Non DubitoA Mind in Many Languages
Search搜索搜尋
EN
DeutschFrançaisEspañol日本語한국어

Who Wrote the Standard?谁写下了这个标准誰寫下了這個標準 · 09

Who Questions the Successful Reform?

改革成功以后,谁来问新标准

改革成功以後,誰來問新標準

Having questioned once does not establish independence from every future premise. Existing safeguards may remain useful when roles change, but the relationship needs another examination.

曾经敢问,不等于今后对所有前提都独立。角色变了,原有保护可以保留,独立的结论仍需重新核。

曾經敢問,不等於今後對所有前提都獨立。角色變了,原有保護可以保留,獨立的結論仍需重新核。

Han Qin ·

A group spends years challenging research practices and finally changes them. Journals adopt new rules; training and forms follow. The reformers become the people best equipped to maintain the result. Success is no disgrace, but a new question appears: who can question the new premises?

The source offers no inevitable tale of reformers becoming oppressors. Drafting, implementing, and maintaining are signals to investigate, not proof that independence has vanished. The narrower question concerns whether resources, standing, and evaluation now require the new premise to remain valid.

Independence is not a permanent medal worn by an institution. It is a relationship to particular premises. An organization can be independent of an old classification and differently placed toward its replacement, or still protected adequately. Independence from what must be stated.

Registered Reports provide a real example. The Center for Open Science describes review of protocols before outcomes and conditional in-principle acceptance, not revoked because of results alone. Protocol adherence and quality assurance still matter. This describes a publication arrangement, not proof of all its intended effects.

A new method can address old problems and acquire questions of its own. Which studies suit advance specification? How should justified departures be handled? Could new blind spots arise? Asking need not mean wishing to restore an absence of transparency or review.

The Center's TOP framework also includes maintenance and improvement through feedback. This shows continuing work rather than a single declaration. It certifies neither independent review of every new premise nor inevitable dependence through maintaining the same framework. Both conclusions would outrun the arrangement described.

The relationship needs checking again. Resources once independent of the old premise might now depend on demonstrating the reform's success. Conversely, protected tenure, funding, and publication rights may still support reporting failure honestly. Existing evidence should be examined rather than discarded for a dramatic conclusion.

Rechecking does not require reversing the verdict. It asks what safeguards still protect and what they miss. Protection against arbitrary external removal may not block evaluation for insufficient defense of a reform. Or actual arrangements may already prevent that pressure effectively.

One observable distinction concerns whether finding failure counts as doing the job. An evaluator commissioned to report accurately can succeed by finding the program ineffective. A unit whose continuation requires proving effectiveness faces a different threat. Both measure; their dependencies differ.

Another concerns closure. Can maintainers alone block evidence, prohibit an agenda, or declare review finished beyond reopening? Good intentions cannot fill the independent role if they can. Where they cannot, the actual restraint deserves recording rather than deletion to preserve a criticism.

Internal separation is a familiar response: one team promotes and another evaluates. It may help without certifying independence through separate offices. Appointments, resources, agenda, publication, and unilateral closure require examination. Different personnel alone cannot answer all these questions.

A 2025 FEDS Notes article by Kay and Migueis discusses different analysts drafting and evaluating rules to mitigate independence concerns. Its footnote notes that both groups ultimately answer to SEC leadership. This reports the authors' caution, not a finding that all SEC evaluation lacks independence.

The example supports insufficiency of personnel separation alone, not impossibility within one institution. Internal evaluators can possess protected tenure, resources, and publication rights; outside bodies can share one controller. An organizational address cannot substitute for function.

Recusal is another distinction. Avoiding review of one's own manuscript may address a specific conflict without establishing independent review of disciplinary premises. Both tasks can matter. Performing one bounded action cannot certify a different scope merely because both involve independence.

Shared training is not incriminating by itself. People with common qualifications can ask the deepest questions; unfamiliar outsiders do not automatically judge independently. Expertise and dependence are different matters. Excluding every insider would not remove the need for an actual inquiry.

A qualification requiring that a premise never be denied changes the case. So does promotion conditional on proving its validity. Those requirements need evidence. Similar membership lists cannot supply a prediction that everybody must protect the same answer. Association alone is insufficient.

Following current rules is also not promising their permanent truth. A researcher can use a procedure while retaining authority to record failure and seek revision. If accurate criticism still counts as competent work, shared practice and independent questioning can coexist.

The internal change need not be called capture. No purchase or outside assimilation is required. A successful inquiry can place its questioner in a maintenance role. Identifying changed roles and dependencies does not rename achievement a conspiracy or presume an external villain.

Success is not even necessary. Imagine funding conditional on proving a proposed method valid before anyone adopts it. The relevant dependence already exists. Conversely, success followed by effective separation of maintenance and evaluation cannot automatically establish lost independence. The relation, not the milestone, matters.

A strong claim must admit counterexamples. Effective separation of appointments, resources, agenda, and outcomes can refute the proposition that participation in maintenance always destroys independence. Renaming every working safeguard a subtler dependence would make the criticism impossible to disprove.

The source lists psychiatric classification, evidence-based medicine, research-design reform, and open science as cases to examine, not completed verdicts. Change and nonchange cannot supply missing evidence about appointments, funding, or assessment. A reform history is not automatically an independence investigation.

Auditing already has traditions concerning self-review and changed circumstances. This essay claims no first discovery that independence is particular. Its focus shifts toward disciplinary premises: what counts as evidence and a valid inquiry, rather than one set of accounts alone.

Borrowing an idea from auditing does not provide every discipline a ready design. Rotation, cooling-off periods, and incompatible roles may have costs. Their existing use requires records rather than an advance declaration that all fields lack safeguards. A research question is not a universal accusation.

One worry is that repeated questioning prevents stability. Yet stability need not mean immunity, and everyday use need not mean voting again daily. A usable standard and a revisable premise can coexist. Scope, authority, and timing are the arrangements still required.

Another worry concerns hostile attempts to reopen settled issues indefinitely. Permanent exemption for maintainers is not the only response. Reception, screening, and reasons for declining review can be procedural and answerable, rather than determined solely by which side the questioner appears to represent.

Reformers' dignity need not require being right forever either. A method can genuinely improve work and later need revision. Admitting a new boundary does not make earlier effort worthless. Separating people from a method's victory or defeat may also make truthful reporting less dangerous.

The essay appoints no unchanging final court. Questioners acquire new duties, and safeguards face new facts. Reform's success cannot close inquiry; it reopens the question of independence in a changed relationship. The explanation belongs to that relationship now.

The standard may be used tomorrow and its reformers may continue maintaining it. A little room can remain: if evidence someday stops supporting it, can the person reporting that still be understood to have done the job, rather than betrayed the reform?

一群人花了多年,终于让自己质疑的研究规范改变。新规则进入期刊,培训与表格也跟着更新,他们成了最熟悉规则、最有能力维护它的人。成功没有什么可耻,却带来一个新的问题:现在谁能问这套新规则?

原作不从这里讲改革者必然变成压迫者的故事。参与制定、实施或维护,都只是需要核查的信号,不是已经失去独立的证明。问题比讽刺更具体:他们的资源、资格与评价,是否开始以新前提必须有效为条件?

独立不是机构永久佩戴的勋章,而是相对于一组前提的关系。同一个机构对旧分类可能独立,对自己维护的新方法则可能不同;也可以仍有足够保护。说“我们一直很独立”,若不说对什么,就还没说完。

开放科学提供一个实际入口。开放科学中心介绍的注册报告,把研究方案的同行评审放在结果出现之前,提供有条件的原则性接收。结果本身不决定撤销接收,仍有遵守方案与质量保证等条件。这里转述的是出版形式,不把设计目的直接当效果证明。

它让一些原先的问题有了新的处理方法,但新的方法也可以被问:哪些研究适合事先固定方案,怎样处理合理改动,是否产生新的盲点?提出这些问题,不等于希望回到没有透明与审查的状态。

开放科学中心的TOP框架也有持续维护与反馈改进的安排。这说明改革不是只写一次宣言;却不能由存在改进安排,就自动确认针对任何新前提的复审都独立,也不能由维护同一框架就预定不独立。

要重新看的,是关系。以前审旧标准时,机构的存续不依赖旧标准成立;如今若以证明新标准成功才获得经费,结构便不同。反过来,原有任期保护、独立经费与公开权若仍能支持如实报告失败,也应照实保留。

所以“需要重核”不等于结论一定反转。它要求问已有证据还护住什么,又漏了什么。防外部任意撤换的安排,不一定防住内部以“维护改革不力”作评价;也可能实际已有足够规则让这种评价不起作用。

一个可观察的区别是:测量失败算不算尽职?受命如实评估的单位,发现方案无效仍完成工作;若单位只有证明方案有效才能生存,失败报告就威胁自己。两者都在测量,同样名称却不是同样依赖。

另一个区别是,谁能让问题结束。维护者能否独自挡住材料、禁止议题,或者宣称已经评估完而无人可再开?若能,独立就不是靠善意补齐的。若不能,实际限制也值得记录,而非为了批评把它抹去。

最常见回答是内部分工:一组推广,另一组评估。它可能有用,却不能只靠换办公室就认证独立。谁任免评估者,谁控制资源和议程,谁能压住结果,谁可单独宣布结案,需要逐项看。

2025年Kay与Migueis发表在美联储网站的研究短文,讨论规则的事后评估,提到让不同分析员参与起草与评估来缓解独立性问题;脚注又提醒,两组最终仍向SEC领导层负责。这里取作者的提醒,不据此断言SEC所有评估都不独立。

这一例支持的是人员分开不自动足够,而不是同一机构内永远做不到。内部评估位置可以有受保护任期、独立资源、报告公开权;外部机构也可能受同一个控制者支配。组织地址不能代替功能。

回避也需要分清。一个人在某篇稿件中不参与审理,可以处理该次利益冲突;它不能单独证明整组学科前提已经有独立复审。两种工作都可必要,不必拿一个做过的动作,代证另一个范围。

共同训练也不是罪证。受过相同教育、持相同资格的人,可以提出最深入的反问;不熟悉问题的外人,也不自动更有独立判断。专业知识与依赖关系是两件事,不能用排除全部内行来省掉核查。

若资格的维持明确以不得否定某前提为条件,就不同了。若晋升只容许证明它有效,也有特定依赖。需要查的是这类要求有没有,而不是从会员名单相似,就猜所有人必然守同一个答案。

依现行规则做事,也不等于保证它永远正确。一个研究者可以暂照通行程序完成项目,同时有权记录失效、提出修订。若如实指出失败仍算尽职,共同实践与独立追问就可能并存。

这种内部变化也不必叫外部俘获。没有谁收买,没有利益集团来同化,改革成功本身就可能把追问者放到维护位置。指出角色与依赖的内在变化,不是把成功改名为阴谋。

成功甚至不是必要条件。设想新方法尚未被采用,起草机构却已只有证明它有效才拿得到下一轮资助,相关依赖已经出现。反过来,改革成功后若维护另有安排、评估有效分离,也不能断言独立已丧失。

因此最强的断言应容许反例。若已有分离在任免、资源、议程和结果上实际有效,就能反驳“参加维护便不独立”。不能见到有效保护,又把它改叫更高明的依赖,让自己的论点永远不会错。

原作把精神医学分类、循证医学、研究设计改革与开放科学都列成待核的案,而不是已判案例。改了或没改,不能替工作组的任免、经费、评价关系提供答案;一项改革史也不等于一份独立性调查。

审计领域早已有自我评价与情况变化后重核的传统,因此本文也不声称首次发现独立必须具体。它的重点是把问题推进到学科运行的前提:被问的不是一个账目,而是哪些结果算证据、怎样的研究算成立。

不过,从审计借来思路,不自动给每门学科现成办法。轮换、冷却期、禁止兼任,都可能有代价;有没有已经在做,需要查记录,不能先宣布所有领域都缺少。这里留的是研究问题,不是普遍控诉。

人们可能担心,这样问下去改革永远无法稳定。可稳定不等于不能被问,日常使用也不等于每天重投一票。可用的标准与可重开的前提可以同时存在,真正需要安排的是适当范围、权限和时机。

另一种担心是,恶意者借独立之名不断重开已解决问题。这个风险也不能靠给维护者永久免检解决;谁受理、怎样筛选、怎样说明不重开的理由,仍应有可问的程序,而非只看提问者站在哪一边。

改革者自己的尊严也不依赖永远正确。一套方法可以真切改善过工作,后来仍需修改;当年的努力不会因承认新边界就全成徒劳。把人从方法的胜败里分出来,也可能让如实报告不再那么危险。

这篇没有任命一个永不变化的终审者。追问者也会换职责,新的保护也须接受后来事实。改革的成功不是关闭问题的理由,只是把下一轮独立问题重新打开,要求大家按新关系说明。

新标准明天还会用,曾经推动它的人也仍可以维护。留下的一点余地是:若有一天材料不再支持它,报告这件事的人,能否仍被当作完成了工作,而不是背叛了改革?

一群人花了多年,終於讓自己質疑的研究規範改變。新規則進入期刊,培訓與表格也跟著更新,他們成了最熟悉規則、最有能力維護它的人。成功沒有什麼可恥,卻帶來一個新的問題:現在誰能問這套新規則?

原作不從這裡講改革者必然變成壓迫者的故事。參與制定、實施或維護,都只是需要核查的信號,不是已經失去獨立的證明。問題比諷刺更具體:他們的資源、資格與評價,是否開始以新前提必須有效為條件?

獨立不是機構永久佩戴的勳章,而是相對於一組前提的關係。同一個機構對舊分類可能獨立,對自己維護的新方法則可能不同;也可以仍有足夠保護。說「我們一直很獨立」,若不說對什麼,就還沒說完。

開放科學提供一個實際入口。開放科學中心介紹的註冊報告,把研究方案的同行評審放在結果出現之前,提供有條件的原則性接收。結果本身不決定撤銷接收,仍有遵守方案與品質保證等條件。這裡轉述的是出版形式,不把設計目的直接當效果證明。

它讓一些原先的問題有了新的處理方法,但新的方法也可以被問:哪些研究適合事先固定方案,怎樣處理合理改動,是否產生新的盲點?提出這些問題,不等於希望回到沒有透明與審查的狀態。

開放科學中心的TOP框架也有持續維護與回饋改進的安排。這說明改革不是只寫一次宣言;卻不能由存在改進安排,就自動確認針對任何新前提的復審都獨立,也不能由維護同一框架就預定不獨立。

要重新看的,是關係。以前審舊標準時,機構的存續不依賴舊標準成立;如今若以證明新標準成功才獲得經費,結構便不同。反過來,原有任期保護、獨立經費與公開權若仍能支持如實報告失敗,也應照實保留。

所以「需要重核」不等於結論一定反轉。它要求問已有證據還護住什麼,又漏了什麼。防外部任意撤換的安排,不一定防住內部以「維護改革不力」作評價;也可能實際已有足夠規則讓這種評價不起作用。

一個可觀察的區別是:測量失敗算不算盡職?受命如實評估的單位,發現方案無效仍完成工作;若單位只有證明方案有效才能生存,失敗報告就威脅自己。兩者都在測量,同樣名稱卻不是同樣依賴。

另一個區別是,誰能讓問題結束。維護者能否獨自擋住材料、禁止議題,或者宣稱已經評估完而無人可再開?若能,獨立就不是靠善意補齊的。若不能,實際限制也值得記錄,而非為了批評把它抹去。

最常見回答是內部分工:一組推廣,另一組評估。它可能有用,卻不能只靠換辦公室就認證獨立。誰任免評估者,誰控制資源和議程,誰能壓住結果,誰可單獨宣佈結案,需要逐項看。

2025年Kay與Migueis發表在美聯儲網站的研究短文,討論規則的事後評估,提到讓不同分析員參與起草與評估來緩解獨立性問題;腳注又提醒,兩組最終仍向SEC領導層負責。這裡取作者的提醒,不據此斷言SEC所有評估都不獨立。

這一例支持的是人員分開不自動足夠,而不是同一機構內永遠做不到。內部評估位置可以有受保護任期、獨立資源、報告公開權;外部機構也可能受同一個控制者支配。組織地址不能代替功能。

回避也需要分清。一個人在某篇稿件中不參與審理,可以處理該次利益衝突;它不能單獨證明整組學科前提已經有獨立復審。兩種工作都可必要,不必拿一個做過的動作,代證另一個範圍。

共同訓練也不是罪證。受過相同教育、持相同資格的人,可以提出最深入的反問;不熟悉問題的外人,也不自動更有獨立判斷。專業知識與依賴關係是兩件事,不能用排除全部內行來省掉核查。

若資格的維持明確以不得否定某前提為條件,就不同了。若晉升只容許證明它有效,也有特定依賴。需要查的是這類要求有沒有,而不是從會員名單相似,就猜所有人必然守同一個答案。

依現行規則做事,也不等於保證它永遠正確。一個研究者可以暫照通行程序完成項目,同時有權記錄失效、提出修訂。若如實指出失敗仍算盡職,共同實踐與獨立追問就可能並存。

這種內部變化也不必叫外部俘獲。沒有誰收買,沒有利益集團來同化,改革成功本身就可能把追問者放到維護位置。指出角色與依賴的內在變化,不是把成功改名為陰謀。

成功甚至不是必要條件。設想新方法尚未被採用,起草機構卻已只有證明它有效才拿得到下一輪資助,相關依賴已經出現。反過來,改革成功後若維護另有安排、評估有效分離,也不能斷言獨立已喪失。

因此最強的斷言應容許反例。若已有分離在任免、資源、議程和結果上實際有效,就能反駁「參加維護便不獨立」。不能見到有效保護,又把它改叫更高明的依賴,讓自己的論點永遠不會錯。

原作把精神醫學分類、循證醫學、研究設計改革與開放科學都列成待核的案,而不是已判案例。改了或沒改,不能替工作組的任免、經費、評價關係提供答案;一項改革史也不等於一份獨立性調查。

審計領域早已有自我評價與情況變化後重核的傳統,因此本文也不聲稱首次發現獨立必須具體。它的重點是把問題推進到學科運行的前提:被問的不是一個帳目,而是哪些結果算證據、怎樣的研究算成立。

不過,從審計借來思路,不自動給每門學科現成辦法。輪換、冷卻期、禁止兼任,都可能有代價;有沒有已經在做,需要查記錄,不能先宣佈所有領域都缺少。這裡留的是研究問題,不是普遍控訴。

人們可能擔心,這樣問下去改革永遠無法穩定。可穩定不等於不能被問,日常使用也不等於每天重投一票。可用的標準與可重開的前提可以同時存在,真正需要安排的是適當範圍、權限和時機。

另一種擔心是,惡意者借獨立之名不斷重開已解決問題。這個風險也不能靠給維護者永久免檢解決;誰受理、怎樣篩選、怎樣說明不重開的理由,仍應有可問的程序,而非只看提問者站在哪一邊。

改革者自己的尊嚴也不依賴永遠正確。一套方法可以真切改善過工作,後來仍需修改;當年的努力不會因承認新邊界就全成徒勞。把人從方法的勝敗裡分出來,也可能讓如實報告不再那麼危險。

這篇沒有任命一個永不變化的終審者。追問者也會換職責,新的保護也須接受後來事實。改革的成功不是關閉問題的理由,只是把下一輪獨立問題重新打開,要求大家按新關係說明。

新標準明天還會用,曾經推動它的人也仍可以維護。留下的一點餘地是:若有一天材料不再支持它,報告這件事的人,能否仍被當作完成了工作,而不是背叛了改革?

Source & reading note原论文与阅读说明原論文與閱讀說明

Ten essays draw on the Disciplines prequel and nine papers. Unattributed applicants, scores, meetings and procedural handovers are imagined. Linked texts, studies and rules retain their dates and evidential limits. The claims that divisions leave something uncontained and that people are more than tools are openly assumed philosophical and normative starting points, not empirical laws. Evidence assessment can itself consider values; professional boundaries, limited authority and reasoned retention are not failures by definition. Missing complete cases are not supplied as established findings; unverified powers are not declared absent, and an unsuccessful search establishes no frequency. These essays prescribe no diagnosis, lending cutoff or clinical action and offer no legal or professional operating advice. English is independently written.

十篇据《学科论》前置篇及九篇正文改写。窗口、评分、会议与程序交接中未另注出处的人物和场景为设想;真实文本、研究与规则另列来源,并注明材料日期和能证明的范围。划分总有未尽、以及人不只是工具,是原作公开承担的哲学与规范起点,不冒充经验定律。证据审查本身可以处理价值,专业边界、有限授权和审后保留不等于失职。原文没有查得的完整实例不补成已证事实,未核不写成不存在;没有发现实例也不据此判断发生频率。这里不决定诊断、信贷门槛或医学处置,不作法律与专业操作建议。简繁中文同文,英文独立重写。

十篇據《學科論》前置篇及九篇正文改寫。窗口、評分、會議與程序交接中未另注出處的人物和場景為設想;真實文本、研究與規則另列來源,並註明材料日期和能證明的範圍。劃分總有未盡、以及人不只是工具,是原作公開承擔的哲學與規範起點,不冒充經驗定律。證據審查本身可以處理價值,專業邊界、有限授權和審後保留不等於失職。原文沒有查得的完整實例不補成已證事實,未核不寫成不存在;沒有發現實例也不據此判斷發生頻率。這裡不決定診斷、信貸門檻或醫學處置,不作法律與專業操作建議。簡繁中文同文,英文獨立重寫。

Read the source paper ↗阅读原论文 ↗閱讀原論文 ↗

Center for Open Science: Registered Reports format and conditions开放科学中心:注册报告的条件与形式開放科學中心:註冊報告的條件與形式 ↗

Center for Open Science: TOP framework and maintenance开放科学中心:TOP框架与维护说明開放科學中心:TOP框架與維護說明 ↗

Kay and Migueis, 2025 FEDS Notes on ex-post evaluationKay与Migueis,2025年规则事后评估研究短文Kay與Migueis,2025年規則事後評估研究短文 ↗