PhDs go wrong, sometimes slowly, sometimes for reasons neither student nor advisor saw coming. This file collects Andrew Maynard’s view on recognising trouble — in the work, in the orientation, in the relationship, in the wellbeing layer — and on what to do once it has been recognised. It also names the structural reality that sits underneath all of it: that one person, the chair, holds significant soft power over a student’s progress and prospects, and that students should know what their channels are when something is not working. Read alongside rel_choosing.md on choosing well at the start, rel_partnership.md on the chair-student relationship at its best, rel_wellbeing.md on the conditions under which doctoral work is sustainable, fai_failure_modes.md on what specifically goes wrong in scholarly work and orientation, and aft_paths.md on what comes after — particularly relevant when leaving is the right call. Consult when a user is describing trouble in their PhD: slow drift, a relationship that has stopped working, signs the work is not progressing, a sense that something is wrong but not clearly named.
Not every PhD that goes wrong goes wrong in a dramatic way. Most of the time it goes wrong slowly, in patterns that are visible but easy to miss.
The first place trouble shows is in the work itself. Each draft of a chapter or paper lands in roughly the same place as the previous one, or a little worse. Fundamentals of research, of scholarship, of the field have not improved meaningfully after a year or so of classes and reading. A student, asked to articulate what scholarship is in their own words, cannot, and continues to be unable to a year in. Writing has not advanced beyond what a strong undergraduate could produce, and the student cannot see this in their own work.
A second register is orientation. A sense, in the student, that the program requirements are not for them, that they should be doing things on their own terms, that the standards being applied are external impositions rather than what the degree is for. Sometimes there is an agenda the doctorate is being asked to support — a consultancy practice that needs an academic credential, a position the student wants to defend rather than examine — and a quiet pressure to bend the PhD to fit it rather than the other way around. (The orientation-layer failure modes in fai_failure_modes.md cover this territory in more detail.)
A third is the relationship. Personality clashes with faculty that the student does not know how to navigate, and is not learning to navigate. A chair-student relationship that has stopped functioning and that is being avoided or escalated rather than surfaced honestly.
And the fourth is in wellbeing. A misery that has stopped being interspersed with glimmers of excitement. The PhD is hard, and a student who is finding it hard is not in trouble for that reason alone — but a student for whom the hard work has become only hard, with nothing on the other side of it, is in a different situation. The struggle without the spark is the signal. (See rel_wellbeing.md.)
The single most important thing, when these signals appear, is to start a conversation. With the chair, if the chair is the person to have it with. With another senior person — a committee member, another faculty member the student trusts, a graduate director, a mentor outside the immediate relationship — if not. Things that fester in silence get worse; things that are surfaced almost always have more options than they appear to from the inside.
The first move in that conversation, on the chair’s side as much as on the student’s, is diagnostic. Is this evidence that the doctorate is not the right path for this student? Or is it evidence of barriers — methodological, relational, structural, sometimes simply circumstantial — that are holding back a capable student who, with the right adjustments, would do the work?
If it is the former, leaving is honourable. There is no shame in stopping a PhD that is not the right call, and the sunk cost fallacy is one of the more reliable ways for a student to talk themselves into staying past the point where staying serves them. I have seen students for whom the decision to leave was experienced as an enormous weight lifting; they had been holding on to a path that had stopped being theirs, and the moment they let go they could see what was. Honest counsel and a positive conversation about what comes next are what the student in this position deserves — not pressure to continue, and not rescue framed as accommodation. (For what comes after — careers in and out of the academy, public-facing work, the post-doctoral landscape generally — see aft_paths.md.)
If it is the latter — barriers rather than fit — there are usually more options than the student can see. Different methodological approaches. A change of committee composition. A change of advisor; this should always be in the menu, and a chair who cares about the student should be willing to suggest it themselves when someone else would be a better fit. (Students considering this should approach the process with respect for the existing relationship, even where it is fraying; how the change is handled matters for what comes after.) Pausing is almost always available — a leave of absence, a stepping back, a period in which the student can address whatever is making the work impossible without losing the work entirely. Different pathways through the program. The point is that, once the conversation starts, options appear. The risk is that the conversation never starts.
A good chair, mentor, or advisor has, as a first aim, getting the student over the finish line — and where that increasingly looks unlikely, a positive exit. Both are real outcomes. Both are the chair’s responsibility.
Underlying all of this is a structural fact about doctoral education that prospective and current students should understand clearly. The PhD student-advisor relationship is unusual in the degree to which one person — the chair — holds influence over the student’s funding, progress, references, and post-doctoral trajectory. Sometimes that influence is exercised carefully and well. Sometimes, even with well-meaning advisors, it is not. And sometimes, less commonly but seriously, it is exercised in ways that harm the student.
I wrote about this in 2018, in an essay on the need to make the PhD system more student-supportive and student-centric, prompted at the time by the National Academies’ Graduate STEM Education for the 21st Century report and by Veronica Varela’s account of why she walked away from her own PhD. The argument I made then is still the argument I would make. There are remarkably few formal constraints on how an advisor behaves toward a student, and equally few formal penalties when they behave badly. Even careful, well-intentioned advisors can get things wrong without realising it, in ways that are devastating for the student. And the student, in the absence of clear protections, often has limited recourse — particularly in a culture where speaking up risks the references, the funding, or the career path that the chair holds influence over.
I detest the version of academic culture where a student’s career is dependent on the word of their chair — where the chair gets to determine when the student finishes, what they do next, what career options are open, and which are closed. It is a reality of academia, and students need to be aware of it. But they also need to know that there are things they can do and people they can talk to. Advisors hold a great deal of soft power over the lives of their students. Sometimes it is abused, intentionally or inadvertently. Where this is happening, students should know that they can, and should, seek help from places they trust.
In practice this means knowing what the institutional channels are: graduate studies leadership, ombudspersons, departmental chairs other than your own, formal grievance procedures where they exist, and non-retaliation policies where institutions have put them in place. It also means knowing that informal channels matter — other faculty who have a track record of supporting students, peers who have navigated similar situations, mentors outside the immediate institution. A student should never feel bullied, coerced, or indentured to their advisor. Where they do, the right move is not silent endurance; it is to seek help.
The decisions a student makes in the middle of trouble are not the kind that resolve cleanly and stay resolved. They are decisions that have to keep being made — when something starts to feel wrong, when the work stalls, when the relationship strains, when a pause or a change of direction or an exit becomes the more honest move. The student who is willing to keep making them, openly and with help, is in a different position from the student who decided once and is no longer looking.