Public standard

Methods & Ethics

How this site knows what it claims to know — and where it refuses to pretend certainty. Threads of Connection is a living historical record of Muslims in the American story: educational, nonpartisan, and checkable.

Version 2026.07 · Updated July 2026 · By Saeb Ahsan

Mission of evidence

The site documents Muslims as integral to the American story — from colonial and founding-era records through institutions, migration, and civic life. It privileges primary documents where they exist; names historians and institutions when it relies on them; and treats American Muslims as many communities, not one voice.

We do not run campaigns, score politicians, or flatten contested history into slogans. The through-line is the record: liberty, law, labor, literacy, service, and place.

Source hierarchy

When sources conflict, higher tiers outrank lower ones. A claim should not rest solely on the lowest tiers.

1Primary documents. Treaties, statutes, court opinions, letters, manuscripts, expedition narratives, government archives, and contemporaneous public records.
2Peer-reviewed scholarship & university presses. Named historians and monographs that can be checked (Diouf, Spellberg, Gomez, Curtis, GhaneaBassiri, Howell, and peers).
3Institutional datasets & public history. National Park Service, Library of Congress, Founders Online, Pew Research Center methodology pages, university special collections.
4Community and institutional histories. Mosque histories, museum labels, oral-history projects — useful for local texture, never sole support for a national “first” claim.
5News and encyclopedias. May corroborate or lead a reader to a primary; not sufficient alone for load-bearing primacy or population claims.

Wikipedia may pair a citation; it is never the sole support for a primacy claim. An institution’s own site may corroborate but does not alone prove its own “first” status (e.g., Mother Mosque: NPS first, community history second).

Evidence Label Legend

A compact key to the five source-evidence badges used sitewide. Each fact, caption, or claim that carries a badge uses one of these labels — never an unlabeled or invented tier. If a claim genuinely rests on two tiers, the stronger one is cited and the second is noted in the surrounding text rather than stacking badges.

Primary documentComes straight from an original record — an archival document, a government or institutional file, a manuscript, or a photograph/artifact held by a library or museum.
Scholarly workComes from an academic book or journal article that has been reviewed by other scholars before publication.
Institutional sourceComes from a survey, dataset, or public-history page built by a research institution or agency (for example NPS, Pew, the U.S. Census, or ISPU) — modeled estimates are always labeled as such, never presented as headcounts.
Oral history / archiveComes from a spoken account — an interview, an oral-history recording, or a family or community narrative — rather than a written record, used where written sources are thin or silent.
Interpretive synthesisA reasoned conclusion drawn from indirect evidence, offered as informed interpretation rather than a documented fact, and always flagged as such.

Inclusion criteria

Timeline entries

A dated event with a checkable source, year-ordered, with a stable title (anchors derive from titles). Body text is limited to what the source supports; “more” fields may add context without inventing certainty.

Landmarks & map pins

Prefer NRHP-listed or otherwise clearly documented sites with founding or dedication dates, place, and historical significance. Coordinates are city- or park-level approximations, never private street addresses for living communities when avoidable.

Claims on cards and briefs

Representation facts (who, when, oath, artifact) over platform characterizations. Religiously mixed diasporas are framed as Muslim communities within larger migrations, not wholesale “Muslim immigration.”

Handling uncertainty

The U.S. Census does not ask about religion. Historical population figures for Muslim Americans are estimates; modern figures are surveys, models, or projections. On this site, numbers carry epistemic labels:

Historical estimate — derived from scholarship and incomplete records; often a range.
Survey (verified) — published methodology from a named research organization.
Modeled estimate — demographers’ synthesis; may run higher or lower than a survey headline.
Projected — future scenario, never dressed as a present count.

Where two reputable series disagree (for example Pew and ISPU race categories), we show both rather than average them into a false consensus.

Fact, inference, and interpretation

Documented fact. Something the named source states: a treaty text, a manuscript page, a listing date, a survey percentage.
Bounded inference. A careful conclusion that follows from the documents but is not a quotation — labeled with “among the first,” “commonly identified as,” or similar hedges when the record is incomplete.
Interpretation / debate. Historians’ competing readings (musical influence hypotheses, exact expedition routes, “first mosque” definitions). Presented as disagreement, not as settled truth.

What historians debate

Disagreement is not a defect in the record — it is part of the record. This site surfaces debates instead of papering them over. Five recurring, concrete examples:

What was “the first mosque in America”?

It depends entirely on the definition used, and institutions do not agree on which should win: first organized congregation, first purpose-built building, or oldest surviving structure in continuous use all point to different sites and years. This site does not declare a winner between “first built” and “oldest surviving” — both facts can be true simultaneously, which is exactly why the question is a definitional debate rather than a factual dispute.

Is there continuity between enslaved-era Muslim communities and 20th-century immigrant Muslim communities?

Whether the Muslim presence documented among enslaved Africans in colonial and antebellum America connects, in any traceable institutional or community sense, to the immigrant communities that built mosques and organizations in the 20th century is an open historiographical question. The documentary record for enslaved-era practice largely goes silent after emancipation, and the 20th-century mosque-building movement is documented as arising from new immigration rather than a demonstrated unbroken line back to antebellum communities. This site asserts neither continuity nor a clean break — the honest position is that the question remains open.

Did Ottoman or broader Islamic-world pluralism influence Locke or the American Founders?

Some writers argue that religious pluralism in the Ottoman Empire or other Islamic-world political models influenced European Enlightenment thinkers like Locke, and through them the Founders' approach to toleration. This claim circulates but is contested and speculative rather than an established historical finding — mainstream scholarship has not established a direct causal link, and this site treats it as a noted, open debate rather than a settled fact.

How many Muslims live in the United States today?

There is no single agreed-upon current figure. Pew's 2017 estimate (roughly 3.45 million, about 1.1% of the population) remains the most recent published absolute headcount; Pew's 2023–24 Religious Landscape Study (released February 2025) instead reports a population share (about 1% of U.S. adults) and does not update the headcount. This site cites both, labeled for what each measures, rather than treating the newer share as a revision of the older count.

Other recurring patterns: route reconstructions for early Southwest expeditions have documentary cores but contested mile-by-mile paths; Spanish-era identity labels (“Moor,” baptismal records) do not always map cleanly onto modern religious categories; and split legal authorities (court majorities and dissents) both appear when both shaped the public meaning of an event.

Representation and ethics

American Muslims are not a monolith — race, theology, region, and generation differ.
Heterodox and separate movements (when historically relevant) are noted for accuracy, not for ranking authenticity in a religious sense.
Enslaved persons are named as scholars, parents, and workers when the record allows; archive gaps (especially women’s literacy) are named rather than filled with invention.
Decorative or AI-era images are labeled illustrative; archival scans are credited as public-domain documents.
Content is CC BY 4.0; code is MIT. Attribution: “Threads of Connection — usmuslim.org.”

Where the record is uneven, and why

The historical record this site draws on is not evenly distributed, and neither is what has survived or been digitized. Where these gaps affect a specific claim, the site tries to say so directly rather than presenting an artificially even picture:

Geographic bias. Documentation clusters around a handful of cities and states with active archives, museums, or historical societies; communities without an institutional archive are underrepresented not because they mattered less, but because less was preserved or digitized about them.
Racial bias. Enslaved and early Black Muslim history in particular survives through fragments — a handful of manuscripts, court records, and secondhand European accounts — compared with the far more complete paper trail available for later, better-resourced immigrant communities.
Class bias. Institutions, published authors, and property owners left more durable records than laborers, domestic workers, and the poor; the record skews toward those who had the means to write, publish, or be written about.
Gender bias. Public records (deeds, mosque incorporation papers, newspaper coverage) disproportionately name men; women's roles in community-building are often recoverable only through oral testimony or secondary mentions in others' records.
Oral-vs-written bias. Communities and eras that relied on oral transmission are structurally harder to document to the same evidentiary standard as communities that left written records — not because oral testimony is less true, but because it degrades and is harder to verify at a distance in time.

How this work is checked

There is not yet a standing formal advisory board. That fact is stated plainly so readers are not misled. Work is checked through a layered review model that can grow into a named board later:

Source gate. Load-bearing claims require primary or peer-reviewed / institutional support before copy is written.
Machine inventory. A committed content-assert script checks counts, chronology, unique ids, and related invariants on every change.
Link and wording audits. Citations are re-checked against institutional pages; overclaims are revised when the record is thinner than the prose.
Open corrections channel. Readers, educators, and specialists can propose fixes; accepted changes cite the better source.

Solo editorial. One person, Saeb Ahsan, researches, drafts, and edits every page. There is no editorial board, advisory committee, or formal peer-review process, and the site does not claim one — that is a real limitation of a smaller review model than an institutional publication's, and the site is explicit about that rather than implying a process that doesn't exist.

Future enhancement: a short public roster of subject-matter reviewers (history, demography, education) who examine major updates. Until that roster exists, this page remains the authoritative statement of process.

Corrections and community feedback

Write to hello@usmuslim.org with the page and section, the claim as written, and — where you have one — a source for the correction (a link, citation, or document). Corrections without a source are still welcome, but take longer to act on, since they must be independently verified before anything changes on the live site. This is the same address the site uses for source and landmark submissions — one contact channel, not several.

What gets reviewed, and how

The claim and the proposed correction are compared against the strongest available source on each side.
If the proposed correction is better-sourced than what's currently published, the page is updated and the change is logged in the changelog.
If the existing claim already reflects the best available sourcing, the page is not changed, and the submitter is told why, where a reply address was given.
If a claim turns out to be genuinely contested among historians rather than simply wrong, the fix is usually to show the disagreement rather than to pick a side — see What Historians Debate, above.

What becomes public

Never published: a submitter's email address or other private contact information — a reply address, if given, is used only to respond.
Published when a correction is made: the change itself, on the live page, plus a changelog entry describing what changed and why. The changelog is public and permanent — corrections are not quietly folded in and forgotten.
Credit goes to the cited source, not to private correspondence, unless a submitter asks to be named and it's appropriate to the context (e.g., they authored the cited work).

Corrections & sources

Suggest a fix or a better source

Every submission is checked against the cited record before anything changes on the site. Prefer primary documents, peer-reviewed work, or institutional pages. Broken links are replaced only with independently verified URLs.

Email hello@usmuslim.org

Update and version policy

The footer carries a “Last updated” month. Major methodology or content rounds also carry a version label (currently 2026.07). Timeline titles of shipped entries do not change (stable anchors). Internal ids used as bookmarks are not renamed without cause. Broken links are fixed only with verified replacements.

Version notes

2026.07 · July 2026

Methods & credibility layer

Methods & Ethics page published as the standing public standard.
Homepage credibility strip, claims boundary, and version notes added.
Demographics and Sources surfaces show measured vs modeled labels more clearly.

This is a living record, not a closed canon. See update policy and project updates.

Research gaps and roadmap

Naming incompleteness is a trust practice. Threads that are well documented sit beside threads that are emerging or thin. Priority areas for future work:

Deeper women’s institutional histories beyond the names already profiled.
Sectarian and theological diversity treated with scholarly care, not polemics.
Southwest and Pacific Northwest archival work (especially pre-1950 Northwest landmarks).
Economic and enterprise threads with primary commercial records.
Elementary-language and multilingual access without diluting citations.
A populated scholarly/community review roster when capacity allows.

Steward notes and contribution guidelines live in the project repository; this page remains the public face of the method.

Suggested citation

Journalists, educators, and students may cite pages as follows (adapt page title and access date):

Threads of Connection. “Methods & Ethics.” usmuslim.org, version 2026.07, July 2026. https://usmuslim.org/methods. Content licensed CC BY 4.0.

For a specific claim, prefer citing the primary source the site links, then this site as the synthesis that led you there.

Source library →Demographics methods in practice →← Home