TOPIC, PURPOSE AND THESIS

Most speeches are lost long before anyone stands up. They are lost at a kitchen table at eleven at night, when a speaker picks something that sounds interesting, promises to “figure out the point later,” and starts collecting sources with no idea what the sources are for. The result is a tour, not a talk — twelve facts about a subject, arranged in the order they were found. This chapter is about the three decisions that prevent that: what you are talking about, what you want the room to do about it, and the one sentence that holds the whole thing together. These are craft decisions, not inspiration. There are formulas, filters and revision moves that work, and you can run them on a topic in twenty minutes. Every idea here comes with a weak version and a repaired version, because the difference between them is the difference between a speech that lands and a speech that merely finishes.

Section 1Choosing a Topic You Can Actually Deliver

Start with an inventory, not a search engine

“Speech topic ideas” typed into a search bar returns a thousand topics that belong to nobody. You will scroll, feel vaguely dissatisfied, and eventually settle on the least offensive item on a list a stranger wrote. That is how people end up giving five minutes on the history of coffee to a room that does not care and to a speaker who cares even less.

The faster route is an inventory of what you already carry. Take a blank page and make three columns. Column one: roles and routines — every job you have held, every shift you work, every place you spend hours (clinical rotation, warehouse floor, church nursery, second-shift call center, a household with a toddler). Column two: arguments you have had — things you have genuinely defended out loud, whether to a coworker, a relative or a comment section. Column three: things you have had to explain — anything a friend, patient, new hire or family member did not understand until you walked them through it.

That third column is gold, and students almost always skip it. Explaining is already a rehearsal for informative speaking. A nursing student’s third column might read: why we ask the same three questions every shift change; what an A1C number actually measures; why a family member cannot just bring in outside medication; how to read a hospital bill. Every one of those is closer to a speech than anything a topic list will hand you, because you have already found the words once and watched them work on a real person.

Here is a worked inventory from a second-year respiratory therapy student, written in about six minutes. Roles: RT student, overnight stocker, youth soccer coach, caregiver for a grandmother with COPD. Arguments: vaping regulation, whether coaches should cut kids, hospital visitor policies. Explanations: how a rescue inhaler differs from a maintenance inhaler, why my grandmother’s oxygen tubing has to be a certain length, how to spot heat illness in an eleven-year-old at practice. That is ten starting points, none of them generic, all of them supportable by someone who has lived near the material.

Brainstorming that produces options, not one idea

The most common brainstorming failure is stopping at the first decent idea. Generation and judgment are different mental activities, and running them at the same time kills both. Set a timer for eight minutes and write candidates without evaluating a single one. Bad ideas are useful; they are the stepping stones to the adjacent good idea. Aim for twelve candidates minimum before you allow yourself an opinion about any of them.

Two techniques do most of the work. Clustering (also called mind mapping) puts a broad word in the center of a page and branches outward: from sleep you might branch to shift work, teenagers, sleep apnea, caffeine, insomnia apps, drowsy driving, hospital noise at night; from shift work you branch again to rotating schedules, nap policy, error rates at 4 a.m., the drive home. The second or third ring is where speech-sized ideas live. Laddering works differently: take a candidate and ask “so what?” three times. Drowsy driving → so what? → crashes peak in early morning hours → so what? → students who work overnights and drive to an 8 a.m. class are running a risk they have not named → that is a speech, and the audience is sitting in the room.

One caution about generation: do not let novelty become the only criterion. Instructors are not scoring originality of subject, they are scoring whether the room learned or moved. A common topic handled with real specificity and fresh evidence beats an exotic topic handled thinly. “Hand hygiene” sounds tired until you build it around the twelve-second observation you made on your own unit.

The three-question filter

Once you have candidates, evaluate them against three questions, in this order. First, investment: will I still be interested at hour four? You are going to spend real time with this material, and boredom shows up in the voice. Second, audience need: does this specific audience have a reason to listen, and can I state that reason in one sentence? Chapter 5 gave you the tools for that judgment; here you simply apply them. Third, supportability: can I find credible evidence in the time I actually have? A brilliant topic with no accessible sources is not a topic, it is a wish.

Then run the constraint check, which is unglamorous and saves people. Does the topic match the assignment type — an informative assignment is not the place for your best persuasive material? Does it fit the time limit? Is it appropriate for this room, this institution, this hour? And can you handle it ethically: a topic drawn from your workplace must never carry identifiable patient or client details, and a topic drawn from your own trauma should be one you can deliver without being ambushed by it mid-speech.

Notice that the total is not what decides. A 1 or a 2 in any single column is disqualifying no matter how the other two score — which is why two topics here tie at ten points and both still get cut. A topic you cannot support becomes a speech of unsupported assertions; a topic the audience has no stake in becomes a well-researched lecture delivered to people checking their phones.

Section 2General Purpose and Specific Purpose

The four general purposes

The general purpose is the broad category of what you are trying to do to an audience. Textbooks disagree on the count. The most common list is three — to inform, to persuade, to entertain — and many add a fourth, to commemorate, for tributes, toasts and eulogies. Some collapse the last two into a single “special occasion” category, and a few argue that all speaking is ultimately persuasive because even a choice of which facts to present is an argument about what matters. Those disagreements are real and worth knowing, but for building a speech the four-way split is the most useful tool, because each purpose carries a different test for success.

That test matters more than the label. If you cannot say what would have to be observably true afterward for your speech to have worked, you do not have a purpose — you have a subject.

The line between informing and persuading is a continuum, not a wall, and that is where most students get marked down. A speech on vaping that closes with “so think carefully about what you put in your lungs” has slid into persuasion. A persuasive speech that spends four of five minutes explaining what a topic is has slid into informing. The fix is not to police every sentence; it is to fix the general purpose first and then let it govern verb choice, evidence type and the shape of the conclusion.

Writing a specific purpose statement

The specific purpose statement is a single sentence, written for you rather than spoken to the audience, that states exactly what you want this audience to know, believe or do by the end. The standard formula is reliable enough that you should just use it: To + general-purpose verb + my audience + the precise outcome.

Six rules make it work. One infinitive only — “to inform and persuade” is two speeches. Name the actual audience, not “people.” State one idea; if you need “and,” check whether you smuggled in a second speech. Write a statement, never a question. Use plain language, not the poetic phrasing you might want in the speech itself. And make it achievable inside the time limit — if the outcome requires twenty minutes, the statement is wrong, not the clock.

Choosing verbs you can check

Most limp purpose statements fail on one word. Verbs like understand, know, appreciate, be aware of and learn about describe invisible states. Nobody, including you, can tell whether they happened. Swap them for verbs that name something an audience member could actually do afterward: describe, compare, identify, sequence, explain, demonstrate for informative speeches; adopt, switch, register, vote, donate, schedule, refuse, reconsider for persuasive ones.

The rewrite is small and the effect is large. “To inform my audience about stroke” gives you nothing to build with. “To inform my classmates how to identify the five warning signs in the BE FAST stroke check and what to do in the first ten minutes” has already told you your main points, your evidence type and your conclusion. A specific purpose statement is not paperwork; it is the outline in compressed form. When students say they are stuck on organization, the problem is usually upstream in a purpose statement that never committed to anything.

Keep the statement on the page in front of you while you draft, and treat it as a gate: any fact, story or slide that does not serve that sentence comes out. This is the cheapest editing rule in public speaking, and it is the one that turns a seven-minute draft into a clean five.

Section 3Narrowing: Making the Topic Fit the Clock

The scope ladder

Scope is how much territory a topic claims. Narrowing is the deliberate reduction of that territory until what remains can be developed, not merely mentioned. The most reliable method is a ladder: write your subject at the top and take one concrete step down at a time, until a rung appears that you could genuinely finish inside the time limit. Do not jump from the top to the bottom in one move; the middle rungs are where you discover the interesting angle.

Rungs one and two are not speeches; they are course catalogs. Rung three still feels like a speech to most students and is still too big — “how resistant organisms spread” is a whole chapter, and you have five minutes. Rung five is where an audience gets something they can hold, because you have room to explain each of the three items, show evidence for each, and still have a beginning and an end.

Time math, done honestly

Speaking rates vary a great deal by person and by material, but most public speakers land somewhere between roughly 120 and 160 words per minute, and planning at about 130 is safe. A five-minute speech is therefore something like 650 words — roughly two and a half double-spaced pages. That number tends to shock people, and it should, because it is the whole argument for narrowing.

Now subtract. An introduction and conclusion together typically eat twenty to twenty-five percent of the clock, so a five-minute speech leaves under four minutes of body — call it 225 seconds. Split that across main points. Two main points get about 110 seconds each; three get about 75. In seventy-five seconds you can state one claim, give one piece of evidence with its source, explain what the evidence means, and transition. That is it. That is the actual unit of a speech, and it is why a four-point speech in five minutes is almost always a list of headlines.

Run the arithmetic before you draft, not after. A student who plans “the causes, effects, treatment and prevention of type 2 diabetes” in five minutes has budgeted about fifty-five seconds per point, which is not enough time to say anything a listener could not have guessed. The same student, narrowed to “three habits that change an A1C reading in ninety days,” has the same clock and suddenly has room for a real example inside each point.

Depth beats coverage: a worked comparison

Take hand hygiene, five minutes, informative. Version one, the coverage draft: history of Semmelweis, the germ theory, types of pathogens, soap versus alcohol-based rub, the five moments for hand hygiene, glove use, compliance rates, hospital-acquired infection statistics, and a closing plea. Eight ideas in the body, under thirty seconds apiece. The audience will retain approximately one of them, chosen at random.

Version two, the depth draft: one claim — alcohol-based rub outperforms soap and water for most routine moments, but only if it stays wet on the hands for the full contact time. Point one: what the contact time is and why the manufacturer sets it. Point two: the two moments in a shift when people cut it short, drawn from your own observation on a unit. Point three: a twenty-second demonstration the audience can imitate at the next dispenser they pass. Same clock, one idea, three developed points, a physical demonstration, and a listener who can now do something differently tonight.

The coverage draft feels safer because it looks thorough. It is not thorough; it is thin in eight places instead of deep in one. The test to apply to your own draft is blunt: if a listener could only carry one sentence out of the room, which sentence is it, and did you spend your time earning that sentence?

Section 4The Thesis: One Sentence That Runs the Speech

What a thesis is, and the three things it gets confused with

The thesis statement — many speech textbooks call it the central idea — is one declarative sentence that states the single main idea of the speech in the words the audience will actually hear. It is not the topic, not the title, and not the specific purpose. Those four things describe the same speech from four angles, and keeping them straight ends a lot of confusion.

Look at all four for one speech. Topic: sleep and night-shift work. Title: “The Drive Home.” Specific purpose (for the speaker): to inform my classmates how rotating night shifts disrupt the body clock and what an anchor-sleep block does about it. Thesis (spoken to the audience): “Rotating shifts break your body clock in three predictable ways — and a fixed four-hour anchor of sleep repairs most of the damage.” The purpose statement is the plan; the thesis is the promise.

Three quick tests. A thesis is a statement, never a question — “Why is nurse burnout rising?” is a title, not a thesis. It is one sentence; if yours needs a semicolon plus an “and also,” you have two speeches. And it must assert, meaning a reasonable listener could summarize what you claimed rather than merely what you discussed.

Informative central ideas still make a claim

Students often think an informative thesis should be neutral to the point of blandness. It should be neutral about advocacy — you are not asking anyone to act — but it should still commit to an assertion about how something works, what it consists of, or why it happens. “Today I will talk about the immune system” commits to nothing. “Your immune system fights an infection in three overlapping waves, and each wave explains a symptom you have felt” commits to a structure, a claim and a payoff.

A useful pattern for informative theses is quantified structure plus significance: name how many parts, name the parts’ nature, and say why it matters to this room. “Three forces set the price on a hospital bill, and none of them is what the treatment cost.” The number does two jobs at once — it makes the claim concrete and it hands the listener a mental filing system before you start filling it.

Persuasive theses and the three claim types

A persuasive thesis must be a sentence a reasonable person could deny. If nobody in the room could disagree, you have written a topic. Persuasive theses come in three families, and knowing which one you have tells you what you owe the audience.

A claim of fact argues that something is or is not true, did or did not happen, will or will not occur: “Twelve-hour shifts increase medication errors in the final two hours.” Your burden is evidence — data, studies, expert testimony. A claim of value argues that something is good, bad, just, unjust, better or worse: “Mandatory overtime in nursing is unethical.” Your burden is a stated standard; you cannot call something unethical without telling the audience by which criterion, and most value speeches fail because the speaker assumed the standard was obvious. A claim of policy argues that something should be done: “This campus should extend shuttle service until 2 a.m. for students leaving night clinicals.” Your burden is the full package — there is a problem, this plan addresses it, the plan is workable, and its benefits outweigh its costs.

Thesis plus preview, and where they land

In delivery, the thesis usually appears at the end of the introduction, immediately followed by a preview statement that forecasts the main points in the order the audience will hear them. Chapter 9 handles introductions in full; what matters here is that the thesis and preview are written together, because the preview is a test of whether the thesis is really one idea.

A model pairing, spoken aloud: “Rotating shifts break your body clock in three predictable ways — and a fixed four-hour anchor of sleep repairs most of the damage. We will look at what happens to your core temperature rhythm, what happens to your reaction time on the drive home, and what an anchor block actually looks like on a real schedule.” Two sentences, about thirty-five words, and the audience now has a map.

Say both sentences out loud before you accept them. Written theses tend to grow subordinate clauses that no listener can follow. If you run out of breath, or if you have to re-read to keep the grammar straight, cut it down. A thesis is written for the ear, and the ear forgives almost nothing.

Section 5Alignment: Testing Topic, Purpose and Thesis Together

The five-check alignment audit

Alignment means that the topic, the general purpose, the specific purpose and the thesis all describe the same speech. Misalignment is the single most common structural problem in student speaking, and it is almost always invisible to the speaker, because each piece looks fine on its own. Run these five checks in order, with all four elements written on one page in front of you.

Check four deserves special attention because it is the one that fails after you have already done the work. You wrote a thesis, then spent three hours in the library, and the evidence turned out to say something adjacent to your claim but not the same thing. The honest move is to revise the thesis. The dishonest move — and it is dishonest, not merely sloppy — is to keep the thesis and stretch the sources to fit it.

Failure patterns you can spot by their symptoms

Topic creep shows up as a speech that keeps adding related material. You started with anchor sleep and now you have a section on melatonin supplements, because it was interesting and it was right there. The symptom is that you cannot state your thesis without the word “also.” The fix is a decision, not a compromise: melatonin is either inside the thesis or it is out of the speech.

Thesis drift is topic creep’s quieter cousin. The speech slowly becomes about the most interesting thing you found rather than the thing you set out to say. Drift is not a sin — research is supposed to change your mind — but it must be resolved. Rewrite the thesis to match what you now believe, then re-check the purpose statement and the outline against the new sentence, in that order.

Purpose-audience mismatch sounds like a good speech aimed at the wrong room: a policy speech urging classmates to lobby a state agency none of them can influence, or an informative speech explaining basics the audience already lives with. Two speeches in one announces itself in the preview: “I will explain what sepsis is, and then argue that our screening protocol should change.” That is a fine ten minutes and an impossible five. And the buried thesis is the speech where the real claim appears at minute four, after the audience has already decided the talk was a list.

A worked case, start to finish

A student writes down one line from her inventory: “I work night shift at a rehab facility.” That is a role, not a topic. Laddering: so what? → the drive home after a twelve is the scariest part of the job → so what? → half our class works overnights and drives to campus. Now there is a topic with an audience attached.

First attempt: Topic — drowsy driving. Specific purpose — to inform my audience about the dangers of drowsy driving. Thesis — “Drowsy driving is dangerous.” Run the audit. Check one passes. Check three fails hard: nobody disputes that drowsy driving is dangerous, so the room has no reason to lean in. Check five fails too, since a thesis this thin can preview anything.

Second attempt, after two hours of reading: Specific purpose — to inform my classmates who work overnight shifts how sleep debt impairs driving and what three countermeasures actually work. Thesis — “After a night shift, your driving is impaired in ways you cannot feel, and only three of the countermeasures drivers commonly use do anything at all.” Preview — what impairment looks like on the drive home, why the usual fixes fail, and the three that hold up. That thesis asserts something surprising, is scaled to three main points in five minutes, aims squarely at the people in the room, and previews cleanly.

Then the research shifts again: the strongest evidence she finds is about a short pre-drive nap, and the material on the other two countermeasures is thin. Rather than padding, she narrows once more. Final thesis — “The only countermeasure that reliably gets a night-shift worker home safely is the one that feels like a waste of twenty minutes: sleeping before you drive.” Same topic, tighter claim, evidence and thesis now saying the same thing. That is what alignment work looks like — not a form to fill out once, but three or four small corrections that keep the speech honest.

Words worth knowing