Thesis Committee Conflict of Interest

The same relationship can be acceptable for a reviewer and disqualifying for an examiner. Four structural differences explain most of the gap, and they are worth understanding because they predict which cases an institution will refuse.

The outcome is a qualification with lifelong effect on a person, awarded by a public or chartered institution, not a decision to publish a paper that can be sent elsewhere. The standard of procedural fairness applied to it is correspondingly higher.

The examination is not anonymous in either direction. In journal review a conflicted reviewer can sometimes be managed by masking, but an examiner sits in a room with the candidate and signs a report under their own name, so the only available remedy is not to appoint them.

The supervisor is a party. A candidate's thesis is also, in institutional terms, an output of the supervisor's work, which is why an examiner's relationship with the supervisor is examined as closely as their relationship with the candidate. Nothing equivalent exists in journal review, where the reviewer's relationship with the corresponding author's mentor is nobody's business.

There is no appeal to a different venue. A rejected paper goes to another journal; a contested viva goes to an appeals procedure inside the same institution, which is slow, adversarial and damaging for everyone. Institutions therefore front-load the checks into the appointment.

COPE's guidance on handling conflicts of interest offers a test that transfers well: a conflict of interest is something that, if left undeclared and discovered later, would make a reasonable person feel misled. Apply that test to the examiner list before the institution does.

Regulations differ in wording but converge on a similar list. Where your institution names none of these explicitly, treat them as the cases to declare and let the doctoral school decide.

Supervision in any form

main supervisor, co-supervisor, mentor of record, or member of the monitoring committee where the regulation separates monitoring from examination. Anyone who has shaped the thesis cannot certify it in those systems.

Co-authorship with the candidate on any part of the thesis, including a chapter published as an article. This is the case that catches people, because article-based theses make co-authors of exactly the people best placed to examine.

Co-authorship with the supervisor within the window the institution applies. Where no window is stated, recent and repeated collaboration is what an institution will look at, and a single paper eight years ago is a different thing from four papers in three years.

Employment or a formal position in the same unit, department or laboratory, and in some regulations the same institution entirely.

Family, partnership, and close personal relationship. Obvious, rarely written down in detail, and the case where saying nothing is most dangerous.

A financial or employment interest in the outcome

an examiner whose company funded the work, who holds a patent the thesis supports, or who is a candidate for a post the supervisor controls.

An unresolved dispute with the candidate or the supervisor. A known adversary is as unfit to examine as a known ally, and institutions treat hostility as a conflict even though its direction is opposite.

Co-publication is where most contested appointments are decided, and it is counted differently from the way a journal counts it. Two rules cover almost everything.

Count the supervisor's record as well as the candidate's. An external examiner with no shared paper with the candidate but three recent papers with the supervisor will usually be refused, and the candidate is often the last person to notice.

Count large-collaboration authorship separately. In fields where consortium papers carry hundreds of authors, being on the same author list is not a working relationship, and most regulations that mention it exclude it. Say so explicitly in the nomination rather than leaving the institution to find the paper and draw its own conclusion.

Check both records before proposing, not after. Search the candidate's and the supervisor's publication lists for the examiner's name in a bibliographic database rather than relying on memory, since a shared conference proceedings paper from six years ago is easy to forget and trivial for an administrator to find.

Editorial relationships are usually not conflicts

having been reviewed by, or having reviewed for, the same journal is not a working relationship. Nor is having cited each other, which our peer review conflict guide also covers for the journal case.

Financial conflicts in doctoral examination are less about payment and more about who has an interest in the thesis being approved. The funded doctorate makes this concrete.

Shared funding is a conflict even without shared publication. Two researchers who are partners on the same grant have a working relationship and, frequently, a shared interest in the project's outputs looking good.

Industrially funded doctorates, such as the French CIFRE arrangement or an equivalent company sponsorship, put a firm in the position of having paid for the work. A supervisor or employee at that firm typically cannot examine, and any examiner with a consultancy relationship with the firm should be declared.

Patents and licensed technology create the sharpest cases. An examiner holding a patent that the thesis validates or threatens has a direct stake in the result, and the fact that this is rarely written into doctoral regulations does not make it acceptable.

Payment of the examiner by the institution is not a conflict. Examiners in many systems receive a fee or expenses through the university, and that is a standard arrangement, disclosed by design.

The institutional conflict is the one with the widest variation between countries, and the one candidates most often get wrong when they propose a name.

The narrow definition is the research unit or laboratory

someone from a different department of the same university may qualify as external. The wide definition is the institution: nobody employed by the awarding university qualifies, including emeritus and honorary staff.

Some regulations extend beyond the university to the site or the alliance. Federated universities, joint research units run with a national agency, and university groupings can mean that two apparently separate employers count as one institution for this purpose.

Joint or cotutelle doctorates multiply the problem, because both partner institutions are the awarding institution, and both sets of composition rules usually apply at once. The composition has to satisfy the stricter reading of both, and that is settled in the cotutelle agreement rather than improvised at the end.

Being an alumnus of the institution is generally not a conflict. Neither is a visiting position that ended years ago. Institutions are concerned with current dependency, not with biography.

The candidate's own future employment matters too

an examiner at the institution where the candidate has just accepted a postdoctoral position is in an awkward position, and it is better disclosed than discovered.

The procedural rule everywhere is the same: disclosure early is administrative, disclosure late is a crisis. The mechanics below are what makes the difference.

Declare at nomination, in writing, in the form the doctoral school provides. Most nomination forms include a declaration field, and a short factual entry there closes the question permanently.

State the facts rather than the judgement. Two co-authored papers in 2021 and 2023 with the supervisor, no shared funding, no shared institution, is a usable declaration. Believing oneself to be impartial is not.

Let the institution decide, and accept the answer. The doctoral school may approve a declared relationship with a note on file, may ask for the member to move from examiner to observer, or may refuse. All three are normal outcomes and none of them is a judgement on anybody's integrity.

If a conflict emerges between appointment and defence, tell the doctoral school the day it emerges. A new joint paper, a new job, or a grant awarded jointly are all changes of circumstance, and a substitution six weeks out is an inconvenience while a challenge afterwards is an appeal.