Why clarify?
Some difficulties do not come from a lack of information.
They appear when several realities have been compacted into the same words, the same sentences or the same frameworks.
A word may appear to be shared while carrying several meanings, functions or levels of evidence.
A sentence, too, can contain several layers at once: an observation, an interpretation, a hypothesis, a value, an assumed causality, a degree of certainty, an implication or even an implicit permission to act.
We help unpack what a word, sentence, text or exchange contains in order to distinguish what is observed, interpreted, assumed, established or still undetermined.
Make a situation clearer before it becomes a decision, a conflict, an instruction or a conclusion.
Distinguish the operations before evaluating the conclusion.
Two ways we work
Unpacking overloaded words
A word can group together several phenomena, functions, frameworks or levels of evidence. The work consists in examining what it groups together, the meanings that coexist, the frameworks in which its use changes, the functions it performs, the places where uses diverge, what it still makes visible, what it may begin to conceal, and the distinctions that become necessary.
This type of clarification can concern words such as risk, respect, responsibility, intelligence, consciousness, evidence, validation, understanding or freedom.
The problem is not always that the word is wrong. It may simply have become too broad for the level of distinction reached.
Unpacking the layers of a statement
A sentence can contain several different operations. Depending on the case, it may be necessary to distinguish observation, interpretation, hypothesis, value, implication, assumed causality, attribution of an internal state, degree of certainty, level of evidence, implicit permission or authority to act, scale of analysis or a still-undetermined element.
We do not create distinctions simply because they are possible. We create them when they become necessary to avoid distorting what is being observed.
What a sentence can contain
A seemingly simple sentence can perform several operations at once. It can, for example, name or classify something within a category, generalize from one or a few cases, interpret behavior or a situation, attribute a cause, assume an internal state, turn a hypothesis into a conclusion, present a possibility as certainty, confuse correlation with causation, present a value or preference as fact, mix levels of analysis, move from an observation toward implicit permission to act, or use a word broad enough to conceal several distinct phenomena.
A sentence can already organize reality through the categories it uses: what is named together, what is separated, what becomes comparable and what remains outside the frame.
Not all of these operations are necessarily present in every statement. The work consists in identifying those that are actually at work in the situation being examined.
A central example: unpacking a sentence
Sentence examined: “People who resist change are afraid of losing control.”
This sentence can contain a categorization: “people who resist change”; a generalization: moving from some cases to “people”; an interpretation: a behavior is described as resistance; an attribution of an internal state: fear; a causal hypothesis: fear would explain the behavior; an additional specification: fear of losing control; and, if the formulation is not cautious, a hypothesis can take the place of a conclusion.
What have we observed? What have we named? What have we generalized? What have we inferred? What have we assumed as a cause? What is established? What remains to be verified?
We do not look only at what the sentence says. We also examine the operations it performs in order to arrive at what it claims.
The unpacking is not intended to prove the sentence false. It is intended to make explicit the steps that would need to be supported for its conclusion to hold.
What a formulation can do
A formulation does not only make a proposition about reality. It can also modify what becomes thinkable, plausible, permissible or justifiable in the rest of the reasoning.
What does this look like in practice?
Example 1 — Risk
A team uses the word “risk.” For some, it refers to a measurable probability. For others, it means a potential consequence, a legal vulnerability or a human perception. An apparent agreement can then hide several different objects.
Example 2 — Interpretive statement
“This person refuses change because they are afraid.” The work can distinguish an observed behavior, its interpretation as refusal, an attribution of an internal state and an assumed causality. The aim is not to declare the explanation false, but to make its assumptions visible.
Example 3 — AI instruction
“If the user seems vulnerable, the system should protect them.” Unpacking the sentence means examining seems, vulnerable and protect: which signals? what epistemic status? what authority? which actions? when does a hypothesis become a justification for action?
Example 4 — Urgency across professions
Two professionals use “urgent.” The criteria may be different: medical urgency, legal deadline, operational risk, cost, possibility of repair or human consequence. Clarifying does not guarantee agreement. It makes it possible to locate where the disagreement actually lies.
Example 5 — Strategic decision
“We have enough evidence to move forward.” Evidence can refer to data, indicators, testimonies, a convergence of signs, an experimental result or a level of confidence judged sufficient for action. It is then necessary to distinguish what is sufficiently established to be treated as true from what is sufficiently supported to justify a given action in a given context.
Example 6 — Respect
“You do not respect me.” Respect can refer to a way of speaking, a boundary, an expectation, recognition, a decision or a value. The work does not decide what the word should mean; it makes the meanings present open to inspection.
These examples are not universal models. They show the type of clarification the work can make possible.
What you can bring to the work
You can bring a word, a sentence, an exchange, a text, a report, a decision in preparation, a disagreement, a meeting or transcript, institutional wording, an AI instruction or prompt, AI output, or vocabulary used by several professions or disciplines.
You do not need to know in advance exactly what the problem is. Sometimes the work consists precisely in locating where the confusion lies.
What this work can make possible
The work can help distinguish a real disagreement from a framework disagreement, identify a word that has become too broad, reveal a hypothesis presented as a fact, distinguish observation from interpretation, locate an assumed causality, detect a hidden unknown, reveal a difference in scale, make a question more precise, make a text or decision clearer, identify the need for additional expertise, or show that a formulation claims more than its elements allow it to establish.
The result may take the form of a structured clarification map, a set of distinctions, more precise questions or a reformulation that makes explicit the different levels present.
Clarifying does not mean simplifying at all costs.
Who is this work for?
This work can serve individuals, teams, organizations, institutions, researchers, professionals from several disciplines, designers or teams working with AI systems.
The function remains the same: to increase the precision of what can be distinguished. This does not mean that the phenomena or mechanisms involved are identical from one context to another.
What we do not do
We do not decide what a word should mean.
We do not decide who is right on behalf of the people or experts concerned.
We do not turn an interpretation into a fact.
We do not turn an analogy into evidence.
We do not replace scientific, legal, clinical, technical or professional expertise where it is needed.
We do not promise that clarification will produce agreement.
Successful clarification can also show that a disagreement is real.
A practice already at the heart of the laboratory
Methodological unpacking of language is already woven into the Aksoydan Laboratory’s research. It is used in interdisciplinary analyses, in work with AI systems and in the examination of formulations before concluding.
Rather than searching for the right definition, the laboratory asks a different question: What does this word or sentence contain here, what function does each element perform, and which distinctions become necessary in order to observe what is happening more faithfully?
We note. We observe. We translate.
Confidentiality
Words, sentences, exchanges, documents and situations entrusted to the Aksoydan Laboratory as part of its work are confidential.
They are not published, distributed or publicly reused unless explicitly requested by the person or organization concerned, within a defined scope.
Entrusting material for clarification ≠ authorizing its publication or public reuse.
Contact us
Would you like to have a word, sentence, exchange, document, decision or complex formulation clarified?