Researcher standpoint and the possibility of objectivity

Epigraph

“The human understanding resembles not a dry light”.

Francis Bacon, Novum Organum (1620), Book I, aphorism XLIX. The quotation follows the English edition; RU uses our own translation. Bacon continues by discussing the influence of will and passions. Edition edited by Joseph Devey.

What this article is about

A researcher wants to improve a city square. He likes quiet spaces, walks a great deal, and considers vehicle traffic the main source of inconvenience. A woman living nearby more often arrives by car with her elderly mother. A trader notices morning deliveries; a teenager notices evening lighting. They share an object of attention, but different questions are available to them.

Can knowledge be objective when research begins with a person who has a particular life and interest? The answer depends on what objectivity means. If it requires the complete absence of a standpoint, the task appears impossible. If it means making the grounds available for examination and refusing to subordinate the result to a wish, standpoint becomes something to work on. This article examines how a question, method, data, and conclusion come together in such a procedure. It neither declares all opinions equally valuable nor promises perfect impartiality. Its practical aim is to distinguish interest, a biased conclusion, and responsible inquiry that allows the world to be different from what its author hoped.

I. A standpoint begins before the conclusion

A researcher’s standpoint is the position from which they ask a question and gain access to its subject. It includes skills, equipment, language, purpose, and relationships with participants, as well as political convictions. Someone who considers themselves neutral still chooses a scale, a unit of observation, and an appropriate explanation.

In our hypothetical urban scene, one could measure the number of cars, the time needed to cross the square, or access to a shop. Every indicator could be accurate. But they answer different questions. If vehicle flow is the only measure chosen, it is difficult to claim later that the place’s convenience for all residents has been studied. The limitation arose at the beginning, rather than when the results were interpreted.

A standpoint is therefore not an admission of unreliability. It can explain why a question became visible at all. Personal experience may reveal a problem, a professional habit suggest a method, and a public purpose make the inquiry significant. Each is useful as long as it does not become a compulsory answer.

For example, someone wants the square to be safer. This is an understandable goal. The claim that a particular redesign will improve safety, however, requires information. One can retain the goal while revising the means. If affection for the proposed means becomes the measure of whether data are true, a standpoint turns into bias.

The first step toward objectivity is to name this sequence. Why am I examining the subject? What question am I asking? What can I see from my position? What does it prevent me from seeing? This account need not become a lengthy autobiography before every measurement. It is enough to communicate the significant conditions affecting the conclusion. Readers need to understand the investigative action, rather than receive an exhaustive portrait of the investigator. A standpoint helps as a map of access and limitations; it obstructs inquiry when it becomes a universal explanation of everything a person says.

II. Objectivity has several jobs

The word “objective” is used for different requirements: measured accurately, confirmed independently, considered consistently, or free from subordination to personal interest. These requirements are related but distinct. One can correctly count an unsuitable quantity and obtain an accurate answer to the wrong question.

In the hypothetical study of the square, two observers count vehicles in the same way. Their agreement increases confidence in the count. It tells us nothing about people who stopped visiting because access was inconvenient. Absent visitors do not automatically enter the observation. The problem concerns the completeness of the question, rather than arithmetic.

Before a dispute, it helps to name the particular job objectivity must do. If the doubt concerns an instrument, check the measurement. If it concerns the sample, check selection. If it concerns the conclusion, examine the move from data to assertion. If it concerns the researcher’s interest, ask whether a result contrary to that interest is possible. One imposing word cannot perform every check at once.

Objectivity is also different from an absence of emotion. Someone may strongly wish to correct a problem and carefully measure its conditions. Another person may speak dispassionately while concealing inconvenient data. Tone creates an impression; it cannot replace access to the procedure. Restraint can help a conversation, but it is not independent evidence of truth.

We must also distinguish an object’s independence from our wishes from an observation’s independence from conditions. A wish does not change the length of the crossing, but measurement conditions may change the recorded time. This does not make the quantity arbitrary. It requires us to say how we obtained it. Once these meanings are separated, objectivity becomes a concrete set of actions. It ceases to be an honorary title awarded to someone supposedly excused from explaining their own grounds.

III. Bias can look like consistency

A person may conscientiously look for confirmation while systematically avoiding anything that could change the conclusion. The danger lies beyond deliberate fabrication. Sometimes the rules of inquiry quietly adjust themselves to an already pleasing picture.

In aphorism XLVI, Bacon describes the understanding’s tendency to support an accepted idea and dismiss contrary examples. This is an early modern philosophical diagnosis of reasoning, rather than a contemporary experimental assessment of everyone. It helps as a question about procedure: what happens to the inconvenient case? Novum Organum, Book I, XLVI–XLIX.

In a hypothetical urban project, a resident’s positive response is treated as an indicator of success. A negative response is described as attachment to the old arrangement. The latter explanation may sometimes be right. But if it is applied to every negative response without a separate check, the result is protected against refutation in advance. The researcher applies different rules to welcome and unwelcome information.

It therefore helps to identify what observation would prompt a revision of the hypothesis. This need not be one decisive fact. A complex project requires several indicators and an observation period. But there must be a possible result that will not automatically be renamed a hidden victory. Otherwise data merely illustrate the conclusion.

The opposite trap is declaring oneself honest simply because one has found a flaw in one’s own idea. A single admitted drawback can make an overall account persuasive while leaving its other transitions unchecked. It is better to consider significant grounds separately: what is confirmed, what is limited, and what is unknown. Acknowledging complexity helps when it changes the conclusion or the conditions of action. A decorative qualification at the end does not repair a confident assertion at the beginning.

IV. A visible standpoint does not remove the need for evidence

Disclosing an interest matters, but it is insufficient. A researcher may honestly say they love pedestrian spaces and then draw a weak conclusion about their benefits. Transparency explains where the question came from; it cannot replace information about consequences.

In a hypothetical discussion, the author answers an objection: “I stated my position from the start.” But the reader asked why the data support the chosen means. The answer addresses another part of the work. We can appreciate the honest disclosure of interest while continuing to ask for justification.

A general list of social characteristics is similarly insufficient. It may reveal significant conditions of access, but it does not automatically explain every methodological decision. The specific connection matters more: the researcher observed the square only during the day, did not use its entrance with a stroller, or obtained information from shop owners. These circumstances make the boundaries of the picture understandable.

It also helps to distinguish an acknowledged influence from an unconfirmed guess about influence. People know why they chose a subject, but do not always know how their habits affected the conclusion. “I accounted for all my biases” is too strong. A more accurate formulation names the checks performed and leaves room for another error to be discovered.

Readers have a corresponding responsibility. An argument cannot be refuted merely by observing that its author has an interest. Interest may explain the choice of question and justify closer scrutiny of the method. But the conclusion’s falsity must be established on its grounds. Otherwise criticism of standpoint becomes a convenient substitute for substantive disagreement. Objectivity begins when authors make their work available for scrutiny and critics actually examine what was said, rather than construct a portrait of the speaker.

V. A method turns a question into a bounded action

A reliable method does not promise to know everything. It establishes which part of a question can be checked under given conditions. That limitation makes the result intelligible. If a method checks crossing time, the conclusion should primarily concern crossing time.

A hypothetical team compares two routes. For the result to mean something, it must specify the start and finish, time of day, ordinary obstacles, and participants. If a sunny day is compared with a rainy evening after the route changes, the entire difference cannot be attributed to the project. Every possible variable need not be included, but significant conditions must be recognized.

The method depends on the question. Counting suits the frequency of events; interviews describe experience; observation can establish a sequence of actions. No instrument has universal priority. A table does not automatically establish causation, and a story does not establish frequency throughout a group. Each approach provides information of a particular kind.

What the method cannot see also needs to be retained. A survey addressed only to current visitors may miss why former visitors left. An average time may conceal especially difficult cases. These limitations do not necessarily invalidate the work. They alter the scope of the conclusion and the direction of the next question.

A method helps make inquiry objective when someone else can understand what was done and relate the result to the task. No absolute observer outside the world is needed. What is needed is specified conditions, transparent transitions, and the possibility of an independent observation. Even a small study can then produce limited but solid knowledge. An indefinitely broad answer is weaker, however impressive it sounds: it promises more than the action performed can confirm.

VI. A plan and a discovery belong to different stages

Research may test a hypothesis in advance or search for a pattern not yet known. Both activities have value. The error occurs when something found after examining the data is presented as having been predicted beforehand. Readers then misjudge the strength of the test.

The Center for Open Science explains preregistration as recording a plan before research or analysis to distinguish planned work from subsequent exploration. It makes that boundary more transparent, but does not turn a bad plan into a good one or prohibit discoveries. Center for Open Science, “Preregistration”.

In our hypothetical project, the team expected the new arrangement to shorten the route. Measurements show no shortening, but reveal fewer stops. This is interesting. It can be described as a new observation and tested further. The number of stops cannot retrospectively be declared the main indicator all along if the earlier plan says otherwise.

Rigid adherence to an unsuitable plan does not make research reasonable either. Conditions may change; a measure may prove inappropriate. A change should then be recorded, its reason explained, and the new analysis’s status identified. Honesty means preserving the sequence of the work, rather than forbidding thought once it has begun.

Formal registration is usually unnecessary in everyday life. Recording beforehand what you want to check and which indicator will be used for comparison is enough. After acting, note the unexpected separately. This small discipline allows learning without rewriting your own intention to fit a successful outcome. The researcher’s position becomes visible over time: what they assumed, what they observed, and what they changed. Objectivity gains both a method and a history of its application.

VII. Criticism is needed where it can be taken up

A single researcher cannot see every assumption of their own. Other people therefore have a role in knowing. They may notice a missing condition, suggest another explanation, or examine a transition the author considers obvious.

Longino identifies four conditions for a critical community: available forums for objection, actual uptake of criticism, shared standards, and qualified equality of intellectual authority. The last does not require everyone to have identical competence on every subject. “Cognitive and Non-Cognitive Values in Science,” p. 40.

In our urban scene, this means more than inviting residents to a presentation. They need to see the grounds, ask a significant question, and receive a response. If the project has already been declared successful and every comment goes into an optional appendix, criticism does not contribute to knowledge. It serves as public decoration.

Taking a criticism into account, however, is different from agreeing with it. An objection may be mistaken or address another task. It can be rejected with a stated reason. If rejection is indistinguishable from ignoring, participants cannot see what happened. A record therefore helps: the substance of the comment, the check, the decision, and the remaining question.

Criticism may yield a negative result for the initial idea and a positive one for the inquiry. A team may abandon one change while retaining another. This is no defeat for objectivity. It shows that the author’s standpoint encountered independent grounds. The more dangerous situation is one where everyone politely confirms a smooth account because they lacked the time, access, or right to change it. Agreement has value through the procedure that produced it, rather than the number of raised hands.

VIII. Diversity cannot replace organization

Different participants can broaden the range of conditions noticed, but a diverse membership does not by itself guarantee reliability. People with different lives may be brought together and given equally little influence. A task may remain unexplained and produce many incomparable answers.

Residents, traders, and specialists attend our hypothetical discussion of the square. Some discuss deliveries, others noise, and others safety. If the facilitator demands an immediate choice of one improvement, differences become competition for a resource. Inquiry first needs to determine which questions address one feature and which require separate criteria.

Diversity becomes useful through a procedure for combining information. Participants describe conditions, the researcher checks comparability, and the group distinguishes established facts from conflicting goals. Some interests really are incompatible. Objectivity need not eliminate that conflict; it helps show its costs without inventing a benefit for everyone.

Opportunities within the group are unequal too. One participant commands professional language, another knows an everyday difficulty, and a third controls documents. If authority follows only presentation style, some knowledge remains invisible. If every personal account is treated as final, scrutiny disappears. Competence must correspond to the subject of the question.

Roles should therefore be distinguished in advance: who supplies observations, who evaluates the method, who decides, and who checks the consequence. These roles may overlap, but must not merge unnoticed. Someone who approved a project does not become independent confirmation of its result simply by later signing the report. Different positions improve inquiry when their differences retain meaning and reveal what one position could not examine independently.

IX. Independence has a specific object

A second voice does not necessarily mean a second independent basis. Two reports may use the same data, two experts the same retelling, and two measurements the same faulty setting. The number of confirmations must be related to their origins.

In a hypothetical dispute, three people cite one table and appear unanimous. They agree with an interpretation of the table; they have not made three separate observations. If the table is wrong, repetition does not correct it. Several independent analyses of one table may nevertheless help check a calculation. The particular kind of independence must be named.

Data, analysis, and decision can be distinguished. Independent measurement checks data; recalculation checks computation; a different argument checks explanation. Confirmation at one level does not automatically confirm the next. A correct calculation can support an excessively broad conclusion.

This approach does not require endlessly checking everything. The cost of an error determines a sufficient scale. A small, reversible decision may need only direct observation and a clear explanation. A significant conclusion requires stronger grounds. A check should resolve a particular doubt, rather than add another authoritative signature.

The researcher’s own independence has limits as well. Absence of direct payment does not imply absence of interest: reputation, attachment to an idea, and team expectations can also matter. This is a reason to disclose significant conditions and provide scrutiny, rather than declare every inquiry unreliable. Independence makes sense as a relation to a particular risk. It cannot be obtained once and forever, then used as exemption from explaining a procedure.

X. Data resist, but do not speak for themselves

A measurement may turn out to be unexpected and inconvenient. This resistance matters: the world need not confirm the researcher’s wishes. But the result’s meaning still requires reasoning. A number does not state its own cause, scope of application, or next step.

In our hypothetical project, crossing time decreases. We can conclude that people crossed faster under the measured conditions. Without further grounds, we cannot claim that every resident’s quality of life improved or that accessibility was solved. Those conclusions concern other features and people.

Distinguishing change from its cause is especially important. If a new transit stop opened nearby, the composition of visitors may have changed independently of the square’s redesign. The observation remains real, but its explanation needs examination. A bounded conclusion is no worse than a broad one: it shows more accurately what work the data actually did.

A negative result also has several meanings. Failure to detect an effect may mean no effect exists under these conditions, the measure is insufficiently sensitive, or the wrong feature was tested. A convenient explanation cannot be chosen automatically. The method and available information need examination. Sometimes the honest conclusion is that there are insufficient grounds to decide between these possibilities.

Such uncertainty does not prohibit practical action. One can choose a reversible step, state its grounds, and check it separately. But a decision under uncertainty cannot be presented as established causation. Objectivity preserves the relationship between the strength of evidence and the strength of language. The world has been given an opportunity to object; the researcher must not drown out that objection with an overconfident account.

The boundary of a conclusion also matters when results are passed on. A brief account can easily lose conditions while retaining only a promise. The main conclusion should therefore carry a significant limitation alongside it: who was observed, which period was compared, and what was not checked. Readers need to distinguish an established change from a hoped-for next effect even when they do not open the complete research account.

XI. An error does not destroy a researcher’s standpoint

Objectivity is tested after an error is found too. Someone may correct a calculation, revise a conclusion, and preserve the history. They may instead defend their reputation by renaming the error an acceptable simplification. The difference appears in dependent consequences.

A hypothetical author discovers that part of the waiting time was excluded from the indicator. Replacing one number is insufficient. The conclusions depending on it must be examined: the promised gain, route comparison, and decision about further action. If the central conclusion survives, explain why. If it does not, correct it along with the data.

A correction history matters for knowledge. It shows what was checked in the new version and which version the previous result concerned. Without it, a familiar “checked” label can accompany an altered text. Readers cannot tell which conclusion survived which examination.

An error does not make someone incapable of knowing altogether. Its seriousness depends on mechanism, consequences, and response. Admitting a bounded error does not mean accepting every criticism. If a comment is unconfirmed, the author can show the check and retain the conclusion. Openness and independent judgment both survive.

The inner task is to stop tying dignity to infallibility. An author’s standpoint can remain significant even when a means fails. The aim of improving the square does not require declaring every project a success. Reliability grows through enough separation from the current conclusion to see new grounds. This is not indifference to one’s work. It respects the work as inquiry whose result can still surprise its creator.

XII. Objectivity restores the right to a difficult choice

Good research does not always deliver one compulsory recommendation. It may reveal several options with different consequences. A decision requires a value choice that cannot be hidden inside the word “scientific.”

In our urban scene, one option reduces noise, another makes deliveries easier, and a third preserves vehicle access for people who find walking difficult. A combination may be sought, but cannot be promised without checking. Even complete data do not automatically decide which compromise the group should accept. They clarify the cost of choosing.

The researcher’s standpoint becomes visible again here. An author may prefer one criterion and openly justify it. But a truthful account of other consequences must be preserved. A value guides a decision; it gains no right to rewrite a fact that makes the decision difficult.

In On Liberty, Mill defends serious consideration of others’ grounds as a condition of understanding one’s own opinion. For our question, this is a useful comparison: a choice becomes more meaningful after encountering a strong alternative. Mill’s liberal argument is not a statistical analysis manual and cannot replace one. Chapter II.

Objectivity therefore does not require leaving human life behind. It requires relationships with the subject and with other people in which desire does not have the last word about what happened. The researcher remains a person with an interest, but the conclusion becomes a shared object of scrutiny. At the beginning, he saw the square from his own position. At the end, he can explain both his view and what it overlooked. This capacity does not cancel choice. It restores its grounds, boundaries, and responsibility for consequences.

What this means in practice

Choose one question on which you want a particular result. Record your desired goal separately from a testable claim. For example: “I want to organize my day more conveniently” and “the new arrangement will reduce unfinished tasks.” These sentences are related but not identical.

Determine what you are observing, under which conditions, and which indicator will support comparison. Name a possible result that would change your hypothesis. No academic format is required for every everyday decision. It is enough to avoid a rule under which every outcome proves success.

After acting, separate the planned check from an unexpected discovery. Both may help. If you change the criterion, retain the reason. Ask another participant about a significant condition you may have missed, and show what happened to their comment.

State a conclusion proportionate to the data. Identify what is established, bounded, and unknown. You may make a reversible decision with incomplete information without turning it into a universal explanation. Check consequences through a separate observation.

If an error is found, correct dependent conclusions and retain a version trail. If criticism is unconfirmed, explain the check and continue. Objectivity is the possibility of this movement: from standpoint to question, from question to specified action, and from result to revision. People do not disappear from inquiry; they make their work clear enough for others to examine what, precisely, it gives them grounds to trust.

Key quotation

“man always believes more readily that which he prefers”.

Francis Bacon, Novum Organum, Book I, XLIX. The quotation follows the English edition; RU uses our own translation. This is a historical philosophical claim about a tendency of judgment, rather than a measured probability of error for every researcher. Text.

Connections between traditions

Comparison	Similarity and boundary
Bacon: resistance to an accepted idea ↔ examining an inconvenient result	The shared requirement is not to dismiss contrary cases. Modern methods have their own justification and do not follow directly from the aphorism.
Longino: a critical community ↔ independent examination	Criticism must affect knowledge. Several opinions without a procedure do not guarantee objectivity.
Preregistration ↔ a personal record of the initial expectation	Both preserve the sequence of intention and result. An everyday note does not become formal scientific registration.
Mill: a strong alternative ↔ checking an interpretation	Understanding benefits from objection. Freedom of discussion does not replace measurement and control of conditions.
Researcher standpoint ↔ responsibility for a decision	Interest explains the question, while data limit the answer. A value choice cannot rewrite established consequences.

Primary sources and research

Francis Bacon. Novum Organum (1620), Book I, XLVI–XLIX. The specified aphorisms were read in the English edition edited by Joseph Devey; a historical critique of tendencies of understanding and the source of both quotations.
Helen E. Longino. “Cognitive and Non-Cognitive Values in Science: Rethinking the Dichotomy” (1996), pp. 39–41. Passages on pp. 39–41 were read: background assumptions and the four community norms.
Center for Open Science. “Preregistration”. The definition and sections on planned and exploratory work were read; an official methodological source, rather than a separate experiment on registration’s effectiveness.
John Stuart Mill. On Liberty (1859), Chapter II. The passage on testing opinion through objections was read; the philosophical boundary between discussion and method.

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 coauthor 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.