Yes — and it is the single word choice in your thesis that carries the most meaning per character. A reporting verb is not a neutral connector. It is a compressed claim about how much you believe the source, how strong its evidence is, and whether you intend to build on it or push against it. A supervisory committee reads that claim whether or not you meant to make one.
This page is a working reference: what each verb commits you to, which ones are safe when your evidence is thin, and the specific substitutions that quietly change your argument. It is written for a Canadian thesis, major research paper or dissertation, and it assumes you already have sources to introduce.
What exactly is a reporting verb?
It is the verb that carries a source into your sentence: Chen (2021) argues that…, Chen (2021) found that…, Chen (2021) claims that…. Every citation you write that is not a bare parenthetical has one, which means a literature review of eighty sources contains roughly eighty small statements of your own position that you may never have consciously made.
The mechanics of the citation itself — author, date, placement, punctuation — are a separate problem, handled in the guide to citing in APA 7th edition for a Canadian thesis. This page is about the word immediately after the bracket.
What are the four things a reporting verb can encode?
Every choice is doing up to four jobs at once, and most weak sentences come from getting one of them wrong while the other three are fine.
- Evidential strength. Does the source have data, or a position? Demonstrates asserts data strong enough to settle something. Argues asserts reasoning.
- Your alignment. Do you accept it? Shows accepts. Claims does not.
- The source type. A theoretical essay does not find anything. An experiment does not contend.
- What you will do next. A verb of agreement sets up an extension. A verb of distance sets up a critique, and an examiner will expect the critique to arrive.
Which verbs say that you accept the finding?
These verbs put your credibility behind the source. Use them when you are willing to be asked, at the defence, to justify the source as well as your own work.
| Verb | What it commits you to | Safe when |
|---|---|---|
| demonstrates | The evidence settles the point. | Strong design, adequate sample, replicated. |
| establishes | The field now treats this as known. | There is a review or a body of work behind it. |
| shows | The data support the reading, and you agree. | The analysis is sound and directly on point. |
| found | This is what the study produced, and you accept the report. | Almost always safe for empirical work. |
| confirms | It agrees with something already established. | You have named the earlier finding first. |
| documents | It records a state of affairs. | Descriptive or observational work. |
Found is the workhorse and the safest default for empirical sources. It reports without inflating. If you are unsure which verb to use in a Canadian thesis literature review, found is almost never the wrong answer, and the synthesis moves that make a chapter out of those reports are set out in the step-by-step guide to writing a literature review for a thesis.
Which verbs put distance between you and the source?
These signal reservation. They are legitimate and often necessary, but they are never neutral, and using one by accident is one of the more common unforced errors in a graduate chapter.
| Verb | What a reader hears | Use when |
|---|---|---|
| claims | The author says so; the writer is not persuaded. | You are about to explain why not. |
| asserts | Stated without adequate support. | The source genuinely offers no support. |
| alleges | Contested, possibly improper. | Almost never in a thesis outside law. |
| purports to show | Openly sceptical. | You have a specific methodological objection. |
| maintains | Holds a position under pressure from others. | There is a live disagreement in the field. |
The most frequent single error is using claims as though it meant says. A candidate who writes Nadeau (2019) claims that participation rates have declined and then builds three chapters on Nadeau has told the committee that the foundation of the thesis is a claim the candidate does not accept. That is a question at the defence, and it is a question with no good answer. The general shape of that questioning is covered in what questions are asked at a Canadian thesis defence.
How strong a verb does your evidence actually license?
Match the verb to the design, not to how much you liked the paper. A rough ladder, strongest at the top:
- demonstrates / establishes — meta-analysis, replicated experiment, or a systematic review.
- shows / found — a single well-powered study reporting its own results.
- reports / describes / documents — descriptive, observational or survey work.
- suggests / indicates / points to — a small sample, a correlational design, a pilot, or a qualitative study with a narrow participant group.
- proposes / theorises / conceptualises — no data at all; a framework or a position paper.
Two rules follow directly. First, a correlational study never demonstrates a cause; the wider problem of causal language in interpretation is worked through in the guide to writing the discussion chapter. Second, if your own sample is small, the verb you use for your own findings should sit at level 4 — and the reasoning behind whether it is small is in how many participants a thesis needs.
Does the convention change by discipline?
Yes, and it changes enough that a verb which reads as normal in one Canadian department reads as naive in another.
- Humanities and interpretive social science. Texts and their authors are treated as present and available for argument: Fanon argues, the poem enacts, Taylor reads Hegel as. Argument verbs dominate because argument is the object of study.
- Experimental and health sciences. Studies are events that happened: the trial found, participants reported, the authors observed. Argument verbs are used sparingly and mostly in the discussion.
- Engineering and computer science. Systems and methods are the subject: the model achieves, the approach outperforms, the authors propose.
- Law. A distinct set, and a strict one: a court held, a judge found, a statute provides, an author submits. Getting this wrong is immediately visible to a legal reader, and the source conventions are set out in the guide to doing legal research for a Canadian law thesis.
The reliable way to calibrate is not a list. Open three recently accepted theses from your own programme in your university repository, take the first thirty citations in each literature review, and write down the verbs. The distribution you get is your department’s convention, and it will be narrower than any published list.
Which verbs weaken a sentence without adding safety?
A separate category: verbs that are not wrong but are doing nothing.
- talks about, looks at, deals with, touches on. Conversational, and they tell the reader nothing about what the source concluded.
- mentions. If it is only a mention, ask why it is in your chapter.
- states. Technically neutral, but it flattens everything to the same level; a chapter built on states reads as a list.
- tries to show. Reads as a sneer without an argument attached.
The test: delete the verb and read the sentence as a fragment. If the source’s actual conclusion is still missing, the verb was not carrying its weight.
Does verb choice have anything to do with plagiarism?
Indirectly, and usefully. Patchwriting — paraphrasing by swapping synonyms while keeping the original sentence shape — is one of the most common integrity problems in graduate work, and it is very hard to do while consciously choosing a reporting verb, because choosing the verb forces you to decide what the source concluded and how much you believe it. That decision is yours; the original sentence structure is not.
It also changes what your similarity report looks like, though not in the way most candidates expect. Correctly attributed reporting still matches the source in the parts that should match. What percentage means what is covered in what similarity percentage is acceptable at a Canadian university.
A verb bank to work from
Grouped by the function you need, not alphabetically. Pick the function first, then the verb.
| Function | Verbs |
|---|---|
| Reporting a result | found, reported, observed, recorded, measured, identified |
| Reporting a strong result | demonstrated, established, showed, confirmed, verified |
| Reporting a tentative result | suggested, indicated, implied, pointed to, hinted at |
| Presenting a position | argues, contends, maintains, posits, holds, submits |
| Presenting a new idea | proposes, introduces, develops, advances, outlines, theorises |
| Agreeing across sources | corroborates, supports, echoes, extends, replicates |
| Disagreeing across sources | disputes, challenges, questions, complicates, refutes, contradicts |
| Signalling your reservation | claims, asserts, purports to show, assumes |
| Describing what a study did | examined, investigated, surveyed, compared, tracked, evaluated |
| Describing a gap | overlooks, omits, does not address, stops short of, leaves open |
The last row is worth practising. Leaves open and stops short of identify a gap without accusing anyone of incompetence, which is the register a Canadian committee expects when the person whose gap you are naming may be sitting on your examining committee or reviewing your grant.
How do you fix a chapter that is already written?
Do it as a pass, not while drafting. Search your document for the author names in your reference list, look only at the word after each one, and ask a single question each time: am I willing to defend this source at this strength? If yes, move on. If no, move down the ladder. Most candidates find between five and fifteen verbs that overstate, and one or two that accidentally sneer.
Second pass: count how many times you used your three most frequent verbs. If states, says or found accounts for more than half your citations, your chapter is reporting rather than synthesising, and no amount of verb substitution will fix that on its own — the synthesis has to be added.
Getting the draft down so you can do this pass at all
None of this is possible on a chapter that does not exist yet. Tesify gets a structured draft on the page — your sources, your argument, your sections, with citations formatted consistently — so that the work you actually get marked on is the judgement layer: which claim you are willing to stand behind, and how strongly. The thinking stays yours. The blank page stops being the obstacle.
Draft your literature review with Tesify
Frequently asked questions
What is a reporting verb?
It is the verb that introduces a source into your sentence, such as argues, found, suggests or claims. It encodes how strong the source evidence is and whether you accept it, so it functions as a compressed statement of your own position on every source you cite.
Which reporting verb is the safest default?
Found, for empirical sources. It reports what a study produced without inflating the strength of the evidence and without signalling scepticism. For theoretical or argumentative sources, argues is the equivalent safe default.
What is wrong with using claims for every source?
Claims signals that the author asserted something you do not accept. Applied to a source you are building on, it tells your committee that your own foundation is unreliable, and it invites a defence question you cannot answer without contradicting your chapter.
Can I use demonstrates for a correlational study?
No. Demonstrates asserts that the evidence settles the point, which a correlational design cannot do for a causal question. Use shows or found for the association itself, and suggests or indicates for any causal reading of it.
Do reporting verbs differ between disciplines in Canada?
Yes, substantially. Humanities and interpretive social science favour argument verbs such as argues and contends. Experimental and health sciences favour event verbs such as found, observed and reported. Law uses a distinct and strict set in which a court held and an author submits.
How do I find out what my own department expects?
Open three recently accepted theses from your programme in your university repository, take the first thirty citations in each literature review, and record the verbs. That distribution is your department convention, and it will be narrower than any published list.
Should the verb change between the literature review and the discussion?
The function changes, so the verb usually does too. A literature review mostly reports what sources found. A discussion mostly positions those findings against your own, which calls for verbs of agreement and disagreement such as corroborates, extends, complicates and contradicts.
Is there a rule about how many different reporting verbs to use?
There is no fixed number, but if your three most common verbs cover more than half your citations, the chapter is probably reporting rather than synthesising. That is a structural problem, and substituting verbs will not fix it on its own.
Does a reporting verb protect me from a plagiarism finding?
Not on its own, but consciously choosing one makes patchwriting much harder, because it forces you to decide what the source concluded rather than reshaping its sentence. Attribution still has to follow the citation rules your programme applies.
Are verbs like talks about and looks at acceptable?
They are not errors, but they are conversational and they tell the reader nothing about what the source concluded. In a thesis they usually mark a sentence that has not yet decided what it is saying.
How do I name a gap without insulting the author?
Use verbs that describe the boundary of the work rather than a failing of the researcher: leaves open, stops short of, does not address. That register matters in a small national field where the author may examine your thesis or review your grant application.
Does this apply to a major research paper as well as a thesis?
Yes. The scale is smaller and the source count is lower, but a major research paper is assessed on the same synthesis skill, and a reader forms the same impression from the same verbs.