ITH The Institute for Transatlantic HumanitiesAn observatory of the humanities · Atlanta & Paris

Principles

Method

The rules the Institute holds itself to, and what they rule out.

This page has moved into Mission.

An observatory is only as good as its discipline about evidence. These are the founding principles. They govern the infrastructure, and they will govern anything the Institute publishes from it.

Founding data principles

Preserve evidence. Raw source material and its provenance are retained wherever it is legal and practical to do so, so that every derived observation can be audited back to what was actually captured.

Separate observation from inference. What a source explicitly states is kept distinct from what a machine classified, what an analyst inferred, and what a scholar judged. Each layer is labelled as what it is.

Canonicalise without destroying ambiguity. Scholars, works and institutions receive stable identifiers; their aliases, conflicting evidence, uncertain matches and historical changes are kept alongside them and never flattened away.

Time is first-class data. The Institute preserves historical states, and does not overwrite the past with the present classification or the current affiliation.

Provenance everywhere. Every significant relationship, classification, measurement and claim is traceable to its evidence and to the version of the method that produced it.

Version the method. A change in how something is computed is a new version, announced, with the history re-run.

Claims

The Institute distinguishes four grades of claim, observed, inferred, interpretive and causal, and holds that the stronger and more consequential the claim, the greater the burden of evidence and of human validation it must carry. A causal claim about the life of a tradition requires substantially more than a temporal association or a correlation, and the Institute will say so when it cannot meet that standard.

Independence

Research is governed by scholarly questions and evidence, not by the interests of donors, sponsors, technology providers or institutional partners. Material funding relationships and conflicts of interest are disclosed. Researchers retain control over findings, sponsored research is distinguished from sponsor advocacy, unfavourable findings are published where contract permits, and methodological limitations are published always.

Pluralism

The Institute exists to understand the humanities, not to make conformity to a preferred philosophical, political, religious or methodological position the measure of their health. Disagreement is not pathology. Heterodoxy is not automatically health, and neither is consensus. Traditional scholarship is not automatically healthier, and novelty is not automatically renewal. Popularity is not significance. Where the Institute crosses from description into judgement, it will mark the crossing.

Copyright and access

Each source is admitted under its own terms, recorded at admission. Full text stays inside the instrument where the rights require it; public products use metadata, aggregates and short quotations, each with its page. Where a source has supplied material under agreement, it is cited as the source of every finding drawn from it.