Appearance
Why Agazan
Why Agazan exists, what it does not claim, how its features are chosen, and a tour of the features that serve its aims.
On this page
Purpose
Agazan encodes psychological concepts and self-improvement techniques into vocabulary and grammar so that using the language nudges you toward better habits. For example, a single word for can’t can say whether the block is only temporary, may still change, or will not. That split keeps a hard moment from hardening into a fixed identity about who you are.
The design aims at three outcomes: compassion, rationality, and empowerment. Each is described with the features that serve it under Psychological features.
Try it in English first
Do you have to learn a whole language to use these ideas? No. Claritish is short lessons on dropping single Agazan words into everyday English, where an English adverb or tag would go. In The package arrives Friday thovum, one word says it could be so, and English has no single word for that. Other lessons cover how you know, how far a label reaches, thanks that name the need, and more.
Every word Claritish teaches is already real Agazan, so nothing you learn there is thrown away. Start with the Claritish introduction; each lesson is short.
Limits
Agazan encodes psychological concepts and techniques, but it is not therapy, counseling, or medical advice, and it is not a substitute for professional care when you need it.
The design aims at compassion, rationality, and empowerment, and prefers research when it exists (feature criteria). That is design intent, not a claim that speaking or studying Agazan has been shown to improve mental health or other outcomes.
The language is currently in beta: grammar, lexicon, and these docs can change as the design settles.
Why language can help
Unhelpful thoughts are often automatic: catastrophizing a small setback into “my life is ruined,” treating two options as the only ones, or labeling a feeling as anger without naming which sake is unmet. (A sake is something that is good for you, such as autonomy, competence, or connection with others.) Healthier framings exist, but they are hard to reach in the moment.
Agazan puts those framings into ordinary words and optional extra pieces on words. When you speak or think in the language, the helpful distinction is already sitting in the sentence shape.
Criterion for features
A psychological feature must meet these criteria:
- Helps with language goals. Supports compassion, rationality, and/or empowerment. Prefer research when it exists; user testing is fine when research is scarce.
- Addresses a common problem. Targets biases or friction that show up for most people. Rare edge cases do not justify heavy machinery.
- Easy to use. Explainable in roughly one paragraph plus a couple of examples; usable in sentences with at most about one extra second of thought after practice.
- Avoids shame. No option should read as the socially “correct” default. Each choice has situations where it belongs.
- Reminders where they are needed. Nudge compassion, rationality, or empowerment mainly where those failures are likely. Some extra nudges are fine; pointless ones should stay rare. Nudges stay optional and targeted.
Psychological features
This section tours how the three aims show up in vocabulary and grammar. The features are grouped by the thinking trap each one answers, and the groups sit under the aims. Each feature says what English blurs, what Agazan marks instead, and what that helps you do; the linked pages teach the actual forms. A group names the habit its features push against, not a bias they have been shown to cure (limits).
Compassion
Support self-acceptance and acceptance of others, and make judgments visible when they arise.
Praise can name the sake a person’s work actually served instead of stopping at a vague good. A flash of anger or anxiety can be described by the unmet sake behind it, how strongly it stirs, and where it sits, so the feeling points at something you can act on instead of stopping at a blank label. And can’t comes in kinds: temporary, modifiable, or irreversible. A block never has to read as who someone is. Because says what made something happen without claiming anyone is to blame, unless you choose to say so.
Treating a moment as identity
A hard moment, a label, or a past pattern hardens into a fixed fact about a person. The features here let you say how far a label reaches, how permanent a block is, and which past events still count today, so a bad stretch stays a stretch.
How far a label reaches
English Alahen is angry can mean one outburst or the kind of person Alahen is. One act slides into a verdict on someone's nature, and a name for a problem (lazy, anxious) starts to sound like its cause.
Label scope lets any label say what kind of claim it makes: about this stretch of doing, about a practiced role, about one relationship (a stranger to Azawan), or about a type that follows the person across scenes.
Marking a label's reach helps you describe what happened without turning it into who someone is, including yourself.
In English: Claritish: Labels.
Can, can’t, and won’t
English I can’t can mean a temporary block, a lasting inability, or a choice not to.
Ability says whether someone can do an activity or have a property, and for can’t, which kind: not right now, not now but it may change, or not ever. Choosing not to is an ordinary won’t, not a kind of can’t.
Saying can’t right now or may yet change instead of impossible helps you stop treating a temporary block as a fixed incapability.
What still counts, and what used to be true
English has… and used to mix two questions: when something happened, and whether it still matters now. Used to also leaves open whether a habit still describes you today.
RESIDUE says the result of a past event still counts now (a debt still unpaid, a door still shut). FORMER says a pattern was usual but you are not claiming it as how things are now (cue: usual weather vs today’s report). Neither word is a past tense: each can combine with what you see now, what you remember, or a forecast.
Marking residue keeps rumination optional: an event you mention without it counts as closed. Marking a former pattern lets “I always was this way” become a description you can stop giving. As-of names the moment that leftovers and patterns are measured from (as of Friday), so you can weigh an old situation from its own date instead of against today. For English if they had … they would, the as-of moment is a placeholder you do not claim is real, which is not the same as pretending.
Treating a feeling as a label
A feeling, a thanks, or a should arrives as a single opaque label, so what is met or unmet and what would help stay out of view. Sakes name what is at stake, and emotions are built from them.
Sakes, gratitude, and oughts
English often blurs I want, I need, I should, and this is good for me, so gratitude, motive, and ought run together.
Agazan’s sakes are a fixed list of things that are good for a person (autonomy, competence, understanding, purpose, relatedness, beneficence, pleasure, physical, or unspecified), and each sake word also says how you stand toward that sake:
| Stance | Meaning |
|---|---|
| Met | This serves the sake, including whether the payoff lasts or is only for now (gratitude / savoring) |
| Motive | Acting for this sake, including where the reason stands |
| Prescription | This act ought to be done for this sake; the word’s ending (its last letter) says whether you are inviting it, offering it, or trying it (invited / offered / trial) |
| Unmet | Costs the sake, including how changeable that is |
Naming the sake and your stance toward it helps you notice gratitude, motive, and oughts instead of blurring them into a vague want, should, or good.
In English: Claritish: Oughts and motives.
Emotions as composition
Opaque labels (angry, anxious, proud) hide which sake is in play and what to ask for.
Agazan builds an emotion from three parts: a sake and how it stands (usually met or unmet), where the feeling sits or what it points at, and its motion (surging, flowing, or still).
Building the feeling from those parts helps you ask for what would actually help instead of stopping at the label.
In English: Claritish: Feelings in three parts.
Thanks that name the need
English thank you says you are grateful but not for what.
Agazan has no set phrase for thanks. You say which of your sakes was met, and saying it to the person who acted is the thanks. The ending says whether the payoff lasts or is only for now. To name someone else's sake instead, put that person right after the sake word (whose stake). Put the sake word right after a thing you own and it thanks that thing, with no separate word for my, so appreciating what you already have takes one word (personal possession).
Naming the need that was met makes thanks specific, which helps both of you notice what actually helped.
In English: Claritish: Thanks and sorry that say why.
Treating a cause as a fault
An explanation of what made something happen slides into blame, and one cause among several carries the whole outcome. The feature here keeps cause, blame, and a share of the cause apart.
Cause, blame, and share
English because of you can mean you made it happen, you did something wrong, or you were one part of it.
The ending on a because word says which claim you make: plain cause with no fault (thavem), a reason that broke a norm (thavel, it's their fault), or one share among other causes (thaver, partly because). Plain cause is the default.
Keeping cause apart from fault helps you explain what happened without blaming, and saying partly keeps one person, including you, from carrying a whole outcome.
Rationality
Surface common biases that make self-delusion easy.
Claims can say how you know: memory, a report, a pattern, clues, a deduction. A general claim says whether it allows exceptions and how many cases it rests on.
A because or an if says what kind of link it claims. When the “must” is really an ought, the sentence shows the value behind it instead of treating a wish as necessity.
Choice lists can say whether they are open or closed, so two options cannot smuggle themselves in as the only ones.
Treating a guess as fact
A guess, a rumor, a hazy memory, or a wish gets treated as settled fact, and a few cases get treated as a rule. The features here show what a claim rests on: how you know, how strong that evidence is, whether it is only possible, and whether a general claim allows exceptions. Forecasts belong here too, but they are taught together with plan and decision under wanting, planning, and deciding.
How you know, and could be
It is easy to treat a could be that keeps looping in your head as settled fact, and to lump memory, hearsay, gut feel, and maybe they left together as if they were equally good grounds.
Evidentiality lets a claim about the world name how you know: a live look, memory, a pattern, clues, a report, or a gut sense. MEMORY is recalling a scene from earlier, and memory can be wrong. A live look (LIVE) is sensing it while it is still in view. Name the channel once and it carries over your next claims, so a whole story can rest visibly on one so they say.
The ending on that evidence word says how strong the evidence is (a vivid memory or a hazy one, a reliable source or a rumor), separately from how likely the claim is. Marking how you know helps you catch overconfidence and hearsay before you treat a take as settled fact.
A deduction is its own channel, apart from strong clues: it says the claim must hold if its premises do, and its ending says how far you checked the steps. It obviously follows is easy to say about a conclusion the clues only make likely; keeping the two apart, and marking whether you grant the premises at all, helps you catch a leap dressed as a proof.
In English: Claritish: How sure are you? and How do you know?.
Always, usually, and how many cases
English always and never sound absolute even when you mean usually, and they do not say how many cases they rest on.
A general claim says in its ending whether it allows exceptions: closed always (hual) allows none, and open always (huam) leaves them open, like usually or as a rule. A PATTERN evidence word says how much the claim rests on, from a well-established pattern (thobal) to only a few cases (thobar).
Saying usually when you mean it, and from a few cases when that is all you have, helps you catch overgeneralizing before one bad day becomes it always goes wrong.
Treating a wish as a law
A link between events, a habit, or an ought gets treated as a law the world must obey. The feature here says what kind of link a claim makes.
Causes and if
Everyday English because, leads to, and have to blur three splits: a condition that is enough vs one that is required, a reason in the world vs your grounds for saying something, and a wish vs a law of nature.
Causation gives each link its own word: if, only if, because, and if and only if. An intended result (so that) uses a separate word, hogom (so that): the action is aimed at that outcome without claiming it happens, and it is not the same as acting for a sake (motive, tho). When the “rule” is really an ought (hard work should earn promotion), say it with a sakes prescription (the) on the act, not with a causation word.
Keeping those splits apart helps you stop treating wishes and habits as if the world must obey them.
Treating a few options as all of them
Two options pose as the only ones, one way of seeing a situation poses as the situation itself, and an unstated standard decides the ranking. The features here show how many options are in play and what standard is being used.
Open lists, closed lists, and false dichotomies
It is easy to present two options as if they were the only ones possible.
Joins mark what kind of list you are building (inventory, exclusive choice, inclusive and/or, rank, …) and whether the listed items are the only ones in play (closed) or still open. Open forms leave room for alternatives you have not listed.
Saying whether a list is open or closed helps you notice false dichotomies, and leaves room for options you have not named yet.
Alternatives, reframes, and judgment bars
Work talk often freezes on the problem or the solution, or ranks options against a standard nobody has said out loud.
Prefer numbered alternatives (ordinary nouns plus an ordinal, like solution 2) so candidates stay plural and revisable. Reframe the same situation with softer wording rather than inventing a second candidate. A comparison like worse than… names its bar, the standard it measures against (than usual, than I expected, than you demand), so the standard stays visible, including when it is only your own expectation.
Numbering candidates, rewording, and naming the bar help you keep choices and standards revisable instead of freezing on a single framing or a hidden standard.
Empowerment
Support authentic choice by making reasons for action visible.
The motive or ought behind an action can be named with the sake it serves instead of collapsing into a bare should. A must or a may says who it comes from. Wanting, planning, and deciding stay separate, and a decision can mark how revisable it is. An attempt can be a low-stakes trial, a plan can name its backup before you start, and enough is measured against what you need rather than against perfect. Consent says whose yes it is and how binding, and an opening can say how long the talk will last before it starts.
Treating a choice as forced
Wanting, needing, and being obliged blur into a bare should, and a plan or a pick sounds locked. Sakes already name the reason behind an action (motive, or an ought you choose to take on). The features here show where a must comes from and keep wanting, planning, and deciding apart, so the choice stays yours.
Who says you must
English must, have to, and is supposed to sound equally binding, whether a law requires something, a person demands it, or people only expect it. May hides the same split for permission.
A requirement word says where a must comes from: a rule, a person who demands it, or only an unwritten expectation that nobody actually demands. A permission word does the same for may: a rule allows it, someone said yes, or nobody objected.
Naming the source helps you notice when a have to is only an expectation, so you can decide whether to meet it.
In English: Claritish: Allowed, required, agreed.
Wanting, planning, and deciding
English will mixes plan and prediction, and I want to can sound like a promise. It is easy to talk as if the future were already known (a certain prediction that fuels anxiety or overconfidence), while a pick often sounds more locked than it is.
Agazan splits them: want for what you wish for, without meaning to do it yet; PLAN for what you mean to do; DECISION for how revisable a pick is; and a forecast for what the world will do. A forecast has no word of its own: you say how you know, with the same evidence words as above, and add bral (later). So every forecast says what it rests on, and its ending says how strong that evidence is.
Splitting these helps you hold a want without owing it to anyone, avoid treating a wishful or “certain” forecast as settled fact (or as a commitment), and keep a revisable pick from sounding locked forever.
Compare with: English will. Use PLAN for intention, an evidence word plus bral for a forecast, and DECISION for how firm the pick is.
In English: Claritish: Wants and plans and Decisions and tries.
Treating effort as all or nothing
Every attempt can feel like a full commitment that either succeeds or fails, every plan like it has to be complete before you start, and every result like it has to be perfect. The features here let you match the effort to the task: probe instead of committing, plan the backup first, call a result not yet instead of failed, and stop at enough.
Trying it out
English try does not say whether you are committing to keep going or only testing whether something works, so every attempt can sound like it will end in success or failure.
Try says that an attempt happens and leaves the outcome open. Its ending says how far the attempt will go: a committed attempt (thudul) keeps going until it works or is ruled out, and a trial run (thudur) is a low-stakes probe where failing is useful to know.
Calling an attempt a trial run helps you start without betting everything on it: if it fails, finding that out was the point.
In English: Claritish: Decisions and tries.
Planning the backup first
A plan made in English often names only the first choice, and either stays vague or has to be complete before you act.
A fallback joins an attempt to its backup with xon: try A; failing that, B. The backup counts only if the attempt before it fails. The PLAN ending says how much of the plan is drawn: backups included, steps filled in, or only a direction. No level is better, and a sketch is often all a small or early plan needs.
Naming the backup before you start helps a setback turn into the next step instead of a dead end, and matching plan detail to the task keeps planning from becoming a reason not to start.
Good enough
English enough and too much often compare a result with an unstated ideal, so nothing ever counts as done.
A sake bar measures an amount against what a sake actually needs: enough, too much, or not enough for that need. When you do not name which sake, the unspecified sake stands in.
Measuring against the need instead of against perfect helps you notice when a result is good enough to stop.
Treating silence as a yes
An act that touches someone else's body, things, or time can go ahead on a yes nobody said, and a yes once given can sound permanent. The features here say whose yes it is and how binding, so consent comes first.
Consent: whose yes, and how firm
English she's fine with it does not say whether she signed on, said yes and can take it back, or was never asked.
Consent names the person the act happens to and says how binding their yes is: a binding agreement made in advance, a yes they can take back at any time, or a yes you are only assuming because nobody objected.
Marking an assumed yes as an assumption helps you check before acting, and marking a spoken yes as revocable keeps it theirs to withdraw.
How long you can talk
English hi does not say whether you can stay, whether you have only a few minutes, whether you have one thing to ask, or whether you are only passing by.
A greeting is your own name said as a sentence (n.). Calling someone is a vocative. A conversation-length bid adds one vowel to your name: open-ended, a few minutes, one thing to ask, or passing by. The other person can answer with their own bid; the shorter bid sets the limit for both of you. When you leave, the same form says whether you are leaving soon, about to leave, or leaving now. The vowel describes the conversation, not either person.
Marking the opening that way helps you agree to the talk without a greeting committing you to unlimited time.
Conceptual metaphors
Abstract ideas are hard to hold; without a concrete picture, the psychologically useful reading of a word slips away in the moment. This cuts across the groups above.
Many roots have an abstract sense beside a concrete one, so an idea you cannot see is tied to a familiar scene that already carries the caution. Examples:
- Memory as fishing (a guess as fact): cast into opaque water; you may pull up a fish or a boot (the memory can be wrong).
- Live look as attesting (a guess as fact): you report what you see while the scene is still in view.
- Residue as unpaid debt (a moment as identity): the result still counts, like a balance still owed; not a past tense.
- Former climate as an empty nest (a moment as identity): the nesting season is over; a pattern that has ended, not a past tense.
- A pattern as paw-prints (a guess as fact): a general claim rests on a trail of cases, and the ending says how long the trail is. Always from a few cases is a short trail, not a law.
- As if as theater (a guess as fact): pretending is framed as a stage scene; what really happened stays offstage.
- Mechanism as gears (a wish as a law): you say how a result comes about, like gears meshing, not only that it follows.
- Affect as water (a feeling as a label): surging, flowing, or still water (motion) plus where the feeling sits or points, instead of an opaque emotion label that hides the sake.
Conceptual metaphors help you reach and remember the healthier frame by tying it to a concrete picture that already encodes the bias you are trying to catch.
Criticisms
Can a language change how you think?
Some people doubt that a language can influence its speakers’ thoughts (the claim that it can is the weak Sapir-Whorf hypothesis). I make no claims here about whether it happens for natural languages, but for Agazan specifically, I give the following argument:
- The way a concept is framed (phrased) affects how people think about it (a well-established phenomenon in psychology and the social sciences generally).
- Agazan’s vocabulary and grammar ensure particular frames for certain concepts (using the design laid out on this page and in the linked grammar docs).
- Therefore, Agazan influences how speakers think.
Ethical considerations
Some are reminded of Newspeak from 1984, where a totalitarian government alters the language to influence the populace into being more compliant. What makes Newspeak unethical is that the language is forced upon an unwitting populace.
For Agazan and similar languages, as long as using the language is voluntary, and the speaker knows how the language influences their thoughts (and the language delivers on those promises), then it is morally good. Learning the language is like enrolling in a self-help course to change your thought patterns: clearly ethical if it changes you in the healthy way you want. Claritish lets you sample the effect in English before you commit to anything more.
Language influences but does not determine thought. Speakers who wish to think or communicate in a way discouraged by the language are still capable of doing so (though the sentences will likely be longer).
Language evolution
Some think that the inevitable change to languages that happens over decades of use will lead to the constructed language eventually being “watered down” to the point where it no longer has the properties that make the language beneficial.
While the language can and should change to keep up with the times, I am optimistic that it will retain or even strengthen its beneficial properties because the people who choose to learn the language want those beneficial properties and so will adopt changes in line with the language’s goals.
Even if the language does lose its potency over a century, having a positive impact for many people over such a long period of time is still a worthwhile accomplishment.
Next
To try the ideas in English first, start with Claritish. To go straight to the language, continue with introduction.md for the name, grammar design, and how to learn these docs, then phonology.md for letters, word-endings.md for citation endings, clause.md for clause shape, speech-moves.md for turns, then dependents.md. Community updates: Agazan on Reddit.