How to prepare for your thesis defense (viva)

The thesis defense, or viva voce, is the strangest exam you will ever sit. There is no syllabus beyond a document you wrote yourself, no past papers, and the examiners have already read every word before you enter the room. That inverts the usual logic of revision. You are not learning new material. You are re-learning your own, and preparing to talk about it under pressure with the two or three people who have studied it most closely.

The good news is that this is a defensible position. Nobody knows your project better than you do. Preparation is mostly about converting years of tacit knowledge into answers you can produce on demand: why this question, why this method, why these sources, and what it all adds up to.

This guide covers the six to eight weeks before the defense: how examiners read, how to re-read your own thesis, how to build and rehearse a question bank, and how to make sure the document you are defending has no soft spots you have forgotten about.

What examiners actually look for in a thesis defense

By the time you enter the room, your examiners have read the thesis and, in many systems, written an independent report on it. The defense is rarely about discovering whether the work is good. It is about confirming that you did the work, that you understand it deeply, and that you can see it the way a peer would: contribution first, limitations acknowledged, choices justified.

That means examiners are listening for ownership. They want to hear you explain why the project took the shape it did, not recite what is on the page. Strong candidates talk about decisions and trade-offs. Weak candidates summarise their own chapters as if someone else wrote them.

It also means the defense is a conversation between researchers, not an interrogation. Examiners disagree with theses they pass all the time. What they will not forgive is a candidate who cannot engage with the disagreement.

Roles differ too. Where you have both, the external examiner typically leads on the field-level questions: contribution, positioning, rival readings of the literature. The internal examiner often carries the regulations and the detail. Read both examiners' recent work with this division in mind, and you can predict a surprising share of the afternoon.

Re-read your thesis like an examiner, not an author

Start with a full cover-to-cover read, ideally in one or two sittings, a few weeks out. You will be surprised how much you have forgotten, especially in chapters written a year or more ago. Read a printed or annotated copy and mark every claim you would want to double-check before defending it out loud.

Build a page map as you go: a one-line summary of each chapter, plus the page numbers of your research questions, your key results, and any passage you already know is contested. In many systems you can bring an annotated copy into the room, and being able to turn straight to page 84 instead of flipping under pressure is worth the effort.

Then write the examiner's report yourself. List the five or six weakest points in the thesis: the small sample, the excluded case, the method you would not choose again. If you name them first, no question about them can ambush you.

Build a viva question bank and rehearse out loud

Most defense questions are predictable in form, even when the specifics are not. Build a bank of thirty or forty questions drawn from three sources: generic viva questions, your own list of weak points, and your examiners' published work, which tells you what they care about and how they think. The generic core usually includes:

• Summarise your thesis in a few minutes. What is the contribution?
• Why did you choose this methodology over the obvious alternatives?
• Which literature most shaped your approach, and what would critics of that school say?
• If you started again tomorrow, what would you do differently?
• What are the weakest parts of the thesis, and why do they not sink the argument?
• Who benefits from this work, and what should they do differently because of it?

Prepare three summaries: 30 seconds, three minutes, ten minutes

Do not write essay answers to your question bank. Write bullet prompts, then rehearse the answers out loud, alone or to a friend outside your field. Speaking an answer is a different skill from knowing it, and the first time you say your contribution aloud should not be in the defense.

Almost every defense opens with some version of 'tell us about your thesis'. Prepare three versions of the answer. The 30-second version is one sentence of context, one of method, one of contribution. The three-minute version adds the research questions and the headline findings. The ten-minute version walks the committee through the argument chapter by chapter.

Rehearse each until the structure is automatic but the wording is not. A memorised script sounds brittle and breaks when interrupted, and you will be interrupted. What you want is a skeleton you can re-enter at any point. The short version matters more than it looks: it is the frame the rest of the defense hangs on, and a crisp opening buys goodwill for the harder questions.

Know every citation you would have to defend

Examiners check references. Some will pick a citation and ask what the paper actually found, whether you read past the abstract, or why you cited it for that particular claim. The soft spots are always the same: sources cited from an abstract, citations inherited second-hand from another paper's reference list, and claims that drifted away from what the source says during revision.

Before the defense, walk your bibliography and flag anything you could not discuss for two minutes. Re-read the flagged papers, or cut them in corrections if the citation was decorative. Also check for retractions: a free retraction check takes seconds, and being caught citing a retracted paper in a defense is entirely avoidable.

This is one place tooling genuinely changes the picture. In CiteDash, every citation is a database object that resolves to a real paper, and the Fact Checker verifies AI-assisted claims against the full-text PDFs you hold before they ever reach your draft. If you wrote the thesis that way, you already know every claim-to-source link has been checked, and the Library badges retracted papers and blocks them from citation.

How to handle hard viva questions without getting defensive

Some questions will be hard, and a few may feel hostile. The technique is the same for both. Pause. Take a sip of water. Restate the question in your own words to check you understood it, which also buys thinking time. Then answer the question that was asked, not the one you rehearsed.

Concede gracefully when a point lands. 'That is a fair criticism, and here is why the finding survives it' is a stronger answer than a fortified defense of every sentence. Examiners are testing judgment, and knowing which criticisms matter is judgment.

Keep a boundary phrase ready for scope questions: 'that is an important question, and it sits outside what this design can answer, which is why I flag it in future work'. Said once, it is confident. Said five times, it is evasive, so use it precisely.

Finally, rehearse saying 'I do not know'. Some questions reach beyond the thesis, and inventing an answer under pressure is far more damaging than an honest boundary. 'I have not tested that, but here is how I would' turns a blank into a small demonstration of competence, which is what the whole exam is really for.

Run a mock defense that actually helps

A mock defense is the single highest-value hour of preparation, but only if it is run properly. A friendly chat with your supervisor, who knows the work almost as well as you do, tests very little. A useful mock looks like this:

• Recruit one or two academics who have not lived the project, ideally including someone outside your immediate subfield.
• Give them the thesis, or at minimum the abstract and one chapter, a week ahead.
• Run it full length and in costume: same opening summary, no breaks, questions until time runs out.
• Ask for a fixed quota of unfriendly questions so you practise composure, not just content.
• Record it. Watching yourself hedge is uncomfortable and extremely corrective.
• Debrief for half an hour: which answers rambled, which questions you dodged, what to rehearse again.

The final week: logistics, regulations, and rest

Schedule the mock about two weeks out: close enough that the thesis is fresh, far enough that you can fix what it exposes. In the last week itself, shift from content to logistics. Confirm the format and length, who attends, whether you present slides, and what happens afterwards. Read your institution's corrections categories so that 'minor corrections' lands as the normal, expected outcome it is, not as a verdict.

Check what you may bring. Many systems allow an annotated thesis and notes. Prepare that copy deliberately, with tabs on your research questions, key tables, and the pages your weak-point list points to.

Learn the outcome categories while you are in the regulations. In most systems the common results are a pass with minor or major corrections, and corrections are the norm rather than a rebuke: they are how examiners hand you a better final document. Knowing this in advance changes how the closing minutes of the defense feel.

Long before defense week, the same discipline applies to the document itself: run the pre-submission checks in Submission Readiness so formatting, references, and declarations were verified before examiners ever saw them. There is a full walkthrough in how to check your thesis before you submit. Then, in the final days, stop revising answers and sleep. Composure is a preparation outcome too.

Walk in as the expert in the room

The defense rewards exactly one thing: knowing your own work deeply and being able to talk about it honestly. Re-read like an examiner, name your weaknesses before anyone else can, rehearse your summaries aloud, and verify that every citation in the document can bear weight.

You have spent years becoming the expert on one specific question. The defense is the one room where everyone present knows that. Preparation is how you make sure you act like it.