Shelving: Treating Premises as Absent
搁置
For a premise to be treated as absent, no one needs to hide it. This paper calls that state shelving. The using side treats it as absent: it goes on using it, does not maintain its reasons, does not register it as unresolved, and the voices questioning it do not land in any procedure that could reopen it; that someone outside is asking does not change shelving. Shelving does not require concealment or active expulsion, and it is not suspension; continued adoption by itself is not enough to judge something shelved: a premise questioned and then retained by decision is not shelved, nor is one paused while still registered as unresolved (S26). Once a premise has entered a scope of warrant, continuing to adopt it does not require reasons to be put forward again; this low burden does not mean that the cost of maintaining it is zero; reopening it means passing a procedural threshold, and analysis should state what the threshold is (S27). A premise treated as absent goes on acting all the same (S28). Quantification is one carrier this paper examines closely (S29). Placed on the methodological map, a situation judged shelved under S26 is one in which the remainder of the method a standard uses is not taken to another quadrant to be dealt with, but is kept filled by a default value, and its status as unresolved is no longer registered; the remainder thus covered is not written off by this, it is still on the ledger and can change into other shapes; whether it shows itself, and whether anyone takes it up, conservation does not guarantee (S30). Five propositions in all (S26 to S30), stated at the end of each section and listed together at the end.
Writing Declaration: This paper was co-drafted with Claude (Anthropic). All intellectual decisions, framework design, and final editorial judgments were made by the author.
Introduction
In 1980 the American Psychiatric Association published the third edition of its diagnostic manual. In its introduction it set out a position. The introduction said that for most disorders the etiology is not yet known; except for disorders whose etiology is established, the manual takes no theoretical position on etiology (APA 1980, Introduction, pp.6–7). The main reason it gave was that writing theories of etiology into the manual would hinder its use by clinicians of different theoretical orientations. This position had its own background; the needs of medical insurance were one of the pressures behind it (Mayes & Horwitz 2005, p.253).
This position was not hidden. It was written in the introduction, for anyone to read. From then on, every diagnosis made by the diagnostic criteria used it; in using it, no one had to say again why it was right. It was later questioned too: the research agenda and the fifth edition's Section III discussed in Paper 2 are instances. What this case shows is default continuation: the premise goes on being used, its reasons need not be restated each time, and reopening it waits for the next revision. What is read here is how a public choice of method went on being adopted; it is not asserted that no one discussed its reasons afterward. Whether it is shelving in this paper's sense, the existing records are not enough to judge.
Shelving has one thing more than default continuation: the using side treats the premise as absent. It goes on using it, does not maintain its reasons, does not register it as unresolved, and the voices questioning it do not land in any procedure that could reopen it. Whether anyone outside is asking does not change this: a premise can be questioned in the journals for decades while the side that uses it goes on treating it as absent. So shelving sits on the using side, not on whether anyone outside is asking.
The reading follows the Prequel. This paper concerns the second domain, standards that act on persons (Prequel §5). §1 says that shelving is neither concealment nor suspension; §2 speaks of default continuation and the threshold for reopening; §3 says that a shelved premise goes on acting on persons all the same; §4 speaks of quantification; §5 places shelving on the methodological map; §6 refers back to Paper 1.
§1 Neither Concealment nor Suspension
When one says that a premise is "looked after by no one," what most readily comes to mind is that someone has hidden it, or driven out those who question it. Shelving needs neither. The position in the Introduction was written in the introduction, not hidden; those who questioned it were not driven out either.
Shelving can also have a prior history of exclusion. A question is first refused on the grounds that it "does not belong to this discipline," and later the using side treats it as absent. The exclusion passes; the shelving remains. So exclusion can be one way into shelving, but it is not a necessary condition of it.
Shelving must also be kept apart from suspension. To suspend a premise is to hold it in suspense deliberately: one knows it is there, withholds judgment for now, and keeps it in mind throughout. Suspension must be maintained; let go of it, and it is no longer suspension. Shelving needs no maintenance: it treats the premise as absent. It is not that one does not know the premise is there; it is that, in using it, one no longer thinks of it.
Finally, shelving must be kept apart from two other states, both of which look like "still in use":
| State | Is it shelving? |
|---|---|
| Questioned, the answer reached the matter, decided to retain | No: this is retention (Paper 2, §5) |
| Not pursued for now, its limits or open status still registered | No: stopping while still registered on the record is not closure (Prequel §4) |
| Still used, not registered as unresolved, treated as needing no registration or handling | Yes |
So to judge whether a case is shelving, continued adoption is not enough; there must be a record showing that the using side has not registered the premise as unresolved. What counts is a positive record showing how it is being handled: it is registered as settled, needing no handling; or a list, at the stage examined, carries the function of registering open matters of this kind, its scope of registration covers this premise, it lists others and not this one, and the related records show no separate registration either. That an item is simply not seen in some list is not enough to judge shelving; where it is only that no registration has been found, the record says "not found," and shelving is not judged (Paper 1, §3). That someone outside is asking does not change this judgment: if the voice asking does not land in a procedure that could reopen the premise, the using side can still treat it as absent.
S26 Shelving: the using side treats it as absent. Shelving is the using side treating a premise as absent: going on using it, not maintaining its reasons, not registering it as unresolved, with the voices questioning it not landing in any procedure that could reopen it; that someone outside is asking does not change shelving. It does not require concealment or active expulsion: exclusion first, turning later into treating it as absent, can also form shelving. It is not suspension: suspension must keep the premise held in suspense, still in mind. Continued adoption by itself is not enough to judge shelving: a premise questioned and then retained by decision is not shelved, nor is one paused while still registered as unresolved; a judgment requires a positive record showing that the using side has not registered it as unresolved; where it is only that no registration has been found, shelving is not judged.
§2 Default Continuation and the Threshold for Reopening
Once a premise has entered a discipline's scope of warrant, continuing to adopt it does not require justifying reasons to be put forward again. Each judgment made by it takes it by default to hold.
This low burden does not mean that the cost of maintaining the premise is zero. For a premise to go on being used, someone must train newcomers in how to use it, someone must code it into software and forms, reviews must review by it, someone must maintain its text. All this takes resources. What is low is the burden of proof, not the cost of maintenance.
Reopening it, however, means passing a threshold. Someone who wants to question the premise again must find a procedure that can take up the question: a revision, an update, a review. Procedures have conditions of initiation: who can initiate, what must be met, when it can be initiated. In analyzing a case, state what the threshold is, rather than saying vaguely that "reopening is hard."
NICE's guideline updates offer an example. When a guideline is updated, the Institute first sets the scope of this update; recommendations falling outside the scope are carried over as they stand. In the stakeholder comments table for the 2017 partial update of the guideline on heavy menstrual bleeding (NG88), an outside comment was directed at a recommendation carried over from the 2007 guideline; the reply said that the recommendation dated from 2007 and was outside the scope of the partial update (NICE 2017, p.175).
What this case proves must be stated precisely. What is seen in it is a recommendation still being carried over, outside what this update reviewed; the threshold for reopening it is that it must fall within the scope of some update. Whether it is shelving of a normative premise, the existing records are not enough to judge: it does not prove that the using side treats it as absent. The table registers both the comments and the replies publicly: a public record makes the manner of handling checkable, while whether a particular residue is acknowledged is judged separately (Paper 1, §3).
S27 Default continuation and the threshold for reopening. Once a premise has entered a scope of warrant, continuing to adopt it does not require justifying reasons to be put forward again; this low burden of proof does not mean that the actual cost of maintenance is zero (training, coding, review and maintenance still take resources). Reopening it means passing a procedural threshold of initiation; analysis should state what the threshold is.
§3 Acting on Persons All the Same
When the using side treats a premise as absent, the premise's effects go on all the same.
This is seen most clearly in classifications and rankings of people. Espeland and Sauder studied the effects of law school rankings in the United States: once the rankings were published, law schools adjusted their admissions, their spending and the data they reported according to the rankings' indicators; those being ranked changed themselves according to the ranking (Espeland & Sauder 2007). A ranking uses a set of indicators and weights, and the indicators and weights contain judgments about what a "good law school" is; the rankings are published every year all the same, and the law schools adjust every year all the same.
So if shelving occurs, it is not a state of rest: that line on the normative ledger is treated as absent, but its effects go on all the time: it classifies, it ranks, it decides who gets through the door. Those it acts on are often the first to feel its remainder.
S28 A shelved premise goes on acting on persons. Classifications, rankings and ratings act on the persons and institutions classified and ranked, who change themselves according to them; a premise treated as absent goes on having its effects all the same.
§4 Quantification
Shelving has many carriers. This paper examines one of them closely: quantification.
When a judgment is turned into a rule, and then into a number, the ought within the judgment goes with it into the rule and the number, and from then on appears in the shape of a number. A threshold, a weight, a cut-off score each records a trade-off: what counts as enough, what weighs more, where to draw the line. Once turned into a number, if there is no associated record of reasons, the trade-off is not easily seen as a trade-off; it looks like the result of a measurement.
Espeland and Stevens called this process commensuration: putting things of different kinds into a single metric so that they can be compared (Espeland & Stevens 1998). Commensuration has its uses: it lets comparison proceed, and lets judgments be repeated. Its cost is that the trade-off put into the metric, if no record goes along with it, does not easily show itself afterward.
This paper does not claim that quantification itself is wrong, and makes no judgment of frequency (it does not say that "quantification often leads to shelving"). A sentence from Institution Theory can be borrowed as a reading: "Forms such as appraisal metrics, rankings and grades are signals, not criteria" (Institution Theory Z13). A number appearing there is a prompt to go and see whether the trade-off it records has anywhere to be questioned; it is not a conclusion. Political Economy applied the same reading to rates of return (Political Economy, Paper 5, E27).
S29 Quantification is one carrier this series examines closely. When a judgment is turned into a rule and a number, the ought within the judgment goes with it into the rule, and from then on appears in the shape of a number. It is not claimed that quantification itself is wrong, and no judgment of frequency is made.
§5 The Method's Remainder Not Carried Across
Methodology Paper 2 drew a map: four methods, deduction, induction, reduction and abduction, each occupying a quadrant; each method has a remainder it cannot remove from itself. The remainder of deduction is: "the source of premises is not within deduction's jurisdiction." The remainder of abduction has two layers, one of which is: "the selection criterion is external" (Methodology Paper 2, §3.1, §3.4). The chisel-construct cycle is someone performing negation at the remainder of one quadrant and crossing to another quadrant to deal with it; the remainder itself does not drive this step. Methodology Paper 2 at the time wrote this step as a crossing forced by the remainder; The First Critique later revised this modality: "the remainder is only a gap, a stretch left unexhausted; by itself it pushes nothing — to say 'the remainder forces the next cut' is to speak of a static gap as if it were a force" (The First Critique, Chapter 1 (Introduction), §2). This paper reads it in the latter way.
A standard is a product of the method it uses; it is what Methodology Paper 2 calls a construct: "negation's sediment, temporarily stable, with remainder" (§5.2). Its normative premises often fall exactly on the remainder of that method. A classification standard is organized by deduction, and where its premises come from, deduction itself does not handle; a selection standard is reached by abduction, and by what "best" is selected, abduction itself does not fix.
A situation already judged shelved under S26, placed on this map, has this structure: the remainder appears, but is not taken to another quadrant to be dealt with; instead it is kept filled by a default value, and its corresponding status as unresolved is no longer registered. Where the premises come from, the using side no longer asks, and fills in a convention; what the selection criterion is, the using side no longer asks, and fills in a prevailing threshold. The default value lets work go on as usual, but the remainder has not been dealt with; it has only been covered. Using a default value for the time being, or not yet dealing with the remainder across quadrants, is not by itself enough to judge shelving: filling in a default value while still registering the matter as unresolved is not shelving (S26).
A remainder covered by a default value is not written off by this. The conservation postulate of The First Critique says that the remainder is "transformable only, never annihilable"; "a question no one asks anymore is still on the ledger — being forgotten does not write it off" (The First Critique, Chapter 1 (Introduction), §2). So it is still on the ledger, and can change into other shapes. Whether it shows itself, and whether anyone takes it up, conservation does not guarantee; the same place says: "conservation is no motor. A remainder persisting on the ledger pushes no one to chisel at it; whether to chisel remains, always, the subject's affair." When it shows itself, it can be a new difficulty; when someone takes it up, it is an alternative method, or a newly established questioning position (what Paper 4 calls "set up only after something goes wrong"). The continuing effects described in the previous section show only that the premise has not stopped taking effect, and have nothing to do with these. Whether anyone goes on to question it returns to the questioning position of Paper 2.
Methodology Paper 2, §7.1 has a sentence: "The compressed inductive remainder does not disappear—it comes back in another form." The first half, "does not disappear," is conservation; the second half, read by The First Critique's revision of the modality, this paper takes only as far as "can change into other shapes," not as a prediction that it "will certainly come back." A reader who does not accept the conservation postulate does not accept the sentence "still on the ledger"; the rest of this paper does not depend on it.
The following situations do not entail one another and come in no fixed order: covered, with no difficulty showing, which can last a long time; a difficulty showing, with no one taking it up; an alternative method arriving; new questioning being established.
A last exclusion condition, for diagnosing shelving: if a discipline's normative premises are, in the records examined, routinely set out and have corresponding reviews, then that is not shelving.
S30 The method's remainder not taken to another quadrant to be dealt with. A standard is a product of the method it uses; its normative premises fall on the remainder of that method (deduction: the source of premises; abduction: the selection criterion is external). A situation judged shelved under S26, placed on this map, is one in which the remainder is not taken to another quadrant to be dealt with, but is kept filled by a default value, and its corresponding status as unresolved is no longer registered; using a default value for the time being, or not yet dealing with it across quadrants, is not by itself enough to judge shelving. The remainder covered by a default value is not written off by this; it is still on the ledger, and can change into other shapes; whether it shows itself, and whether anyone takes it up, conservation does not guarantee. When it shows itself it can be a new difficulty; when someone takes it up, it is an alternative method or newly established questioning. These situations do not entail one another and come in no fixed order. An exclusion condition for diagnosing shelving: if a discipline's normative premises are, in the records examined, routinely set out and have corresponding reviews, that is not shelving.
§6 The Counterpart in the First Domain
This paper speaks of the persistence of that line on the normative ledger in the second domain, standards that act on persons. The first domain has a counterpart: a claim whose original commitment to verification is no longer borne once results come, which Paper 1, §8 calls escape (S7). Shelving and escape both stop registering something as unresolved at a given point; one falls on the normative ledger, the other on a theory's relation to testing.
Paper 1, §7 discusses the logical starting point of parochialism; this paper discusses how shelving persists, and does not take parochialism as its necessary prior history.
Where This Paper Stops
This paper does not say whose fault shelving is; it does not say that quantification is bad; it does not take parochialism as the necessary prior history of shelving.
What it hands to Paper 6: where a shelved premise can be withdrawn or its adoption stopped.
The remainder it leaves: the cases of the diagnostic manual and NICE prove only default continuation and the threshold for reopening, and whether they are shelving, the existing records are not enough to judge; the exclusion condition in §5 was not checked discipline by discipline; the sentence "still on the ledger" rests on the conservation postulate, and readers who refuse that postulate set up a branch of their own.
The Five Propositions of This Paper
- S26 Shelving: the using side treats it as absent. Still used, its reasons not maintained, not registered as unresolved, the voices questioning it not landing in any procedure that could reopen it; that someone outside is asking does not change shelving; concealment or expulsion is not required; it is not suspension; continued adoption by itself is not enough to judge, and a positive record showing that it is not registered as unresolved is needed. (Conceptual distinction)
- S27 Default continuation and the threshold for reopening. Continued adoption needs no fresh justification; a low burden of proof does not mean a zero cost of maintenance; reopening means passing a procedural threshold. (Descriptive; stating the threshold in analysis: notation)
- S28 A shelved premise goes on acting on persons. Classifications, rankings and ratings act on persons and institutions, who change themselves according to them. (Descriptive; Espeland & Sauder 2007)
- S29 Quantification is one carrier this series examines closely. A judgment turned into a rule and a number carries the ought in with it, appearing in the shape of a number; without an associated record of reasons, the trade-off is not easily seen; it is not claimed that quantification is wrong, and no judgment of frequency is made. (Descriptive; the reading takes over Institution Theory Z13)
- S30 The method's remainder not taken to another quadrant to be dealt with. In what is judged shelved under S26, the remainder is kept filled by a default value and no longer registered as unresolved, and a default value alone is not enough to judge; the covered remainder is still on the ledger and can change into other shapes, and whether it shows itself, and whether anyone takes it up, conservation does not guarantee; the situations do not entail one another; there is an exclusion condition for diagnosing shelving. (Descriptive; extension of Methodology Paper 2, own undertaking; phrasing follows §5.2 and the first half of §7.1; the modality follows The First Critique, Chapter 1 (Introduction), §2; "still on the ledger" rests on the conservation postulate, and readers who refuse the conservation postulate do not accept this sentence)
References
All SAE works below are cited by concept DOI, which always resolves to the latest version.
I · This system
- Qin, H. (2026). The Epistemological Map of Chisel-Construct: Four Methods, Four Remainders, One Cycle (SAE Methodology, Paper 2) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.18918194
- Qin, H. (2026). SAE: The First Critique [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.21439120
- Qin, H. (2026). Not Crossing the Line (SAE Institution Theory, Paper 2) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23270384
- Qin, H. (2026). Growth (SAE Political Economy, Paper 5) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23874150
- Qin, H. (2026). Parochialism (SAE Theory of Disciplines, Paper 1) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.24177748
- Qin, H. (2026). Standard-Setting (SAE Theory of Disciplines, Paper 2) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.24177754
- Qin, H. (2026). Seats (SAE Theory of Disciplines, Paper 4) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.24177765
II · Rule texts and institutional documents
- American Psychiatric Association. (1980). Diagnostic and Statistical Manual of Mental Disorders (3rd ed.). Introduction.
- NICE. (2017). Heavy menstrual bleeding: assessment and management (NG88). Consultation comments and responses (consultation 3 August to 13 September 2017). nice.org.uk.
III · External literature
- Espeland, W. N., & Sauder, M. (2007). Rankings and reactivity: How public measures recreate social worlds. American Journal of Sociology, 113(1), 1–40.
- Espeland, W. N., & Stevens, M. L. (1998). Commensuration as a social process. Annual Review of Sociology, 24, 313–343.
- Mayes, R., & Horwitz, A. V. (2005). DSM-III and the revolution in the classification of mental illness. Journal of the History of the Behavioral Sciences, 41(3), 249–267.
Interfaces
Main upstream: Methodology Paper 2, §3.1, §3.4, §5.2, §7.1; The First Critique, Chapter 1 (Introduction), §2 (the revision of the modality, the account of conservation), Chapter 4, §2 (the register); Institution Theory Z13; Political Economy, Paper 5, E27; Paper 1, S7.
Also used: the introduction to DSM-III; Mayes & Horwitz 2005; the NG88 comments table; Espeland & Sauder 2007; Espeland & Stevens 1998.
This series: S26 to S30.
Handed to Paper 6: where a shelved premise can be withdrawn or its adoption stopped.