A thesis paper is two things at once: a document and a learning process. As a document, it is—depending on the examination format—the script for your oral examination or the argumentative roadmap for your term paper. As a learning process, it is both monological, because you alone must undertake the intellectual work of turning a topic into a problem, and interactive, because the paper is refined over several rounds with your instructor. You thus assume responsibility for the dramaturgy of your own examination: you determine which competencies specified in the study regulations (critical analysis, conceptual work, argumentative ability) are to become evident in relation to which subject matter. The following guide explains, step by step, what a thesis paper must accomplish and how the working process proceeds.

Step 1: Problematize Rather Than Merely Introduce a Topic

The most common mistake at the outset is that you identify a thematic field (“Social Media and Friendship,” “Justice in Sport”) but do not articulate a problem. A question becomes philosophical only when it formulates an intellectual conflict, a tension, or an aporia—something that cannot be resolved without argumentation. Test your question using four negative criteria:

1. No yes-or-no questions. “Is X compatible with Y?” invites a one-word answer. Reformulate it: “In what respect…?”, “Under what conditions…?”—questions of this kind compel differentiation.
2. No either-or questions. “A or B?” divides a complex field into a false alternative. Instead, ask about the conditions, criteria, or limits under which something holds.
3. No empirical questions. “Does X influence Y?” is a question for sociology or psychology. The question becomes philosophical only when you inquire into the conceptual, normative, or ontological status of this relation.
4. No trivial questions. If part of your question is uncontroversial, it is philosophically redundant—remove it and focus the question on what can reasonably be disputed.

Step 2: Derive Rather Than Posit

Your question must not “fall from the sky.” The introduction to your thesis paper must show why one is confronted with precisely this problem—as the necessary consequence of an intellectual conflict, not merely as an expression of interest or a historical episode. Ask yourself: From what field of tension does my question necessarily arise? Anyone who reads your derivation should feel the problem themselves before you state it.

Step 3: The Placeholder Test

Look for vague relational terms in your guiding question: “appropriate,” “bound,” “relevant,” “problematic.” Such words obscure the criterion that is actually at issue. Replace the placeholder with the criterion—it is often already present in one of your theses and merely needs to be incorporated into the question. But bear in mind that every clarification comes at a cost. Anyone who sharpens a question makes an interpretive decision and introduces presuppositions. Make this decision explicit in the paper rather than making it tacitly.

Step 4: Venture a Thesis

A thesis is a risky position, not a finding. “It is more complex than expected” is not a thesis. “Various positions can show that…” announces a literature review, not an argument. A viable thesis is:

• defensible and open to challenge—one can argue for and against it by giving reasons;
• demarcating—it recognizably excludes alternative answers;
• argument-selecting—it determines which arguments and sources are relevant;
• stated at the outset—it appears at the beginning and guides the text rather than being asserted retrospectively in the conclusion;
• supported by the text—if you attribute a position to an author, your thesis must not overstate that position. Ask yourself: Does the text really say this, or does it make only a weaker claim?

The most productive form combines both: first demonstrate that you have understood a position (reconstruction), then add an open follow-up question or criticism of your own (problematization).

Step 5: Concepts as Supporting Infrastructure

Explicitly identify the central concepts of your work and decide for each one whether it will be defined (because it serves merely as a tool) or problematized (because it is itself being examined). The standard of quality is the capacity to draw distinctions: Can you distinguish between “good,” “right,” and “just”? Between “equal treatment” and “identical treatment”? Avoid two typical errors: suggestive, unanalyzed formulas (forceful words that merely assert meaning) and quotations used solely as appeals to authority. A quotation is a site of reflection: it is conceptually analyzed and subjected to argumentative scrutiny, not presented as proof.

Step 6: Structure Functionally Rather Than Enumerating Additively

An outline resembling the table of contents of an introductory textbook (“Foundations—Author A—Author B—Comparison—Conclusion”) has failed. Every chapter requires an explicit argumentative function in relation to your thesis. The diagnostic question for every transition is: “Which problem from step N compels me to proceed to step N+1?” If you cannot answer this question for a transition, the sequence is arbitrary—and therefore requires justification. Proven basic dramaturgical patterns include:

1. Problem → Proposed solution A → Failure of A → Proposed solution B → Synthesis;
2. Theory → Objection → Repair → Renewed criticism → Your own position.

Formulate headings that make substantive claims rather than merely indicating formal sections (“Why the Coherence Argument Succeeds” rather than “Conclusion”), and include a brief signpost in every chapter telling the reader where she stands in the argument.

Step 7: Choreograph the Literature Rather Than Merely Reporting It

The guiding distinction in philosophical scholarship is: doxography vs. validity assessment—reproducing what has been said as opposed to examining whether it is true. A linguistic symptom of report mode is: “As X explains…,” “Y shows that…”—anyone who writes in this way moderates the debate rather than participating in it. Avoid both extremes: the single-source vacuum (the entire paper depends on one author) as well as the omission of research literature (primary texts only, with no positioning within the field of existing interpretations). Choreograph your literature: assign each source a distinct methodological function (source of the problem, opposing position, test case, proposed repair) rather than reporting on them in parallel. And first make the opposing position as strong as possible before criticizing it—a response to a weak version will convince no one.

Step 8: Feasibility and Focus

Scale your project consistently to the examination format: How many pages do the examination regulations stipulate? How many minutes does the oral examination last? (Research this yourself—it is part of the competence required for the examination.) Two typical errors are:

1. Multiple projects in one. If your draft contains three possible papers, one must become the central project; the others become test cases or avenues for further inquiry.
2. Excessive granularity. Those who have successfully narrowed their question often increase the level of detail again within the remaining topic (thirty subsections over fifteen pages). Countermeasures: treat subsections as internal argumentative steps rather than subchapters; assign each chapter a page budget; identify and remove redundancies between chapters.

The Working Process: Ping and Pong

A thesis paper emerges through two or three rounds of revision with your instructor. Three rules apply to this process:

1. Guarantee of continuity. Feedback does not replace your line of argument; it sharpens it. You do not need to start over after receiving criticism—you refine the existing question and thesis. Endless restarts are the opposite of progress.
2. The necessity of choosing between alternatives. Feedback often presents two or three well-developed options for narrowing the focus, together with their respective advantages and disadvantages. The choice is yours—because you must take responsibility for the position in the examination. Authorship cannot be delegated.
3. Saturation criterion. The process has an endpoint. A thesis paper that already answers every question has been overworked: it becomes a short treatise that settles too many answers and leaves no room for the examination discussion. At this stage, open points are not deficiencies but productive starting points for follow-up questions—the paper should provide points of friction for the discussion, not completeness.

Using AI—as a Sparring Partner, Not an Author

You may use AI during the exploratory process, but the roles must remain clear: AI is an examiner, critic, and sparring partner—not a generator of ideas from nothing and not an author. The responsibility clause is: Do not adopt any suggestion that you have not fully understood and cannot defend yourself in the examination. Productive forms of use include:

• Request counterquestions and objections to your draft, not summaries.
• Conduct a Socratic dialogue governed by strict rules: “Do not accept mere repetitions of my thesis paper.”
• Use perspective-switching prompts: Have the opposing position formulated as strongly as possible (steelman), and then develop the response yourself.
• Simulate the examination with a graduated progression of questions—from questions of comprehension to questions of validity.

Enter the AI session with your own prepared proposals rather than allowing yourself to be carried along by improvisation. And transparency applies to both sides: feedback from instructors that was produced with AI support will also be identified as such.

Form and Format as Part of the Substance

Questions of format are not superficial matters but epistemic conditions of scholarly communication. A thesis paper must be citable, commentable, and versionable: submit a PDF with page numbers. Photographs of texts (JPEG) prevent annotation, version comparison, and text searches; continuous formats without a page structure make it impossible to refer precisely to individual passages; freely editable files undermine a document’s status as an authoritative version. The format is part of your text’s claim to validity.

How Your Work Will Ultimately Be Assessed

The criteria governing the exploratory process and those used in the final assessment are the same. Evaluations of term papers examine precisely what the feedback on the thesis paper is designed to develop—according to a fixed framework:

1. derivation of the question,
2. development of the thesis,
3. methodological reflection on the structure,
4. line of reasoning and engagement with the literature,
5. use of quotations and conceptual analysis,
6. linguistic and stylistic form.

This predictability is the purpose of the shared framework for discussion: there are no hidden standards. Anyone who takes the process seriously will know on the day of the examination the criteria by which their performance will be assessed—because they will already have put them into practice in the thesis paper. Further guidance on formulating questions in written work is available on the Writing Center’s page on research questions.

Using This Page as a Tool for Your Use of AI

The present text is not intended merely to be read. You can paste it directly into an AI session and use it for two purposes:

1. As instructions for the AI to guide you purposefully. Add this page to the AI’s context before you begin working on your thesis paper. This allows the AI to assess your work according to the same criteria applied by your instructor rather than according to general notions of a “good paper.” At the beginning of the page, there is a function that displays it as markup text. This is an AI-friendly format. Opening prompt: “Here is my instructor’s guide to the thesis paper [insert text]. From this point onward, act as a sparring partner according to precisely these criteria. Begin by asking me questions about steps 1 and 2—problematization and derivation—before we proceed.”
2. As a simulation for anticipating your instructor’s response. Before submitting your thesis paper, ask the AI to assume the role of your instructor and draft feedback based on this page. This allows you to identify potential points of criticism before the discussion takes place. Opening prompt: “Here is my instructor’s guide to the thesis paper [insert text] and my current draft [insert draft]. Give me the feedback that my instructor would likely provide according to these criteria—including strengths, weaknesses, and concrete revision steps.”

In both cases, the responsibility clause continues to apply: AI provides questions and assessments, not ready-made answers. Whatever you ultimately submit must be something you have understood fully and can defend yourself.