About This Research

This site is a personal research portfolio. Each entry takes a hard question, surveys the best available evidence, and lays out what it does and does not support. Finally it explores the current work on the issue.

Authorship and accountability

The author, Jason James, is responsible for all content on this site — its claims, framing, conclusions, and any errors. Nothing here is published without human review.

How it was produced

The research was produced with Claude Code. The model was used to gather and organize sources, draft content, and check internal consistency. It is a tool in the workflow: it carries no accountability for the content and is not cited as a source. The author provided the structure of the research questions and responses, and iterated each page several times with follow up questions. The iterative process first worked through big picture logic and structure of arguments, counters, and resolution. Then proceeded section by section, and page by page with dozens of rounds of follow up questions and rewriting.

Verification

Every factual claim is tied to a primary or peer-reviewed source, cited inline. Citations were independently verified — confirming that each source exists, says what it is cited as saying, and resolves to a working link — rather than taken on the model’s word. Where the evidence is genuinely contested, the text says so and presents the disagreement.

Sources and citations

Citations point to the underlying scholarship — journals, working papers, datasets, and books — not to the AI tool. External links open in a new tab. Disclosing the model here, as part of the method, follows emerging norms for LLM-assisted research: the tool gets a methods disclosure, the sources get the citations.

Privacy and data

The research draws only on public sources. No personal, confidential, or copyrighted full-text material is fed into the tooling.

Corrections

Found an error, or know a better source? Open an issue or pull request. Corrections are welcome.