Writing a PhD thesis review in France

There are two external PhD thesis reviewers. They should not share any co-authorship with the PhD candidate, nor affiliation.

The following note summarizes what does the doctoral school expect to find in the PhD reviews that the reviewers will write during the review time (about six to 8 weeks).

It is a public document. It will be distributed to the PhD student and the committee members before the defence. If the candidate is successful, the reviews will be part of their application dossier for academic and faculty positions in France.
It should contain a decision at the end of the review on whether the reviewer is favourable to the defence.
It is expected to be 2-3 pages long.
The typos, small mistakes, English corrections and minor comments go to a separate email sent directly to the PhD student, who will modify the thesis accordingly before the defence.
Reviewers are 100% allowed to interact with the PhD student during all the review phase for example by email, phone call… to ask for clarifications, details…

Tips on French “rapport de thèse” (reviews of a PhD thesis).

Two-step process: the PhD thesis reviews, the PhD defence.

In France, there is a preliminary step toward the PhD defence: the PhD thesis review. Usually, two reviewers are chosen and contacted by the PhD advisor(s) to write a review on the PhD thesis.

A PhD reviewer (rapporteur de thèse in French) is required to be a recognized expert in the domain of the PhD, and either be an associate professor or full professor (tenured), or a faculty member with HDR in France (HDR = a second thesis), or have an equivalent senior position (senior scientist/researcher in a lab of a company for example).

The two reviewers have about six weeks to read the PhD thesis, write their review and send it to the doctoral school and the advisor(s). They are not expected to send the review to the PhD student directly, the advisors and the doctoral school will do it. The review is an official document and the explicit approval of each of the reviewers is required for the doctoral school to allow the PhD student to defend his/her PhD. In France, if someone fails the PhD thesis for lack of scientific content for example, it’s at this point, not the day of the defence. If the two reviews are positive (they both agree that the thesis is good enough to be defended), then the doctoral school allows the student to defend his/her PhD thesis, and the second step can be organized: the PhD defence. If the reviews are not positive, then for example the PhD student will continue a few more months to improve the thesis, to address what is required by the reviewers.

The reviews can be written in French or English. In France, these reviews are part of the application dossier for any faculty or academic position (assistant professor for example). A review is expected to be about 2 to 3 pages long.

What to do / not to do in a PhD review: Separate clearly the official part for the doctoral school (2-3 pages) on the thesis, sent to the doctoral school and the advisors, then a separate email sent to the PhD student with the English mistakes, typos and minor comments.
It is possible (allowed) for a reviewer to invite the PhD student for a seminar in his/her lab so that the PhD student explains some parts of his/her PhD. It is less common nowadays.
Why accept or decline to review a PhD?: When receiving the invitation to review a PhD thesis, the thesis itself is not finished, hence it is not possible to decide to review or not based on the PDF of the thesis. However, the invitation should contain the list of publications and preprints on which the thesis will be based, together with a link to these papers. -> A potential reviewer should check that he/she is competent to review the thesis based on these papers (at least competent for part of the thesis / some of the papers). Reviewing a PhD is a lot of work, if the contacted expert does not have time, he/she can decline and maybe suggest other names, like for reviewing a paper.

Writing a PhD review/report

In the beginning review, there should be an explanation on which parts (topics, chapters) of the thesis the reviewer is competent (expert), and on which this is not the case. The set of reviewers is supposed to cover the whole PhD thesis. Usually each PhD reviewer knows who is the second reviewer when receiving the manuscript to review.
The main sentence is the conclusion at the end: a positive review should have a mention saying that the reviewer agree and is in favour of the defence. (in French, it looks like “Je suis très favorable à la soutenance de thèse”).
The audience of this report are the committees and jury members for faculty positions (in France), who are computer scientists but not necessarily cryptographers (they are not expert on the topic of the PhD).
Usually the review starts by explaining the context and the content of the thesis, of the chapters, without judgment (without evaluating the work)
The report continues with an evaluation chapter by chapter.
Conclusion: assessment of the work done by the thesis, what are the contributions, what are the good sides, what is missing (in terms of references or of technical content)
Time needed for a thesis review: about the equivalent of two weeks full time.
In England the reviews are confidential, in France they are public.
The review is important in France for the faculty position hiring committees as it is a review from a specialist, independent of the PhD student and of the advisors.
later, a hiring committee member can contact a PhD reviewer and ask precise questions on the thesis.

What shall a reviewer do if the thesis is insufficient for a PhD. In France, a borderline PhD cannot be changed into a Masters degree.

the reviewer shall contact the advisor(s) in advance (do not write an email to the PhD student a Friday evening)
consider a “mild” review, or
ask (by email) modifications and improvements of the thesis, and assess again the resubmitted manuscript, with a non-enthusiastic review.
When agreeing in reviewing the manuscript, the reviewer engages in sending the review, otherwise, the thesis cannot be defended without two positive reviews!