A tool as an extension of memory and reasoning

Epigraph

In an explanation, simplicity is power.

Clark and Chalmers, The Extended Mind, sentence within the notebook argument. Original English; not a general instruction to choose the simplest account without grounds.

What this article is about

A written name need not be reconstructed from memory each time. A diagram can be changed, compared with its previous version and shown to another person. Arranging cards sometimes reveals a relationship someone has not yet expressed in words. In such tasks, a tool participates in reasoning rather than merely storing a completed answer. Yet preserving a record does not establish its truth, and quickly obtaining an answer does not eliminate questions about its grounds. This article considers how external resources can help remembering and thinking while retaining distinctions between access, checking, understanding and responsibility. The central example of preparing a small book exhibition is invented. Clark and Chalmers's philosophical argument, a passage from Plato's dialogue and a religious image of a path raise different questions; we do not declare them one theory. A practical result can be modest: identify what your record does, and avoid assigning it a decision for which it provides insufficient information.

I. A record preserves definite content

Imagine an organiser writing down the titles of six books for a small exhibition. The list allows them to see which titles they entered without reconstructing the sequence each time. It does not establish that every book is in the room or that its owner has agreed to display it. Definite content has been preserved, rather than the whole situation. Other information must be obtained separately if needed. This distinction is especially useful when a tidy record creates an impression of completed checking. Good handwriting, a neat table or the right number of rows does not certify relationships between entries and objects.

The list's purpose can be named explicitly: "Books I propose considering." A book's absence from the room does not then contradict the record. Changing the heading to "Books ready for display" introduces other informational requirements. The heading participates in reasoning as much as the titles beneath it. It determines what follows from a book's inclusion. Check not only whether a line was copied correctly but whether the heading matches the action actually performed. A changed purpose cannot be replaced by attractive formatting of the same entries. It is a separate decision worth identifying.

A record can accurately represent yesterday's intention while being unsuitable for today's choice. The organiser intended to show one book, then received news of another possibility. The earlier intention does not become fabricated, but it no longer automatically determines the current decision. A date or brief status note helps retain that distinction. Its purpose is not to collect as many administrative details as possible. Such information matters where readers would otherwise mistake an earlier state for the current one. Keeping a history helps only within the inquiry we want to reconstruct; the author's whole life need not become a log.

II. A tool can participate in a process

In The Extended Mind, Clark and Chalmers propose a philosophical account of external resources participating in cognition. Their notebook thought experiment concerns a record's role and informational accessibility; habitual use matters, while past endorsement is discussed as a debatable criterion. It is an argument, not a clinical notebook trial or a rule declaring every object part of a mind.

In our exhibition, the organiser transfers titles to separate cards. They can now place books they want to compare beside one another and move one card without rewriting the list. The tool changes the available operations on represented content. It stores titles and makes an arrangement available for inspection. If the organiser notices two books sharing a theme, we still need grounds for that relationship. Cards' proximity displays a choice rather than proving a book's subject. Nevertheless, the choice becomes visible and can prompt a question that was awkward to discuss within a single long list.

Describing this case does not require first settling the dispute about the mind's boundaries. We can name definite operations: recalling through a record, choosing a pair, changing an order and comparing alternatives. A philosophical position helps raise the question of external participation; describing the work identifies what happens in our illustration. These levels do not replace each other. Cards helping a particular task is not independent proof of the entire philosophical position. A reader who rejects that position can still understand why the organiser made content available for rearrangement. Practical clarity survives different answers to the wider dispute.

It is especially important to distinguish the role of a habitual record from recommending unquestioning belief in it. The organiser can repeatedly consult a list while correcting its contents. There is no contradiction: the resource participates in work while a particular entry remains open to question. Trust in a storage method does not automatically become trust in every message entered. Our practical and ethical choice is not a new criterion attributed to the philosophers. We separately propose retaining opportunities for clarification when there is a relevant reason. Regular use should not mean prohibiting recognition of an error.

III. Accessibility differs from correctness

Suppose the list lies in a folder the organiser cannot find before the meeting. Its contents may be preserved but unavailable when needed. This is a definite limitation of the resource's function. It establishes neither that the record lacks information nor that the organiser remembers nothing. Another available ground may be used, or part of the question left open. If titles are reconstructed from memory, this new route should not be presented as reading the missing list. A similar result does not remove the difference between the message's source and what can currently be checked.

Now imagine the opposite: the folder is at hand, but one title was entered incorrectly. Fast access makes reproducing the erroneous line easy. Speed of retrieval and correctness therefore need different assessments. Convenience may be a genuine advantage despite particular content needing correction. We need neither abandon the record altogether nor declare it correct because it has been useful elsewhere. A definite check concerns this title, perhaps comparison with information on the chosen book if the book is actually available. This does not promise verification of the entire list at once.

The correction should also be identified. If the organiser changes an entry while a copy previously shown to a collaborator remains unchanged, two differing records exist. The new one does not automatically update the old. The organiser can say which title changed and offer access to the current version. Sending one message does not establish that the collaborator read it or accepted the change. Our example can retain that uncertainty. A tool helps communicate information, but transmission, reading and another person's subsequent choice are different events. Combining them creates an unwarranted sense of completion.

IV. Rearrangement does not prove a relationship

The organiser places the cards in three groups labelled "History," "Travel" and "Personal experience." A clear diagram results. Yet a book may concern several subjects, and its assigned label reflects a particular exhibition plan. Claiming it belongs exclusively to one theme would require other grounds. A narrower statement is possible: "We place it under travel because we discuss this passage." The diagram then represents an organisational choice rather than the work's complete essence. A card remains useful when its position is not expected to communicate more than it actually does.

Rearrangement creates another possible sequence for conversation. Visitors are shown a map, then a diary, then an account of returning home. The organiser may find that sequence intelligible, but it does not prove a historical relationship between the works. Neighbouring positions on a table differ from relationships in the world. A relevant connection should be stated and supported separately. If this is simply a chosen presentation order, say so. This distinction permits free use of diagrams without treating every personal choice as an objective law discovered. Our cards make a choice inspectable rather than abolishing questions about its origin.

An empty space may help too. The organiser reserves it for a book not yet selected. The space represents an open question rather than an unknown book already discovered somewhere. A collaborator asking why no card is there can be told about the selection task. Filling the gap with the first suitable title merely to finish the diagram's appearance is unnecessary. Visible incompleteness sometimes represents the work more accurately. It permits discussion of a criterion before choosing. Emptiness itself, of course, guarantees no honesty; its meaning must match what the organiser actually knows or intends to investigate.

V. Preserved words cannot explain themselves

In the consulted Phaedrus passage, Socrates considers written speech's limits in answering questions. This is a dialogue's argument accessed through Jowett's English translation, not an experiment showing records harmful. Our question is narrower: what explanations does this list need for another reader to understand its purpose?

Suppose a helper sees "done" beside a title. The organiser meant the title had been copied, while the helper inferred the book had arrived. The mark was preserved without distortion; its function was understood differently. Clarify its meaning or replace it with a more definite phrase. Every record needs no endless commentary. Identify the transition relevant to the task. A mark used only as a personal reminder may be clear to its author. Once shared, that earlier private understanding cannot automatically be attributed to all future readers. The situation of use changes even when the entries remain unchanged.

Clarification may not arrive immediately. The helper can retain the question and suspend an action dependent on its answer. This is a definite result: an ambiguity was noticed rather than replaced by a guess. The whole record need not be declared useless. Titles and other intelligible information remain visible. One mark's limitation can nevertheless be acknowledged. If the organiser is unavailable, another reprinted list need not supply the missing explanation; it preserves the same mark. The next conclusion requires new grounds rather than merely more copies of the original record.

A discrepancy may also arise because the copy itself differs from the initial message. In the invented note "do not move before an answer," losing "not" reverses the instruction. Accurate reproduction of the remaining words does not remove that difference. Preserving the instruction requires comparing negation, condition and timing rather than only the familiar title. This does not require suspecting every copy. We consider a definite transition where a mismatch has already been found. It can be corrected, the correction's grounds retained, and a separate inquiry made about any earlier action based on the other version. Corrected wording does not retroactively cancel that event.

VI. Counting depends on what is counted

The list contains six lines. The helper counts them and says, "Everything is here." In our story, that is excessive so far. Six lines may concern six distinct books or include a duplicate. Counting lines produces a definite number. Checking the presence of six selected books requires connecting each entry with its corresponding object. Counting does not supply that relationship. Name the question first, then choose an action: counting lines, checking distinct titles, or comparing entries with books. None must establish the others' results without separate examination.

Suppose one card was copied and another lost. Six cards still lie on the table. The number of elements remains while their composition changes. This simple invented case shows why quantity differs from correspondence. Where composition matters, convenient automatic counting does not complete the task. An accessible comparison with the selected set is needed. If that initial set is itself uncertain, claiming everything matches it would be dishonest. We can determine which titles are present and separately leave completeness open. A bounded result does not become false because a wider result remains unavailable.

Action marks work similarly. A tick may mean a title was checked, the book arrived, or its owner replied. When all three matter, one general tick can easily conceal unfinished work. A complex system is unnecessary. Choose clear terms for relevant distinctions and omit unnecessary marks. Our exhibition might need just "title checked" and "book in the room." Neither establishes its owner's agreement, considered separately. The tool shows definite events; its format does not authorise treating every desirable state as achieved.

VII. A shared record creates no shared decision

The organiser shows the helper a diagram and asks whether they agree with the order. Viewing and changing it are different conditions. If the helper sees the sheet but cannot move cards without invitation, no changes does not establish agreement. They may not know suggestions are expected. Explicitly name the question and a way to answer. The record then becomes an object of discussion, while the decision remains with participants. Agreement should not be inferred merely from access to material. Nor must someone object to every entry to retain the right to reject the overall plan.

Access should relate to content. Discussing books' themes need not involve exposing a private note about an owner or correspondence they did not intend for the group. The organiser can create a separate diagram with necessary information. This is the authors' ethical proposal rather than an assessment of a particular law. It distinguishes assistance to memory from disclosure of another person's information. Keeping several kinds of information together is convenient, but convenience does not answer who should see them. Giving someone a tool may change access; that change deserves identification before being treated as a purely technical operation.

Authority to change a decision matters too. A helper may correct a title's spelling without being authorised to cancel the book's display. Both operations can look like changing one card, but their consequences differ. Specify the entrusted work: checking words, proposing an order, or making the final choice. Our participants can agree directly without creating permanent offices or rules. If the assignment is unclear, its scope matters more than editing speed. The tool displays changes but establishes no authority for everyone physically able to move the paper.

VIII. A resource retains the question of direction

The selected Proverbs passage uses a path image within religious and ethical instruction. It establishes neither a method for modern lists nor a planning guarantee. We use it as an occasion for our own question about action's direction while preserving the source's different purpose.

For the organiser, the question is definite: why hold this exhibition? If the purpose is introducing different travel narratives, discuss the cards' order in relation to it. A changed purpose allows retaining the previous diagram as history and choosing another. The tool does not decide which purpose deserves participation. It can represent consequences of particular options, but people affected by them make the assessment. Our participants may choose a small display instead of a large one or abandon part of the idea. Already completed cards do not create an obligation to finish the original plan.

After much work, a finished diagram may look more persuasive than an unformatted proposal. Presentation does not establish a plan's value. Another participant may name an important fact in one sentence, such as a chosen book being unavailable. The message cannot be dismissed merely because it disrupts an attractive sequence. Check the relevant report and decide what to change. The tool helps reveal consequences when its content remains discussable. It obstructs conversation if its finished appearance becomes grounds to prohibit questions. This difference concerns how the diagram is used rather than paper's nature.

A plan and an account of a completed exhibition need different captions. Rearranging cards after the event does not alter the earlier order of books in the room. It can represent another proposal or another way to narrate what happened. These purposes should be named separately. Otherwise a helper seeing the current diagram may think they are reconstructing the past arrangement. A resource can retain history and propose a future, but one drawing does not certify both functions simultaneously. Where needed, preserve the earlier version with a brief caption and develop another alongside it. The history distinguishes events rather than prohibiting changed views or making the earlier choice obligatory.

IX. Obtaining a result and understanding an operation

Consider a separate artificial example. For a rectangular sheet with stipulated sides of thirty and twenty centimetres, the numerical product is six hundred. A calculator can help obtain the number without establishing that an actual sheet has those dimensions. Answering about that rectangle's area requires its shape, units and correctly entered values. If actual dimensions were not measured, honestly report a conditional calculation. Operational speed does not turn assumed conditions into observed facts. This illustration is not advice for construction, material purchases or another decision about a real object.

Obtaining the number differs from understanding why this operation was chosen. Someone can check the entered values while not knowing why multiplication answers the question. Name the missing part instead of declaring yourself incapable of using the tool. Seek an explanation of rectangular area or choose another question. Conversely, understanding the operation guarantees no freedom from typing errors. Knowledge and checking input therefore need not be opposed. These are different portions of a task that may assist each other. A tool has a definite function; further responsibility does not disappear when an answer appears on screen.

Return to the exhibition. Software can sort titles by a chosen feature, but the suitability of that order depends on the organiser's purpose. Alphabetical order differs from the sequence of historical events. Obtaining one does not permit claiming the other from a neat table. Check conformity to the named rule, and separately discuss whether the rule suits the task. A tool can accurately perform an operation unsuitable for a particular plan. Acknowledging this does not diminish its work. It preserves the person's freedom to choose the question instead of merely accepting the most readily available form of answer.

X. Extending memory retains boundaries

At preparation's end, the organiser can state definite accomplishments: titles preserved, cards arranged, a mark clarified and an erroneous entry corrected. Where one owner's agreement remains unknown, a position in the diagram does not replace it. Where the helper has not read a changed-version message, sending remains sending. Different states should be retained when the next step depends on them. There is no need for endless checking of every detail. Simply avoid declaring complete a part for which grounds have not been obtained. Other people's honest use of the result depends on this.

An unfinished diagram can be a stopping point too. The organiser leaves one book out pending further information or postpones the exhibition altogether. The list preserves previous work without demanding continuation. Its significance depends on a chosen task, and tasks can change. A person may retain, transform or stop using a tool if another approach suits their present question better. Neither convenience nor the philosophical image of extended memory obliges belief in every preserved word. The resource provides definite assistance that can be assessed without promising intellectual improvement overall.

A useful reader's question is: what exactly does my external record let me do now? The answer may concern recall, comparison, discussion or checking one feature. Then it is easier to name what the record does not establish: agreement, content's truth, completeness or the appropriateness of a goal. That boundary does not weaken the resource. It restores a precise purpose to its work. External means can participate in reasoning through their content and transformations; participation alone should not transfer someone else's responsibility to them. We can trust a definite operation, retain an open question and freely choose whether to continue.

What this means in practice

If you want to try the distinction, choose one ordinary record of your own without others' private information. Name its present purpose: recalling a title, comparing possibilities, or showing a proposal. Moving all records into a new tool is unnecessary. Consider an existing sheet, or simply retain the question. Choose a task that creates no unwanted commitments and requires no publication of its content.

Read one entry and distinguish what it states from what you are willing to infer. A listed title, for example, shows a record was made rather than an object being present. If the next conclusion matters, seek suitable accessible grounds. If unavailable, retain uncertainty. Where useful, clarify a heading or mark's meaning; the change should address a definite question rather than a desire for a finished-looking table.

Finish with a definite account of the result. Identify what was preserved, checked or left open only where these distinctions serve your task. Another person need not adopt your record's format. This is an original voluntary exercise rather than treatment for memory or a measured method for improving intelligence. Its purpose is a resource's clear function and a free decision about the next action.

Key quotation

Ponder the path of thy feet

Opening clause of Proverbs4:26, King James Version. Religious-ethical context retained; not a planning-efficacy rule.

Connections between traditions

Clark and Chalmers / cards: external participation is compared with definite authorial operations; the invented exhibition proves no entire philosophical position.
Plato / shared record: written response is compared with clarifying a mark; the ancient argument becomes no study of present tools’ harm.
Clark and Chalmers / Plato: resource roles and limits of written explanation are different questions; no unified theory or identical teaching is claimed.
Proverbs / our plan: path and action’s direction are compared while religious ethics and a contemporary card order remain distinct.
Three sources / responsibility: content checking, consent and freedom to continue are our practical proposal, without memory/intelligence-improvement guarantees.

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.