändert haben und eine Satzungsänderung nach § 85 Absatz 2 bis 4 nicht ausreicht, um die übertragende Stiftung an die veränderten Verhältnisse anzupassen, oder wenn schon seit Errichtung der Stiftung die Voraussetzungen für eine Auflösung nach § 87 Absatz 1 Satz 1 vorlagen, 2. der Zweck der übertragenden Stiftung im Wesentlichen mit einem Zweck der übernehmenden Stiftung übereinstimmt, 3. gesichert erscheint, dass die übernehmende Stiftung ihren Zweck auch nach der Zulegung im Wesentlichen in gleicher Weise dauernd und nachhaltig erfüllen kann, und 4. die Rechte von Personen gewahrt werden, für die in der Satzung der übertragenden Stiftung Ansprüche auf Stiftungsleistungen begründet sind.

Addendum to Appendix 8 · Record of Disputed Points Resolved Against the Text in Force

The passages cited in this paper from the German Civil Code (BGB) were checked against the text in force since the first of July 2023, as reproduced in a consolidated mirror of federal legislation with the authority for entry into force supplied for each provision. This comparison resolved three points concerning source attribution:

LocationIssueResolutionDirect effect on the claim
List of Sources, Part IIIInclusion of §82a among the provisions used, although it appears neither in the main text nor in footnote 51The citation was corrected: provisions relied upon in the argument were separated from those recorded for comparisonNo effect; no claim had been based on it
Footnote 51The stated source status of the text—obtained from legal scholarship mirrors and the government bill (Drucksache 19/28173)—conflicts with the limits of the evidentiary record stated at the beginning of this appendixThe citation was corrected: the source is a consolidated mirror of federal legislation with the authority for entry into force supplied for each provision; the bill is not used as the source for the text of any provisionStronger source attribution, not a stronger claim; the text’s source status improved, but no conclusion changed
11.1, first claimReliance on the renumbering of §85, although the provision in force concerns the conditions for amending the charter, not the beneficiary’s legal positionThe conclusion was confined to its proper scope: the paper explicitly states that the inference rests on the renumbering, not the provision’s substance, and identifies §86 no. 4 as the provision concerning beneficiariesThe first of the two pieces of evidence for replacement was weakened; the test result stands because the second (§86 no. 4, §83(2) and the 2016 judgement) was unaffected

The result of the replacement test stated in the abstract and conclusion did not change: neither claim collapses upon replacement with the German comparator, and on this element waqf is not distinct from its comparator within the family of purpose institutions.

Appendix 9 · Passages Relocated from the Main Text

These passages were relocated from the main text to the appendix verbatim; no qualification, evidence or statistical figure was removed.

A · The Case Record of al-Minshāwī’s Waqf Before Dār al-Iftāʾ: Relocated Text

Because the evidentiary authority of documents is tested in actual practice rather than abstract description, the paper traced a single waqf through successive fatwās. The production programme studied the portfolio of Aḥmad al-Minshāwī Pasha’s waqf from the perspective of its creation; here it is read from the perspective of those who claimed benefit within it.

There are five successive episodes between (1907 and 1921): the sum allocated to pupils at Ṭanṭā School, where the questioner was the Ministry of Public Instruction; the inclusion of waqf recipients of cash and land among the charitable purposes stipulated to take precedence; the hospital’s disbursement purpose, confined by the literal terms of the allocation to the sum of two thousand pounds rather than revenue in general; the qualifying descriptions of teachers at al-Aḥmadī Mosque, elaborated by Muḥammad Bakhīt in accordance with the founder’s stipulation; and the interpretation of a stipulation added by the founder through a deed of amendment issued in 1314 AH, concerning the exclusion of those who incurred debt (the episodes, their questioners and their issues are detailed in Appendix 2, with their footnotes in the corresponding locations).

What remains constant throughout the record is that the deed is the sole reference point: a copy is invoked in every episode, and the juristic answer rests on the wording of its stipulation. Only the types of questioners and issues submitted change (from the Ministry of Public Instruction to an individual nāẓir to the leadership of an entire institute, and from interpreting the stipulation to managing its implementation).

Over fourteen years, then, a single waqf becomes a live file of successive claims, all administered through a single document. This is the first system at the height of its operation: a government register has no central role; everything centres on a waqf deed preserved in a court record office, whose official copies circulate among a nāẓir, a ministry, an institute’s leadership and a muftī.

B · The Second Explanation: The Editorial Layer

This explanation holds that the counters measure the conventions of Dār al-Iftāʾ’s office in drafting and publishing fatwās, rather than the work of issuing them. It retains its force after the comprehensive competing explanation has been rejected, since it concerns wording rather than the composition of the material.

Its strong form was addressed through four methodological procedures: first, basing the coding on the body of the answer rather than the title; second, excluding the heading from the scope of matching; third, counting the opening formula separately from mentions of the document; and fourth, measuring the two trends separately.

The formula “We have examined” peaked between 1934 and 1945 at (96%), then fell to twenty-nine per cent (29%); mentions of the document, by contrast, peaked a full decade later (1954–1959), at (79%). Had they measured a single template, they would have moved in the same direction and peaked at the same time.

Their divergence in both the timing of their peaks and their direction ruled out the explanation in its strong form (that they are a single effect of a single cause). Its weak form remains: part of the magnitude of each trend is attributable to editorial custom, and cannot be estimated before comparison of the printed editions. The explanation is thus rejected as to direction but remains viable as to magnitude.

C · The Third Explanation: The Composition of Issues in the Trend in Document Invocation

What was ruled out in section 10.1 was the explanation of the trend in law citations through the composition of issues. The question remains for the trend in document invocation: does its peak during the decade of liquidation reflect the composition of issues rather than the persistence of a system? The issues arising in that decade inherently require the deed to be read.

An initial check was conducted using an explicitly stated lexical flag, and recalculated on the frozen release91. The flag is the occurrence of a reference to the liquidation law, devolution, dissolution of the waqf or land registration in the body of the record. For 1954–1959, document invocation reaches seventy-seven per cent (77%) in instances tagged as liquidation: twenty-seven out of thirty-five. It reaches eighty-five per cent (85%) in untagged instances: eleven out of thirteen. Instances without the liquidation flag therefore invoke the document no less frequently than those with it; indeed, they do so more frequently. In the period before 1946, the lexical flag does not distinguish the two categories at all: the rate is fifty per cent (50%) among tagged instances and forty-nine per cent (49%) among the rest.

It follows that the competing explanation does not explain the peak in document invocation: if the composition of issues were the cause, a statistical difference would appear between the two categories within the same period. The check has two limitations: first, the flag is a crude lexical indicator, no substitute for coding the type of issue; second, the denominators in the untagged category are small. The required test is specific: recode the type of issue using a single rule declared before reading, then measure the trend in document invocation within the same type.

D · The Fourth Explanation: Generational Shift

This was addressed in section 8.3 through checks at the level of the muftī and the year; it was ruled out by controlling for a single muftī whose tenure spanned both decades (such as Shaykh Ḥasanayn Makhlūf). A limited role remains for this competing explanation: the arrival of later muftīs with legal training helped the new form become established more quickly. This is an interaction, not an alternative explanation.

E · Entitlement as Economic Value: Relocated Text

Was entitlement a charitable hope, or a position with material value traded in the market? The corpus answers through a three-part conceptual apparatus: the economic potential of entitlement. Its three dimensions are: (susceptibility to valuation, liability to attachment and use as security, and transferability). These dimensions are not equivalent: a creditor may gain access to monetary revenue while the entitlement-holder has no power to sell the underlying legal position (this is the author’s term in this paper, confined to the definition just given).

Liability to attachment is documented from the beginning of the corpus: the superintendent of al-Ḥilmiyya Awqāf presented the case of an entitlement-holder who had incurred debts and whose creditors had attached his entitlement. A financier had been found who would lend him funds to discharge his debts, on condition that he waive his share of the revenue in the financier’s favour for his lifetime so long as he remained indebted, secured by an undertaking from the nāẓir. Shaykh Muḥammad ʿAbduh ruled that “a waqf recipient entitled to revenue may not waive his right to it in favour of another, for, like inheritance, it is not extinguished by waiver”. He permitted the creditor to be appointed as an agent to collect the revenue towards repayment of the principal, but prohibited the nāẓir from giving the undertaking “because the waqf has no independent legal capacity to incur debts (dhimma), so that debt cannot attach to it”92. The situation thus contains the full set of financial elements: (attaching creditors, a financier who reschedules the debt, and a waqf flow whose receipt is pledged).

The order of priority in attachment is itself governed by jurisprudence: when creditors’ judgements competed with a maintenance order for the entitlement-holder’s wife and two daughters, Bakrī al-Ṣadafī ruled, on the basis of passages from Tanqīḥ al-Ḥāmidiyya [Revision of al-Ḥāmidiyya] and al-Fatāwā al-Khayriyya [The Khayriyya Fatwās], that creditors have no exclusive claim to the debtor’s entitlement, but receive only what remains after his maintenance and that of his wife and dependants93. Entitlement even became an asset against which family judgements could be enforced: a divorced woman attaches her former husband’s entitlement to recover her maintenance94. Waqf benefit is thus protected up to the level of subsistence before it becomes available for enforcement. The text of the answer appears in Appendix 4.

Transferability is narrower and more tightly regulated: assignment of entitlement to another entitlement-holder for a specified period is permissible and does not constitute an acknowledgement extinguishing the right95. An alleged sale of entitlement is judged by the standards of jurisprudence rather than those of the market96. Inheritance is ruled out as a matter of principle: waqf entitlement does not form part of the deceased’s estate, so the deceased entitlement-holder’s debts are not discharged from it, because the successor receives under the founder’s stipulation, not by inheritance97.

The founders themselves entered the dispute: stipulations excluding those who incurred debt are documented in four instances in the corpus between 1921 and 1955. The strictest makes exclusion take effect ten days before the attachment judgement is issued; their full wording appears in Appendix 4 98. The founder seeks to protect the flow from the creditors of his descendants, while the muftīs regulate his stipulations through the standards of signification. (The assertion that these stipulations become more frequent after a particular year is a claim about a temporal trend that these instances cannot support: the dates on which the deeds were drawn up have not been systematically enumerated, and their enumeration is an item in the testing programme.)

This section has one limitation concerning enumeration: the review of excluded records conducted at freezing identified an instance of the sale of entitlement in 1903 (in which Shaykh Muḥammad ʿAbduh issued a fatwā concerning an indebted entitlement-holder’s sale of his entitlement for the next six years), classified as a confirmed miss. It belongs squarely within this section but was not included in the frozen release. What is presented here is therefore a lower bound on what the collection contains concerning capacity for monetisation, rather than an exhaustive account99.

The conclusion remains proportionate to the evidence: entitlement is valued, attached and pledged through its receipt, and the entitled position passes according to the founder’s stipulation. It is an economic value fully open to transactions, even though the underlying position cannot lawfully be sold. Aḥmad Amīn’s account of resident students at al-Azhar selling their regular provisions at the gates should be read in these terms.

Footnotes

  1. The initial check of the composition of issues using the lexical flag, whose formulation appears in the appendix, rerun on the frozen release WEC-432/1.0. Its findings for 1954–1959 are: 27 out of 35 among instances tagged as liquidation and 11 out of 13 among the rest. For 1946–1953: 5 out of 6 and 24 out of 34. Before 1946: 8 out of 16 and 131 out of 265. The document indicator used here is the corrected one, after excluding the heading and the month name. It should be noted that an earlier calculation produced counts differing from these by one instance in each subgroup. Those recorded here are the results of rerunning the calculation on the release; this did not change the direction of the check. ↩

  2. The same source, “Ḥukm isqāṭ al-rayʿ” [Ruling on the Waiver of Revenue], fatwā of Muḥammad ʿAbduh, Ramaḍān 1321 AH, location 2994; the account is supplied by the superintendent of the al-Ḥilmiyya Awqāf. The three quotations come from the answer itself, not from the compilation’s principles. ↩

  3. The same source, “Nafaqat al-mustaḥiqq wa-zawjatihi wa-ʿiyālihi muqaddama ʿalā sāʾir al-duyūn” [Maintenance for the Entitlement-Holder, His Wife and His Dependants Takes Precedence over All Other Debts], fatwā of Bakrī al-Ṣadafī, Ṣafar 1329 AH, location 2976; the questioner is Ṣāliḥ Effendi, agent for the affairs of al-Sitt Fāṭima Hānim. The quotation comes from the answer, in which the muftī bases his ruling on passages from Tanqīḥ al-Ḥāmidiyya and Fatāwā al-Khayriyya [The Khayriyya Fatwās]. The sentence “Maintenance for the entitlement-holder, his wife and his dependants takes precedence over all other debts” is the compilation’s title wording, not the muftī’s wording. The heading gives no Gregorian date; Ṣafar 1329 AH corresponds to February 1911. ↩

  4. The same source, “Istimrār ḥukm nafaqat al-muṭallaqa fī al-ʿidda” [Continuation of a Divorced Woman’s Maintenance Order during the Waiting Period], fatwā of Muḥammad Bakhīt, Rabīʿ al-Thānī 1336 AH / 22 January 1918, location 583. ↩

  5. The same source, “al-Tanāzul ghayr al-iqrār” [Assignment Is Not Acknowledgement], fatwā of ʿAbd al-Majīd Salīm, Ramaḍān 1347 AH / 14 February 1929, location 2736. ↩

  6. The same source, “Al-tanāzul ʿan al-istiḥqāq fī al-waqf wa-bayʿuhu” [Assignment and Sale of Waqf Entitlement], fatwā of ʿAbd al-Majīd Salīm, Dhū al-Qaʿda 1356 AH / 22 January 1938, location 3032. ↩

  7. The same source, “Al-istiḥqāq fī al-waqf laysa tarika” [Waqf Entitlement Is Not an Estate], fatwā of ʿAbd al-Majīd Salīm, Shaʿbān 1353 AH / 13 November 1934, location 3030. ↩

  8. See note 42. ↩

  9. See note 5. ↩

Appendix 10 · Testing the Six Elements Whose Findings Were Reversed Against This Paper’s Claims

The findings from the corpus were retested using three comparators from two families, reversing the findings on six elements. Each is examined below against this paper’s claims: (did the paper base a claim on it, requiring correction, or did it make no such claim, in which case the check itself is recorded):

ElementDid the paper base a claim on it?Action taken
Direction of the purpose condition: positive or negativeNoThe paper does not address whether a purpose qualifies; its subject is establishing beneficiary eligibility. No effect
Status of the governmental act at formationNoThe paper makes no claim about the origin of the status. What is mentioned in 11.1 is a renumbering, not the substance of formation. No effect
Devolution of net assets upon dissolutionNoThis has no place in the fatwā corpus examined or in the two claims. No effect
Perpetuity, considered separately from locking the endowed principalNoThe paper covers 1894–1979 in examining how eligibility is established, and makes no claim about the continuity of the status. No effect
Gap-filling authorityNo; this reinforces its accountIts claim is that the document continues to carry the right, not that jurisprudence alone supplies a gap-filling authority antecedent to it. The axis is threefold: a will-based gap-filling authority in the German case, a normative and customary one in the Indian case, and none in the French case. No claim is therefore corrected; it is recorded that the paper did not claim uniqueness
Beneficiary standingYes; it was correctedThe denial was confined to its proper scope in 11.1: absence of standing against the foundation, not against the authority. §86e(2) was recorded, as was the comparator from the third family in sections 34(1) and 34(4)

A seventh point was affected by the comparative round but is not one of the six: the “rank of the founder’s stipulation”. This was corrected in the abstract, in 11.1 and in the conclusion together, shifting the claim from an unqualified denial of distinctiveness to identifying waqf’s position among three families.