Another person’s testimony: grounds for trusting what others know

Epigraph

A wise man, therefore, proportions his belief to the evidence.

Hume, Enquiry Concerning Human Understanding, X.4, SBN110–111; Hume Texts Online. Sentence within Of Miracles.

What this article is about

Much of what we know comes through other people's words: what happened while we were absent, how unfamiliar work is organised, or where a document can be found. Checking everything personally is impossible, but accepting every report provides no adequate basis either. Between these extremes lies the question of testimony: why can someone's words support our knowledge, and when should that support be clarified? Here, trust concerns a particular claim rather than a person's overall worth. We distinguish sincerity, access to an event, understanding of the subject, and the route of transmission. Limited passages from Hume, Reid, and Proverbs offer different positions for comparison. Philosophical argument and religious advice are not psychological experiments. The proposed exercise leaves readers free to choose the depth of inquiry. It allows a specific question about a report without requiring suspicion towards everyone or promising infallibility. Sometimes ordinary assistance suffices; sometimes another source is needed; sometimes uncertainty remains.

I. Other people's words enter our knowledge

Imagine an acquaintance tells us a meeting has moved to another time. This is a hypothetical situation. We were absent when the decision was made, yet the report allows us to change our plans. Understanding comes through another person who transmits information we cannot directly access. This need not be an inferior substitute for knowledge acquired alone. The acquaintance may have participated in the decision and received an accurate notification. The relevant question concerns the connection between the report and event. When that connection is clear enough for the task, the report provides a workable basis. When it is unclear, identify the missing support without turning its absence into an accusation of deception.

Demanding personal verification of everything soon encounters its own conditions. Locating the meeting may involve a map, signs, and someone's directions. Checking a document can depend on information about its preparation supplied by someone else. No statistical claim about the prevalence of second-hand knowledge is needed here; notice the structure of the task. Complete independence from transmitted information is not a necessary condition for a reasonable decision. Independence can instead mean understanding whom we trust in a particular respect, which limits remain, and what would warrant clarification. Someone else's help does not remove our responsibility for the next action we choose on the basis of it.

Distinguish trusting a report from handing over the entire capacity to judge. We can accept information about the meeting time without sharing the acquaintance's opinion of its value. They may know the schedule reliably while speculating about the organiser's intentions. Making that distinction does not demean them. It recognises that different claims have different grounds. An ordinary conversation need not examine every ground in detail. When doubt arises, return to the specific statement: what was reported, and which action depends on it? This gives the inquiry an object and prevents one difficulty from becoming a verdict of general unreliability. It also leaves the helpful person room to know one thing well without claiming to know everything surrounding it.

II. Sincerity and accuracy are different

Someone can say what they believe and still be mistaken. This is a logical possibility, not a diagnosis of a particular witness. They may mishear the hour, treat a proposal as a final decision, or miss an updated notification. If the report proves inaccurate, first reconstruct how it arose. Deliberate deception is a separate hypothesis requiring its own basis. Do not choose it automatically to explain every discrepancy. Correcting the meeting time may require only identifying which notification the person saw and when. Their motives may be irrelevant to finishing this limited task, even though motives could matter in a different inquiry.

The distinction works in the other direction too. A pleasant manner does not certify content, and an unpleasant manner does not refute it. Someone we dislike can provide accurate information; someone we love may need to clarify theirs. This does not require abandoning relationships for cold examination. It avoids making a relationship perform work it cannot do. Care, agreement, competence, and a report's accuracy are distinguishable, even when circumstances connect them. If a friend passes on an outdated schedule, checking it does not require declaring the friendship mistaken. If a stranger offers the correct link, that alone does not make them a reliable adviser about every other subject we might encounter.

Questions can concern the process rather than the personality: did you see the latest notification, was that an agreed decision or a suggestion, and have I understood the date correctly? These are the authors' proposed formulations, not a tested means of changing another person's behaviour. The person may answer, decline, or lack the information. A specific gap then remains. Keeping it visible is more useful than filling it with a confident story about character. Accuracy concerns the connection between words and their subject. Sincerity concerns the speaker's relationship with their words. Many decisions involve both questions, but an answer to one cannot replace an answer to the other. An error does not by itself establish intentional misrepresentation.

III. Ask what the speaker could access

What could the witness actually know? An imagined participant reports that a meeting reached a unanimous decision. They were in the room but may have left before the discussion or heard only the final announcement. Another person witnessed the vote without hearing the preceding disagreement. Their accounts concern different parts of the occasion. We need not immediately classify one as truthful and the other as false. Establish which part supports each statement. I was there begins an explanation of access; it does not certify knowledge of everything that occurred. The observation has a position and limit even when the person's presence is undisputed.

Access can be indirect yet identifiable. A person may miss the discussion but receive a message from its organiser or see a signed outcome. Distinguish observation from transmission. Saying I was told does not necessarily weaken a useful answer; it may make it more accurate. We learn which further source could address the remaining question. An up-to-date notification may suffice if the purpose is arriving on time. Reconstructing the discussion requires something else. Sufficiency depends on the conclusion, rather than a wish to condemn every indirect route or treat every official route as final. Naming the route lets us use the information according to the work it can actually support.

Identify absent access too. A person may know an action occurred without knowing the motives. They may remember words without knowing whether they were a joke, or see an outcome without knowing its conditions. Such limits do not cancel the whole account. They let us use the established part without attributing more to it. In the meeting example, accept the time while leaving the reason for postponement unknown. If the reason matters, a new task arises with another basis. Do not silently fold it into the checked part of the report. Each expansion needs support of its own. Otherwise, confidence about one fact becomes permission to speculate about neighbouring facts that the original speaker never claimed to know.

IV. Clarify the area of understanding

Access differs from the ability to understand the subject. Someone may see a complicated instrument without understanding its reading. Another may explain its operation without attending the particular test. These hypothetical cases provide different grounds. An answer may need observation and specialised explanation together, with each role identified. One witness need not have both. A mistake occurs when physical presence becomes universal competence, or professional competence becomes presence at an event. Asking how the claim is known can distinguish the roles when it serves a particular task rather than demonstrating suspicion towards every statement. The question should help locate assistance rather than force one person to fill all possible roles.

Competence also has a field. Someone may understand their own work well and know little about an adjacent system. Confident speech outside their experience does not automatically inherit the original support. This concerns the structure of evaluation rather than an empirical ranking of professions. Ask what belongs to their expertise and which part of the answer is tentative. We can use assistance without first mastering the whole speciality. Even when every technical argument is beyond us, we may assess whether boundaries are stated and a further source is identifiable. Acknowledging this limit need not make us helpless; it identifies the kind of help required for the next decision.

Professional language can create an impression that an answer is complete. A term may merely identify the subject without establishing its properties in this case. If the explanation is unclear, ask for its connection to the claim: what follows for our question, and under which conditions? The specialist may reasonably say a brief explanation is inadequate. We can choose a suitable scope of work or leave the question open. Complexity need not vanish on demand, yet it does not prove correctness. Between a simple retelling and a complete expert assessment, limited assistance can make the subject, conditions, and unchecked matters visible. That space allows help without either pretending to have acquired expertise or surrendering judgment altogether.

V. Trace the route of transmission

A report can pass through several people. Imagine five acquaintances repeating that the meeting has been cancelled. All received the claim from one participant who mistook a possibility for a decision. There are five transmissions but one initial observation. The number of repetitions does not establish independent support. Another situation is different: two people independently received the final notification and can identify its content. The transmission route differs. Comparing the cases requires no suspicion of collusion. Ask where the information originated and whether the first link is shared. A shared link matters to the strength of corroboration even when everyone transmitted it sincerely.

When passing a message onward, retain its status. If someone says the time is still under discussion, no new basis supports announcing that it has changed. Shortening can remove precisely the condition that matters: possibly, provisionally, or according to someone else. Those terms may seem expendable while carrying the degree of certainty. Someone receiving our shortened version no longer sees the initial limit. Check both the fact and the promise made to the next reader. We need not attach the whole conversation chain, but a material qualification belongs in the main formulation rather than a private note the recipient never sees. Transmission is an action for which we share responsibility, even when we did not originate the claim.

Tracing also helps correct mistakes. Locate where a change arose: the initial observation, understanding, retelling, or our own transmission. The correction follows the actual difficulty. We may need updated information from the first source, or simply restore a qualification omitted in our message. One weak link need not invalidate every part of the route, yet a discovered gap cannot leave the old confidence untouched. State the finding specifically: the five messages originate in one provisional report, so cancellation is not established. That concerns the basis, not the participants' dignity. It leaves room to obtain another piece of evidence and revise the answer when new knowledge genuinely appears, rather than counting the old message once more.

VI. Hume: relate confidence to its support

In Enquiry Concerning Human Understanding, X.3–8, Hume considers testimony through experience of its relation to facts and circumstances affecting assurance. The epigraph occurs in X.4. This passage belongs to Of Miracles, so it should not be presented as a neutral modern communication manual. Our limited philosophical point concerns differences in the strength of accepting a report. We do not adopt all Hume's subsequent religious conclusions or treat his argument as experimental validation of the questions proposed here. His position retains a particular purpose and historical language; bringing it into our discussion does not erase either of those conditions.

For the imagined meeting, the distinction permits comparison without choosing only complete belief or complete rejection. A current notification and a memory of a provisional discussion provide different support. We can favour one while retaining the possibility of clarification. Confidence need not become a number. Without a defensible calculation, provisional report and confirmed notification communicate more accurately than an invented percentage. Let the reader see what supports the answer. Numerical presentation cannot repair a missing source or turn an impression into a measurement. A clear description of circumstances can convey the state of knowledge more honestly than an impressive-looking estimate with no available method, especially when the question concerns a particular message rather than a repeated measured process.

Past reliability is no permanent guarantee either. Someone who previously knew the schedule well may now have missed an update. This follows from distinguishing earlier and present grounds, not a study of that person. We can appreciate a history of help while asking about current access. Caution then concerns the report. If it becomes a demand to prove sincerity continually, the task has changed into control of a relationship. The reader need not accept that transition. Clarification calls for what relates to the question: which notification, when, and under what condition. A judgment about character is unnecessary when the gap concerns whether one message is current. Such proportion keeps an inquiry directed towards information rather than converting every check into a demand for reassurance.

VII. Reid: trust as a starting possibility

In Inquiry into the Human Mind, VI.XXIV, Thomas Reid considers dispositions to speak truth and believe others. The relevant electronic transcription identifies its text with an1813 edition; it was not collated against the printed pages. Reid connects his account with nature's benevolent arrangement and divine author. This is a philosophical and theological foundation, not contemporary honesty statistics. We use it as a position distinct from Hume's appeal to experience. Reid's confident generalisations about children or lying are not converted into measured findings. The primary passage allows us to consider his proposal while retaining those limits on its use.

The difference exposes a difficulty. Learning from another requires some initial openness to their words. If every statement needs evidence that is itself reported by someone else, checking can continue indefinitely. Yet the possibility of listening does not require unconditional acceptance of every claim. Begin communication, notice specific grounds for doubt, and clarify when the task calls for it. This working order is the article authors' proposal, not a final resolution of the philosophical dispute about testimony. We retain the question: what makes transmission possible at all, and what makes a particular trust adequate for action? A useful limited procedure need not pretend to answer both questions completely in order to help with the next report.

Initial openness should not become an obligation to trust someone whose specific support has failed. Readers may set conversational boundaries and choose another source. Respect for human speech does not cancel an identified error, absent information, or pressure. Conversely, one inaccuracy does not make all learning through others impossible. Between those sweeping conclusions remain particular tasks: establish the new time, locate a witness, or recover a condition. Philosophical positions explain trust differently, while our decision concerns a specified report. Keeping those levels distinct preserves the depth of the question and lets a small inquiry finish. We need not resolve the whole theory of knowledge or accept another person's metaphysics before deciding how to handle one piece of information.

VIII. Proverbs: attention to one's own way

Proverbs14:15 contrasts accepting every word with the prudent person's attention to their own way. The passage was consulted in the King James Version with neighbouring verses. It is religious wisdom advice, not a witness-verification procedure or an account of every Jewish or Christian reading practice. Our limited connection concerns a report and the reader's next action. One verse cannot establish a universal duty of distrust. Nor should its older moral vocabulary become a diagnosis of people who make mistakes or an experimentally supported recommendation. The comparison identifies a question without making these different functions interchangeable.

Do not manufacture agreement among the sources. Hume examines grounds of assurance, Reid proposes initial dispositions within a particular understanding of nature, and Proverbs offers moral orientation. Our reading question touches all three, while their answers belong to different tasks. Similarity helps ask about receiving someone's words; difference prevents claiming that all traditions prescribe one verification sequence. A religious reader may hear personal guidance in the verse; another may approach it as material for comparison. Discussing responsibility for transmitting a report and acting on it does not require an identical religious response. Each reader can consider the connection while retaining their relationship with the tradition and their own account of what gives the text authority.

Remember the next action: what am I planning to do on the basis of this report? A gap tolerable in a preliminary discussion can matter to a final claim about another person. This is our distinction between tasks, not a legal standard or a promise of safety. Broader consequences call for a clearer account of what is known. A report may justify asking a question without establishing the reported fact. We can seek clarification without circulating an unsupported evaluation. Attention to our own way then means identifying the transition from words to action. Responsibility for that transition remains ours, even when the initial statement came from a respected person and was passed on without intending to harm anyone.

IX. Disagreement does not cancel all testimony

When accounts differ, first ask whether they contradict each other about the same subject. One person says a decision was made; another says the discussion left them uncertain. Both may be true. Two incompatible meeting times create a specific contradiction. Check the date of the report, version of the notification, and each person's access. This is a proposed order of inquiry, not a way to guarantee agreement. A contradiction can remain unresolved. Preserve it rather than inventing an averaged account no one supports. The next action may require another source or a temporary pause, depending on the task. An unresolved difference is still a finding when we can describe what exactly differs.

Distinguish uncertainty from accusation. If we cannot yet establish the time, that is what we know. It does not automatically mean someone lied. Specific evidence of distortion can be considered separately. Caution need not require emotional distance from the person. Someone can remain dear to us while their information needs clarification; someone unpleasant can possess the right document. Choosing a basis concerns the report. Relationships need their own attention and cannot be inferred from one check. This leaves room to discuss an error without turning it into a comprehensive judgment of personality or requiring disclosure of an entire life. The inquiry should retain the scale of the uncertainty that actually gave it a purpose.

When new support appears, update the particular answer. If an account relied on an old notification, say that specifically. If the contradiction persists, retain both versions and identify what would distinguish them. Do not defend the first accepted formulation merely because it has already been passed on. Correct proportionately: a new time, a restored condition, or reduced assurance. Readers need to know what changed rather than follow the history of our self-evaluation. Testimony remains discussable when its connection with evidence can be revisited. Trust in this sense allows clarification. It need not promise that the words of someone reliable on one occasion will never need a question again, nor that every new question reverses the earlier reason for trusting them.

X. Trust specifically and ask respectfully

Specific trust can be described more clearly than I believe them about everything. In our example, the person received a current notification and transmitted the time. That supports using the time while keeping other questions separate. We need not announce this formulation in every conversation; it helps when a dispute arises or the report must be passed onward. We understand both the answer and its limit. If the next action depends on the reason for postponement, another inquiry is needed. This does not cancel friendly assistance. It protects assistance from becoming an obligation to know more than the person actually knows and preserves their ability to say that they have not investigated another part.

A respectful question concerns an available basis. Ask how the time is known or whether it remains provisional. That task does not create an entitlement to passwords, private correspondence, or information the person cannot share. Refusing to disclose a document does not prove the report false. It identifies a limit of our access, after which we can choose another route or a less confident formulation. The speaker retains personal boundaries, while we remain free not to accept an unsupported claim. Both can coexist. Transmitting knowledge requires identifying what is available without making every speaker a fully transparent source. Clear limits may leave a question unresolved, but that is preferable to pretending either unrestricted access or certainty has been established.

Returning to the beginning, testimony offers an ordinary route to understanding through people who share different parts of experience with us. Its strength concerns the words' particular connection to their subject, not sympathy or a famous name alone. We can examine the connection calmly: sincerity versus accuracy, access versus competence, and new evidence versus repetition. Philosophical and religious differences can remain without being required to solve our task identically. A finished result may be modest: the time clarified, a condition recovered, or a source identified. Sometimes the question remains open. In each case, readers can receive assistance while answering for their own transition from someone else's words to a claim or action they choose to make.

What this means in practice

Choose one report on which you plan to base a small decision. Record it without enlargement: what is claimed, by whom, and about which occasion? Ask two distinguishable questions: what access did the person have, and what helped them understand what they saw or heard? An interrogation is unnecessary. One appropriate clarification or accessible reference may suffice. Distinguish the source's observation from transmission through others. If several reports share one origin, record that rather than counting repetitions as independent corroboration. The point is to make the information route clear enough for the decision actually at hand.

Decide which action the available grounds support. Use the established part, preserve a condition, or postpone a broader assertion. Name what remains unknown. This proposed exercise promises no infallibility and is not a validated psychological intervention. Stop if further inquiry breaches someone's boundaries or is unnecessary for the task. A brief note can distinguish what is established, reported second-hand, unchecked, and where useful evidence might be found next. If information changes, update the relevant link and communicate a material correction to people who received the earlier version where appropriate. Trust and examination serve a specific question here, rather than evaluating the dignity of those taking part in the conversation.

Key quotation

Truth is always uppermost, and is the natural issue of the mind.

Thomas Reid, Inquiry into the Human Mind, VI.XXIV; stated English transcription. A thesis of his philosophical account, not a measured finding about every person.

Connections between traditions

HumeX / ReidVI.XXIV: both consulted passages treat testimony as a philosophical subject. Their grounds differ: experience of reports’ relation to facts for Hume, original dispositions in divinely arranged nature for Reid. Their dispute is not declared resolved.
HumeX / Proverbs14:15: degrees of accepting a report and attention to one’s way share a reader’s question. Philosophical argument differs from moral guidance; neither is an experiment here.
ReidVI.XXIV / Proverbs14:15: openness to trust and attention to accepted words can be compared when asking how communication begins. This is our connection: the verse does not confirm Reid’s theory, and the theory does not determine the verse’s interpretation.
Reid on language / Hume on circumstances: a connection between words and the speaker’s thoughts differs from a connection between a report and an event. These levels help us ask about sincerity and accuracy separately, without making philosophy a validated test of a person.
Three texts / present application: the article authors construct the meeting-time procedure. The passages raise questions about trust; they do not instruct us about this invented occasion or guarantee its resolution.

Primary sources

About the authors

Andrey Zarinsh is the author and founder of United Wisdom. He explores inner transformation, the unity of traditions, intercultural dialogue, and contemporary spiritual discipline. Instagram: @andrey.zarinsh.

Maria Evilkut is a co-author of United Wisdom. Alongside her research, she seeks to test these ideas in practice and observe how principles of unity and mutual understanding work in real life between people and cultures. Instagram: @ilkutki.